Technical Analysis For The Trading Professional Pdf Free Download

Even though Dynamic Technical Analysis obtained combined evaluations each about this site, as well as through traders/investors individually recognized, it had been ultimately bought due to the fact from it becoming suggested through Steve Bollinger, the actual guy which initially created Bollinger Rings.

1. Technical Analysis For The Trading Professional Pdf free download. software
2. Technical Analysis For The Trading Professional Pdf Free Download For Windows 10

Download technical analysis for the trading professional or read online books in PDF, EPUB, Tuebl, and Mobi Format. Click Download or Read Online button to get technical analysis for the trading professional book now. Technical Analysis For The Trading Professional Pdf Free Download Que Es El Swap En Metatrader 5. Indicadores forex mt4 When rolling a position over to a. [FULL]Stock Market Ebook Collection PDF Download. Intermarket Technical Analysis – Trading Strategies.pdf John J Murphy – Technical Analysis Of The Financial Markets.pdf. Professional Stock Trading.pdf Market Making And Reversal On The Stock Exchange.pdf Market neutral strategies.pdf.

You may not have to try them all; just work your way down the list until you find the one that works for you. Ps2 usb controller driver download. But don’t worry. TIP: If both of your usb keyboard and mouse stop working, use the touchpad, a PS2 mouse or a PS2 keyboard to perform the following procedure. All steps apply to Windows 10, 7, 8 & 8.1. There are three methods for you to fix this issue.

My personal just rue is actually formerly stalling the actual buy. It’s a truly excellent guide so far as content material — the most crucial a part of such guide. It offers realistically organized, succinct, total info which will help to make investors as well as traders cash. Exactly the same skillfully described methods function in most Time Frames. Therefore for all those which industry several TimeFrames, you can easily changeover in between all of them. Also it comes after which indiviDuals focusing on intraday buying and selling, or even individuals focusing on expense profile handle by way of every week in order to month-to-month fast inspections may also be nicely offered with this particular understanding.
Additionally, it seems which since it is really comprehensive as well as obviously offered, it’s ideal for anybody simply getting started within buying and selling, or even controlling their very own profile. Like a issue associated with truth, in some instances the guidelines offered tend to be divided for individuals who tend to be more skilled, or even other people who’re beginners.

Sample questionnaire pdf free download. Health questionnaires have quite a lot of sensitive questions for the patient to comprehend, so be nice and provide guidance (especially if respondents find something confusing).

A few graphs tend to be submitted, a regular graph with regard to GMCR, as well as a good intraday graph through these days (12/22/11) with regard to BIDU. GMCR is really a share exchanged prior to the DTA guide had been bought, as well as had been accustomed to observe how the actual method might have proved helpful inside a lengthier (daily) TimeFrame on the share along with that we had been acquainted. Really a few (a subset) from the DTA methods had been being used once the GMCR industry had been on-going, probably simply because Bollinger Rings happen to be employed for a great number of many years, as well as perhaps found had been comparable DTA methods previously. A far more total group of DTA methods had been utilized in really buying and selling BIDU within actual Time.

Jan 8, 2019 - Here's list of 2019 top 10 torrent sites for music which offer massive high. You can dig out latest hit songs and albums by popular artists. Cz12 Full Movie In Tamil Dubbed Free Download By Torrent >>> test. Tamil Movies 2016 full movies 2015 full movies 2014 full. TAMILTOP 3 SITES TO DOWNLOAD LATEST TAMIL MOVIES IN TORRENT IN TAMIL. Torrent download, tamil video songs, tamil dubbed. 8 May 2017 - 3. Jan 4, 2015 - Making a list of the world's most-visited torrent sites has been a long-standing tradition at TF, which we continue today. At the start of 2015 KickassTorrents is pulling in most traffic followed. In addition, we include last year's ranking for each of the 10 sites. 5 Ways To Download Torrents Anonymously. Download Latest / New Tamil Movies Torrent Free HD p p test. 2015 Hd Video Songs 1080p Blu-ray Tamil Video Songs Torrent. Mission Impossible Fallout Download 2018 Torrent Movie Full HD. Tamil Movies 2016 full movies 2015 full. All Latest Online Telugu Tamil Hindi Dubbed Movies, Hollywood 3D P P. Pl torrent download, tamil video songs, tamil dubbed. Latest tamil songs 2015 download torrent.

Smart Sync with Beatgrids gives you even more time to get creative. Load and customize your choice of high quality FX powered by iZotope including: Delay, Echo, Ping Pong Delay, Reverb, Phaser, Flanger, Distortion, HPF, LPF and a Combo HPF/LPF. Adjust and customize a range of parameters for each effect to shape your sound. Serato DJ delivers all the features and performance to meet the demands of DJs at the top of their game and sets the new standard for controller DJs. Lock your tracks together for seamless grooves. Serato scratch live 2.5 download.

The Stylistics) 9 - The Whole Town's Laughing at Me (feat. Teddy pendergrass turn off the lights mp3 download.

Others Searched For

Technical Analysis For The Trading Professional Pdf free download. software

• quantitative technical analysis howard bandy

Technical Analysis For The Trading Professional Pdf Free Download For Windows 10

IP'
TECHNICAL
ANALYSIS FOR T H E
TRADING PRDFESSIDNAL Strategies and Techniques for Today's Turbulent Global Financial Markets SECOND EDITION
CONSTANCE M. BROWN, CMT Mc Graw Hill New York Chicago San Francisco Lisbon London Madrid IWexico City Milan New Delhi San Juan Seoul Singapore Sydney Toronto
The McGrawHill Companies Copyright © 2012 by Constance M. Brown. All rights reserved. Printed in the United States of America. Except as permitted under the United States Copyright Act of 1976, no part of this publication may be reproduced or distributed in any form or by any means, or stored in a database or retrieval system, without the prior written permission of the publisher. 123456789
10 DOC/DOC 1 9 8 7 6 5 4 3 2 1
ISBN 978-0-07-175914-4 MHID 0-07-175914-X
CONTENTS
e-ISBN 978-0-07-175915-1 e-MHID 0-07-175915-8 This publication is designed to provide accurate and authoritative information in regard to the subject matter covered. It is sold with the understanding that neither the author nor the publisher is engaged in rendering legal, accounting, securities trading, or other professional services. If legal advice or other expert assistance is required, the services of a competent professional person should be sought. -From a Declaration of Principles Jointly Adopted by a Committee of the American Bar Association and a Committee of Publistiers and Associations
Art credits of permissions info or referral to a credits section at the back of the book. Library of Congress Cataloging-ln-Publicatlon Data Brown, Constance M. Technical analysis for the trading professional: strategies and techniques for today's turbulent global financial markets / by Constance Brown. — 2nd ed. p. cm. ISBN 13: 978-0-07-175914-4 (alk. paper) ISBN-10: 0-07-175914-X (alk. paper) 1. Speculation. 2. Investment analysis. 3. Stocks-Charts, diagrams, etc. I. Title. HG6041.B74 2012 332.64'5'dc23
Part
I
Foreword v i i Acknowledgments x i i Disclaimer xiv
1 DISPELLING SOME COMMON BELIEFS ABOUT INDICATORS
Chapter | 1
Oscillators Do Not Travel between 0 and 100 3 Oscillators, contrary to popular belief, can be used to define market trend. A discussion focusing on RSI and Stochastics reveals how specific indicator ranges can confirm larger trends and identify an upcoming trend reversal.
2011041019
McGraw-Hill books are available at special quantity discounts to use as premiums and sales promotions or for use in corporate training programs. To contact a representative, please e-mail us at [email protected]
Chapter [ 2
Dominant Trading Cycles Are Not Time Symmetrical
27
Rhythmical fluctuations can be more than just a fixed interval. Explore how three great North American analysts, Benner (1875), Dewey (1930s), and Cann used cycle analysis to examine the trends and risks in their time and ours presently.
This book is printed on acid-free paper.
Chapter j 3
Choosing and Adjusting Period Setup for Oscillators
59
The correct method for determining the time period for an oscillator is discussed. Recommendations are given as to when to change initial oscillator setups
• Mi •
iv •
Contents
and the methods for making these adjustments. This chapter also reveals how institutional traders can benefit from inexperienced technical traders.
Chapter | 4
Chapter | 7
79
Chapter [ 8
Contents
• v
169
Price Objectives Derived from Posihve and Negahve Reversals in the RSI
Signals from Moving Averages Are 107
Price Projechons by Reverse-Engineering Indicators
The concept of reverse-engineering an indicator to forecast price objectives is discussed using Microsoft's Excel software. This chapter includes a detailed, stepby-step picture illustration of how to export data from Omega Research's TradeStation to Microsoft's Excel software for advanced custom analysis.
Dominant Trend Lines Are Not Always from Extreme Price Highs or Lows
Trend lines of greatest significance may not originate from major price highs and lows. Visual exercises help demonstrate geometric proportions that extend beyond traditional charting techniques. Learn how gaps and strong bars define the angles to create channel lines. The market timing of high-risk trend changes can also be identified by projecting trend lines forward from specific chart signals.
Chapter j 5
185
RSI can be used to calculate price objectives from specific indicator patterns called Positive and Negative Reversals. Signal probability and price objectives can be dramatically improved by measuring the amplitude of the oscillator. The price projection method and filtering technique are fully explained and illustrated.
Frequendy Absent in Real-Time Charts
Time-sequenced charts show that trading signals from moving averages and indicators that incorporate averages may not be present in real-time market analysis. Knowing how these indicators change their screen positions when the current bar is displaced with new data will allow traders to adjust to real-time conditions and miss fewer signals.
Part I 2
Chapter | 9
Gann Analysis: Calculating Price and Time Objectives
211
A detailed introduction to W.D. Gann's use of price, time, and diagonal analysis to determine market movement. You will be introduced to the laws of vibration and natural laws, and harmonic ratios. Examine a 700-year history of the Wheat market. Explore the use of a Gann Wheel, and the critical geometric relationships between a Gann square, Gann Fan, and Square of Nine Gann price projection. Included are the formulas to create your own Gann Wheel.
CALCULATING MARKET PRICE OBJECTIVES
Chapter 6
Chapter | 10
Using Oscillators with the Elliott Wave Principle
Adjusting Tradihonal Fibonacci Projechons for Higher Probability Targets
247
A market move that requires several days to develop is used to walk readers through a real-time, stepby-step progression to show how to apply the Elliott Wave Principle. As the market unfolds through a time-sequenced event, all the rules, guidelines, and patterns are discussed to show how they are used to predict future market action.
13 5
Weaknesses exist in the industry's standard use of Fibonacci relationships. The industry in general determines 0.382, 0.500, and 0.618 retracement levels from distinct price highs and lows and also projects 618, equality, and 1.618 market swings from the same pivot levels. This chapter reveals methods for more accurate price projection to accommodate the normal expansion and contraction cycles present in all markets. In addition, it demonstrates why price spikes or key reversals should not be used to determine Fibonacci retracement targets.
The prior chapters are also applied to clarify how the different techniques are combined. Wave counts in hindsight versus developing wave scenarios unfolding in real-time to predict future market movement require different skills. The differences are discussed, and common misunderstandings in the industry are identified.
vi •
Contents
The chapter concludes with an in-depth look at an analytic method used to develop long-horizon wave interpretations for the S 6 P 500.
Part 1 3
NEW METHODS FOR IMPROVING INDICATOR TIMING AND FILTERING
FOREWORD
PREMATURE SIGNALS Chapter | 11 V o l a h l i t y Bands o n Oscillators
357
A volatility band formula is discussed that has a different character from Bollinger Bands. The chapter also discusses the importance of establishing independent variables for upper and lower bands because markets do not decline in the same manner as they advance.
Chapter
12
T h e Composite Index
431
Appendix A
403
W r i t t e n by a b r i l l i a n t trader for only those seasoned traders w h o are w i l l i n g to work at their analysis o f the markets i n a disciplined way, this book contains the most advanced methodology I've ever seen! C o n n i e Brown's credentials come i n the f o r m o f nine years o n
force to reckon w i t h i n the financial markets. A t the same t i m e , she publishes a daily b u l l e t i n o n the D o w , the S&P, and Bonds. T h i s is faxed to some o f the world's most sophisticated, large traders. Her predictions as to price objectives and trend o f the market are unequaled anywhere i n the industry. T h e r e are 14 separate chapters i n this book, each a separate subject. Six of these subjects have been written o n before, and these chapters serve as improvements o n o l d indicators. T h e r e are, also, 15 major breakthroughs i n technical analysis! Seven o f these breakthroughs are new—never-before-revealed material! Eight more dissect, change, and improve o l d concepts. I n her discussion o f Stochastics and o f RSI as oscillators, she introduces the concept that oscillators do n o t necessarily f l u c t u -
429
435
Appendix B
the front line as a research analyst and f u n d trader. She is herself a disciplined professional, w h o has grown to the p o i n t where she is a
369
The Composite Index is fully disclosed and discussed in numerous global charts such as China's Shanghai Composite, American Corporate Debt BBS Bond Funds, and the USDCHF Forex market. The oscillator was developed to warn traders when RSI is failing to detect trend reversals.
Chapter | 13 T h e Principles of D e p t h Perspective A p p l i e d to T w o - D i m e n s i o n a l C h a r h n g
This chapter discusses the strengths and weaknesses of common indicators in order to isolate desired characteristics for developing a custom formula. Personal biases contribute to the effectiveness of an indicator's signal. Depth perception is discussed to clarify how we evaluate charts graphically in a twodimensional environment. Depth perception needs to be understood as it will affect our judgment of chart signals and will dictate how indicators should be plotted to accommodate personal strengths and weaknesses.
Credits
437
Appendix C Index
ate between 0 and 100 and that all signals do not fall w i t h i n the traditional default overbought and oversold bands. T h e oscillator may actually travel w i t h i n a larger or a narrower range that can be pinpointed w i t h precision. To correct what the writer perceives
439
. vii .
viji •
Foreword
Foreword • ix
the music, they cannot perceive the beauty of the composition
ing band may travel up and down within the range, and that it may
people are 'wave-deaf'; just as a tone-deaf person cannot hear
missed. She also introduces the concept that this effective signal-
begins. Understanding their frustration, Connie agrees that some
range to help identify signals that might otherwise have been
because the wave count appears to change when a larger cycle
of an upper resistance band and a lower support band within this
industry. For some, the Elliott wave is frustrating i n the extreme
as a flaw in commercial software packages, she suggests the use
expand or contract. She suggests that the trader should adjust this effective signaling range to compensate for the idiosyncrasies of strongly trending bull and bear markets, and even suggests some better parameters! This alone would change the way we look at oscillators—and, consequently, our entry timing.
because they are caught up in counting the beats and analyzing the notations. She stresses that it is necessary to understand the structure, but more important to keep a sense of proportionality to the analysis. Then she teaches the three simple rules that form the basis of
ponent of a hypothesis I've long espoused: Stochastics can prove
duces the Composite Index she created to accompany RSI.
lators, Fibonacci ratios, and Gann into her analysis. She is a pro-
and filtering indicators with variants of different lengths. She intro-
real-time 'to the T,' showing as she goes how she integrates oscil-
generation indicators created by applying oscillators on oscillators,
explanation of flats and zigzags, and analyzes a number of charts
to discuss the application of moving averages over oscillators, third-
Elliott wave analysis, takes the reader through an 'easy to take'
But this inventive young trader does not stop there. She goes on
In a theme she returns to frequently, she kids the 'Stochas-
Elliott wave—and help clarify an indistinct wave count!
swing objectives, Connie's projections are plotted from numerous
passionate case for keeping a flexible state of mind.
retracement. In her discussion of the use of multiple Fibonacci
data to construct responsive, customized indicators. She makes a
order to properly calculate the correct price objective of the affected
of attention to procedures for determining and inputting the proper
and that the trader has to remove the differential of the gap in
but does it anyway. She remedies this deficit by giving a great deal
She rightly points out that markets may gap past a price objective
clue as to why, and the educated but lazy trader who knows better
chapter on Fibonacci measurement truly upgrades this old friend.
default values in software and tries to use them to trade without a
Connie also discusses Fibonacci methodology in depth. The
tics Default Club'—both the uneducated public that accepts the
I was particularly impressed with the discussion and the upgrades.
of a weighted factor to them, versus Fibonacci cycles. She explains
support and resistance levels, which are useful in and of themselves.
and decay,' which leads to asymmetrical cycles, and the application
pivot levels. She has found that these levels tend to cluster into tight
To the subject of cycles, Connie introduces the concept of 'growth
the use of charts with differing time cycles to perfect cycle timing.
This chapter has been badly needed. The discussion on spikes by the reader. The explanation of the Fibonacci price projection
naturally turns to the Elliott wave, her starting point i n the
and internal Fibonacci guidance is to be especially appreciated
Approaching the subject of market price objectives, this writer
X • Foreword
method—and specifically the use of multiple Fibonacci swing projections—is worth the price of the whole book!
Foreword • xi
In doing this, she correctly arrives at the conclusion that Fibonacci and Gann took two routes to arrive at much the same
with even greater respect.
required to work outside the mindset established by the puzzle. So,
or price between the different methods she uses should be treated
from the page. To come up with the correct answer, the reader is
the same indicator should be respected, confluence in areas of time
to connect the squares with four lines, without removing the pencil
areas of confluence in time or price within different charts with
writer challenges us to solve a puzzle, the Nine Squares. The task is
place. This has led her to another valuable concept: that, just as
Before tackling the subject of trend lines as price predictors, the
be done—in the next book, of course!
the intersection of trend lines as a timing tool—a subject that
money in the markets. Connie, I challenge you to prove to us it can
ignoring spikes or using intermediate highs or lows. She discusses
a non-Mensa trader such as myself to actually use Gann to make
in the market, she prefers less conventional approaches, such as
she is holding back on some of the specific information needed for
she believes spikes at tops and bottoms are caused by aberrations
can be used to locate dates of changes in trend and price objectives,
lines must be established from absolute highs and lows. Because
However, while she explains how Gann's time and price wheels
too, the writer asks us to suspend our preconceptions that trend
has needed clarification for years! Then, she demonstrates a very
—George C. Lane
unconventional use of trend lines to 'reverse-engineer' a triangle that can be bisected into two right triangles by a line extended into the future that will point to a final bottom. She goes on to introduce an entirely new approach to trend lines—the intersection of trend lines from divergent highs on an oscillator with the long-term trend line. The results are astounding! This is 'eyeball training' from which a good chartist can profit! The Nine Squares connected by four unconventional trend lines in the formation of a pyramid is an excellent lead-in to the subject of Gann. Because his methods seemed enigmatic, it has been suggested that Gann used astrology to arrive at his predictions, and his work has been obscured by the veil of occultism. Connie has correctly perceived that this is not the case and has done an exceptional job of returning his use of an astronomical clock, the third oldest calculator known to humankind, from the occult to the realm of science and simple mathematics!
ACKNOWLEDGMENTS
Acknowledgments
• xii!
divergence signal that developed between two peaks of eight years apart is problematic. Learn to use every method as a weighting factor. This method versus another will either compliment, contradict, or add confusion. A l l are valid as the latter means stay out of a market when all the evidence is unclear. That is a valid opinion as well. But we
isolation are outdated.
There has been so much personal growth since this book was first
our market opinions now. The times of looking at one market in
this project to revise Technical Analysis for the Trading Professional.
have to use multiple time horizons and global markets to develop
I would like to thank Morgan Ertel of McGraw-Hill who initiated
markets. But the key thing for all of us students of markets to keep
involved in producing a book takes as long as it does to write it.
one needs skills to read charts from all financial and agricultural
critical step as well. Most people do not realize the design and work
South African Rand and the other in U.S. Dollars? Today every-
text that flows easily to read. The layout and clarity of graphics is a
diverge will you understand the reason is because one is based in
words from someone who is not a wordsmith and mold them into
stock traders need only study stocks. But when two Gold stocks
mation on to you. Books need a very skilled team to take the raw
Readers need skills in all the financial markets. Some believe
written it was a wonderful opportunity for me to pass this new infor-
ing. By doing well today we earn the right to participate tomorrow.
him continued success.
presented itself to you. Take it. Run with it. Answer it. Assume noth-
list of major contributing authors who feel similarly. We all wish
become aware of a question that the opportunity has just politely
advisor and expert contact in the book publishing worid. He has a
to find the opportunities and risks. Always recognize when you
I have eight books from various publishers. He remains my primary
goal is to stay ahead of the ebbing tide of market changes in order
ent publishing firms, I would stay loyal to him. It is the only reason
methods to resolve the cash flow puzzles that lie in our path. Our
made this book possible in 1998. As his career moved h i m to differ-
in mind as the goal is not the mastery of our tools, but to use our
A special mention must be given to Stephen Isaacs who first
chairs.
G M T (Ghartered Market Technician) designation. The experience
changes are endlessly challenging and keep us on the edge of our
paring for the final exam of the Market Technician's Association
I guess that explains why few in our industry ever retire. The world
I would like to take a moment to encourage the readers pre-
contact them.
that will build an opinion on the future move of a market. A
the Gredits at the back of this book for a detailed list and ways to
historians when the focus needs to be on searching for evidence
who made the illustrations in this book possible. Please refer to
have addressed in this second edition. Young analysts tend to be
There are several market software vendors and chart services
of grading these exams shows a few common misunderstandings I
* xii •
DISCLAIMER
PART
1
It should not be assumed that the methods, techniques, or indicators presented i n this book w i l l be profitable or that they w i l l not result i n losses. Past results are not necessarily indicative o f future results. Examples i n this book are for educational purposes only. T h i s is not a solicitation o f any order to buy or sell. T h e N F A requires us to state that:
'Hypothetical or simulated performance results have certain inherent limitations. Unlike an actual performance record, simulated results do not represent actual trading. Also, since the trades have not actually been executed, the results may have under- or over-
DISPELLING SOME COMMON BELIEFS A B O U T INDICATORS
compensated for the impact, i f any, of certain market factors, such as lack of liquidity. Simulated trading programs in general are also subject to the fact that they are designed with the benefit of hindsight. No representation is being made that any account will or is likely to achieve profits or losses similar to those shown.'
• xiv •
chapter
1
O S C I L L A T O R S DO NOT TRAVEL BETWEEN 0 AND 100
A I h y does it appear that conventional technical indicators V V are failing us as we approach the twenty-first century? What has changed?' Thirteen years ago this was an opening which did not know the volatility changes that would be ahead. However, this method of describing oscillator movement to determine trend and entry/exits in volatile conditions has stood up to the test of time. I believe strongly that a method you favor should be able to handle market changes; therefore, the original text requires no revision. The only suggestion I would offer is to stop reading occasionally and study how a 14-period Relative Strength Index (RSI) has moved in your own charts. Take any time horizon or market. Study global indexes. Gold, O i l , Bonds, and Forex trend or trendless markets, as the method described here will stand up to the challenge. The ranges defined in this chapter remain valid and of value. However, the astute analyst and trader will know there • 3 •
4 • D I S P E L L I N G SOME COMMON B E L I E F S ABOUT INDICATORS
Oscillators Do Not Travel between 0 and 100
• 5
in w i t h instantaneous execution.
methodology for you and Chapter 12 w i l l continue the discussion
through the 20 line.' I n mass, the orders f r o m the same signal pour
current charts. Therefore, this chapter w i l l accurately describe the
crosses the 80 line w i t h divergence, and buy w h e n you roll back up
and b o t h the Composite Index and RSI w i l l be displayed i n more
manual that proclaims, 'Sell your Stochastics when it rolls over and
t i m e the chapter o n the Composite Index w i l l be f u l l y disclosed
all too easy to set up a chart and then read a quick description i n a
of a c o m i n g major trend reversal. For this there is a solution: this
of indicators available to t h e m by the click of a mouse button. It is
are times w h e n the RSI w i l l fail to give any divergence w a r n i n g
i n a more current market environment.
I do not fear what the professional trader m i g h t do. N o r do I have
on the defaults that failed t h e m i n the past.
sis software; professional systems and retail software products alike
the new model and this model is of little value i f the majority rely
quote vendors use the same default variables w i t h i n their analy-
institutions are f a r m i n g out their research to t h i r d parties. Alpha is
were w o r k i n g independently w i t h their o w n technical tools? A l l
W i t h the collapse i n the financial b a n k i n g sector around the world,
H o w d i d this group become so tightly l i n k e d together w h e n they
However, the professional faces a new risk. T h e buzzword is 'alpha.'
to do well, so must you. N o one should look at markets i n isolation.
come to change conventional t h i n k i n g about technical indicators.
nected. W e are waking up to the new awareness that for m y country
avoided but also used to the professional's advantage. T h e t i m e has
i n front of us. But technology makes everything more tightly con-
ogy is indicator failure and capital erosion. This group cannot only be
cal analysis has proved that it w i l l hold up to whatever the world puts
and dangerous mass. T h e impact of this new breed of mass psychol-
m a k i n g older methods obsolete? Neither suggestion is true. T e c h n i -
group is u n k n o w i n g l y operating w i t h i n this new technically armed
inating. Have the indicators failed, or have the markets changed,
card. T h e professional trader who fails to move forward beyond this
technical analysis; the problem is clearly widespread and undiscrim-
novice technical traders who operate as one large institutional w i l d -
phenomenon. T h e traders affected utilize b o t h eastern and western
problem is not w i t h the professional but w i t h the growing mass of
Europe, Asia, and the U n i t e d States seem p u z z l e d by this same
G e r m a n Dax and D o w Jones Industrial Averages are out of line. T h e
have changed. Traders employed by major institutions throughout
ket becomes out of line. N o w programs are triggered because the
tions are that the technical studies that brought a trader prior success
triggered when the spread between the S & P Cash and Futures mar-
lecture or seminar far more frequently than any others. T h e implica-
strong views about market realities such as S & P programs that are
I a m asked these two questions by professional traders before a
fications.
these default variables as they are overwhelmed w i t h the l o n g list
the global volume of f u n d a m e n t a l factors and cross-market r a m i -
setup periods and formulas. Less experienced traders rarely change
analysis grew because it became increasingly d i f f i c u l t to manage
i n the world w i t h c h a r t i n g capabilities starts o f f w i t h the exact same
traders apply their tools. A t the same t i m e the need for technical
m o m e n t . Every quote system shipped to a new location anywhere
T h e 1990s brought dramatic changes to the way technicians and
still use the exact same defaults. T h i n k about that statement for a
M o r e people continue to discover the value o f c h a r t i n g
6 • D I S P E L L I N G SOME COMMON B E L I E F S ABOUT INDICATORS
techniques. However, to evolve beyond the foundations of technical
Oscillators Do Not Travel between 0 and 100
* 7
JY Daily
analysis, we must change the way we utilize technical studies. Traders still working under the premise that there are two groups of technical indicators—indicators for trending markets like moving averages and then indicators for nontrending markets such as the oscillators M A C D , RSI, and Stochastics—are now very outdated. The books that segregated indicators into two primary groups are not wrong. Do not lose sight of the fact that the original works provided us with the foundation on which our industry is growing today. The important distinction is that eady books on technical analysis will eventually be viewed as classics, but traders who fail to evolve beyond these original concepts face a far less pleasant fate: extinction. A good place to begin to dispel some of the common beliefs about our technical indicators is with oscillators. The mainstream
Figure 1.1 Aerodynamic Investments Inc., ® 1996-2011, www.aeroinvest.com
believe that oscillators generally travel between a scale of zero and The general rule to follow for a bear market is that RSI will oscil-
is an overbought market. This is incorrect.
30. There will be many more examples to reinforce this concept.
100. Generally 20 and below is viewed as oversold, and 80 or above
or time horizon.
bear market within this time horizon. The graph showing the RSI
upper resistance zone of 55 to 65. This is true regardless of market
RSI with a daily bar chart of Yen futures. The Yen is falling in a
late within a range of 20 to 30 at the low end of the scale up to an
In Figure l . I the standard default period of 14 is used for the
RSI to decline only as far back as the 40 to 50 zone.
oscillator then declines to a support zone within a range of 20 to
minor advances that lead to more complex consolidations allow the
former downtrend and establishes new lows against the dollar. The
broken. The strong Yen rallies push the oscillator into the 80s. Even
time the indicator tests the range from 55 to 65, the Yen renews its
lator falls to a support zone near 40 to 50. The 40 level is never
a Futures chart, which will be inverted from the spot market.) Each
market or the dollar is weak. Each time the Yen declines, the oscil-
through the 65 level. (Spot traders need to keep in mind that this is
market but over a weekly time horizon when the Yen is i n a bull
is the Yen strong enough to push the RSI oscillator successfully
resistance zone of 80 to 90. Figure 1.2 shows the same Yen futures
zone for the indicator. Study the indicator tops closely. At no time
a range marked by a support zone of 40 to 50 toward an upper
from 60 to 65. A lower band marks 23 to 28 to highlight a support
In a bull market the RSI will shift and begin to oscillate within
indicator has an upper black band marking a range of resistance
8 • D I S P E L L I N G SOME COMMON B E L I E F S ABOUT INDICATORS
JY-Weekly
Oscillators Do Not Travel between 0 and 100 • 9
back up to the 25 zone as the market would then target a new price low equal to the decline that preceded the minor rebound from the oversold condition. Examples for this price projection method from Stochastics will be offered in their correct context in Chapter 14, but the point to make at this time is that oscillators can be used to forecast market trends, which is contrary to popular belief today. Let's build on the introduction of the range rules for RSI by moving on to Figure 1.3 showing a weekly Dow Jones Industrial Average chart. The chart shows clearly that the market corrections in 1996 and 1997 all pressed the RSI down to the lower support zone defined for a trending bull market. The indicator holds the
Figure 1.2 Aerodynamic Investments Inc., © 1996-2011, www.aeroinvest.com
40 to 50 support zone and signals correctly that the market will DJIY-Weekly
Do these RSI ranges defining bull and bear markets apply to other oscillators? Yes, when the period used for the indicator has been correctly defined. We will cover how to find the correct period in the next chapter, and we will readdress the issue of buy and sell ranges as other oscillators are discussed. As an example, in Chapter 14 a price projection method is described for the Stochastic Oscillator which gives the trader permission to buy a market when the Stochastics indicator falls from an extreme high down toward the 75 area. Yes, buy, as the signal will warn the trader that the market could target an additional move equal to the rally that preceded the minor pullback which allowed the indicator to decline from its extreme high over 80. This is only one example of instances when it would be incorrect to sell just because the Stochastics indicator has crossed the 80 range. Conversely a trader would have permission from Stochastics to sell the market when the oscillator moves
'36
'97
'98
Figure 1.3 Aerodynamic Investments Inc., © 1996-2011, www.aeroinvest.com
10
•
D I S P E L L I N G S O M E C O M M O N B E L I E F S ABOUT INDICATORS
Oscillators
Do Not Travel between 0 and 100 • 1 1
zone is i n d i c a t i n g a t o p p i n g f o r m a t i o n w i t h i n a bear market.
range rules that help to define a market's trend. A dramatic example
and M a r c h . A market that can press the R S I only to the 55 to 65
equity indexes w i l l all display this characteristic of predetermined
the 55 to 65 resistance level t h r o u g h o u t the advance i n February
at the 75 level or higher. Asian, European, and N o r t h A m e r i c a n
chart was captured for this book. T h e RSI oscillator fails to exceed
a b u l l market i n each follow-up advance and then finds resistance
warn that the decline is incomplete as of June 3, 1998, w h e n this
rally toward new highs. T h e oscillator moves to the upper range for
t h i r d test seems required.
N o r t h America. T h e daily chart i n Figure 1.4 shows the decline i n
key b o t t o m as the 45 zone is being tested a second t i m e now and a
equity correction that unfolded i n late October 1997 i n Europe and
m o n t h l y chart such a capitulation spike d o w n c o u l d then f o r m a
Asian stock markets. T h e Asian woes c o n t r i b u t e d to t h e global
i n the daily t i m e h o r i z o n c o u l d occur, but i n t h e context o f the
gered a c h a i n reaction w h i c h was u l t i m a t e l y felt t h r o u g h o u t the
price b o t t o m is not i n place as of June 3, 1998, and a sharp freefall
concerns about the strength o f the fixed Chinese r e n m i n b i trig-
about the H a n g Seng that the daily chart i n Figure 1.4 warns that a
to contend w i t h a weakening Yen. T h e H o n g K o n g dollar peg and
chart has fallen only to the 40 to 50 support zone. I t can be said
ket entered a sharp decline i n 1997 w h e n the Asian currencies had
is w i t h i n the context of a larger b u l l market. T h e RSI i n the m o n t h l y
E q u i t y Index i n b o t h a daily and m o n t h l y t i m e h o r i z o n . T h i s mar-
T h e m o n t h l y chart shows a brighter picture as the severe decline
can be seen i n Figure 1.4, w h i c h charts H o n g Kong's H a n g Seng
the H a n g Seng Index f r o m M a r c h to June o f 1998. T h e range rules
H o w can you tell w h e n the trend is about to change? T h e bear market illustrated i n t h e daily H a n g Seng chart allows t h e R S I
HHAY-Dally
HHAY-Monthly -16000
-11500
oscillator to travel between 20 and 65, consistent w i t h t h e range rules for a bear market. W h e n the market's trend is about to reverse, so too w i l l the oscillator rangesTTn the case of the H a n g Seng daily -14000
itii
•10500 -12000
i,
9500 -10000 -8500
RSI
chart, the osciTfator w i l l develop the f o l l o w i n g RSI pattern w h e n i t is ready to trigger a transition f r o m a bear market to a b u l l market (not presently shown i n the current chart data offered i n Figure
the lower support zone reserved for a b u l l market between 40 to 50.
60.00
trend reversal w i l l come w h e n the market moves t h e R S I only to
75.00
•80.00
55 to 65, t h e market w i l l t h e n decline. T h e first i n d i c a t i o n o f a
90.00
•100.00
1.4). After an R S I failure at the upper boundary for a bear market,
-3000
•60.00 -40.00 -20.00
T h a t is why this market does not have a b o t t o m i n place w i t h i n this
45.00
chart. W h e n a b o t t o m is f o r m i n g , the RSI w i l l not break 40. True, 'D '
' 'F'
' W
' ' 'A
ZSi
'94
'95
'96
'97
W
sharp secondary declines that fail to make a new price low can pro-
Figure 1.4
duce a similar indicator formation. E l l i o t t wave traders w o u l d call
12 • D I S P E L L I N G SOME COMMON B E L I E F S ABOUT INDICATORS
Oscillators Do Not Travel between 0 and 100 • 1 3
to know how to interpret an approaching trend reversal that we will
market would fail to establish a new price low.
tion from a trending bull market to a bear market. It is so important
depending on the internal structure of the decline, but in both the
intervals. The reverse would be true for a market preparing a transi-
such a decHne a 'deep second wave down' or a 'fifth wave failure '
Regardless of a new price low or only a double bottom, the mar-
go through this transition step by step with the next chart.
RSI will travel within a bear market. The discussion that follows
tion will clearly find support for the first time at this prior zone. The
The 30 and 65 levels have solid single lines that mark where the
trends, use that as your guide. The market that is making a transi-
mate range the RSI will travel within a trending bull market.
the support range for this market was 37 to 45 in prior bull market
and 80 are marked with a double line that denotes the approxi-
more accurately the range that will be tested in the transition. If
1995 to the bull market that follows into 1998. The levels at 40
use some common sense. Use the market's past history to define
a trend reversal from a bear market for the dollar prior to April
tor that declines to 39 or 38 will still fit this transition phase, so
chart in Figure 1.5 covers a five-year time interval which includes
this discussion about trend reversals at this time. An RSI indica-
Deutsche Marks per U.S. Dollar (DMK/$). The weekly DMK/$
method will be discussed in Chapter 10, so let's not stray from
tor will now focus on one of the largest markets in the world:
it is ready to reverse its former trend. An actual price projection
Our discussion about trend reversals using the RSI indica-
ket decline will move the RSI to the support zone near 40 only if
rally that follows will likely be insufficient to press the RSI through the upper resistance range defined for a bear market. The failed attempt to exceed 65 will be met with another price decline that
[XU5DMMUI -Weekly
ixviIvS
moves the RSI oscillator back to support within the 40 to 50 range once again. Should the market produce an RSI oscillator decline that successfully holds the 40 to 50 support zone a second time, it should be interpreted as a clear warning to a sleeping bear-biased trader that conditions are changing. The market is attempting to reverse and will eventually produce a price rally from this oscillator position that is strong enough to press the RSI into the upper resistance range reserved for a bull market—75 to 85 and higher. It is not necessary for a market to test the new lower boundary more than once before successfully breaking out into the higher range, but it is a common occurrence to see double bottoms develop, especially in longer horizon charts such as weekly, monthly, and quarterly time
6
Figure 1.5 A e r o d y n a m i c i n v e s t m e n t s Inc., © 1 9 9 6 - 2 0 1 1 , www.aeroinvest.com
14 • DISPELLING SOME COMMON BELIEFS ABOUT INDICATORS
Oscillators Do Not Travel between 0 and 100 * 15
bear market.
the RSI in Figure 1.5:
30 level, which is normal support for this indicator in a
addresses each of the oscillator pivot levels that are numbered on
Point 4. The dollar attempts a rebound from point 4 in
forming at a precise Fibonacci retracement level within the
three times before it is finally broken and declines to the
is also forming a very bearish M pattern at the peak that is
oscillator declines to the 40 level. The RSI tests the 40 level
with price data; instead, use the range itself The oscillator
stated that the Deutsche Mark becomes stronger, as the RSI
point 4 to determine the Fibonacci relationship as we would
dollar then begins to weaken, or it might be more favorably
oscillator high at point 3 and then take the oscillator low at
market. This is of far greater value and significance. The
0.618 Fibonacci retracement of the zone. Do not measure the
occurs within the 55 to 65 resistance zone denoting a bear
oscillator range. The pivot high in the RSI at point 5 is a
judgment should see that point 3 is an oscillator peak that
between the two single black lines that denote a bear market
develop in the dollar, the trader with more experienced
will see it with just the naked eye. Measure the distance
oscillator suggesting that another minor correction could
occurring at an extremely interesting level. Some readers
will likely catch the divergence between price and the
a bear market. This particular indicator pivot at point 5 is
higher levels. W h i l e the less experienced technical trader
an M pattern well below the upper resistance zone defining
oscillator is lower than point 1 though the dollar closes at
Point 5. The dollar then rebounds, allowing the RSI to form
RSI forms a bearish divergence pattern. Point 3 in the
focus on the oscillator ranges.
Point 3. As the dollar squeaks past the prior high, the
other tools and indicators that will be far more obvious. Just
the RSI topped at point 1.
data, do not worry at this time, as we will be introducing
a marginal new high relative to the high that occurred when
If you do not see the small contracting triangle i n the price
point 2 was broken. The dollar proceeds to rally and makes
30 level just beyond the pivot low that was labeled point 4.
trend reversal was developing unless the support zone at
makes a new low as the RSI forms a second test of the
remains in force. At this point we would not know that a
have been favored, and the chart shows that the dollar
strength to attempt a new high and that the prior trend
market is extremely bearish. A downward resolution could
stay above 40 should warn us that the dollar has sufficient
unable to break above the lower boundary used for a bull
a double bottom near the 40 level. The ability of the RSI to
this same period of time, and the fact that the RSI was
Point 2. The market finds support, allowing the RSI to form
market actually forms a bearish contracting triangle over
dollar declines and moves the oscillator down to point 2.
the 30 and 40 levels as a market consolidation occurs. The
resistance zone reserved for a trending bear market. The
Figure 1.5, and the oscillator becomes trapped
Point 1. The RSI pivot at point 1 has occurred near the upper
between
16
• D I S P E L L I N G SOME COMMON B E L I E F S ABOUT INDICATORS
Oscillators Do Not Travel between 0 and 100 • 17
divider over without changing the measured ratio. The points on
dollars for Deutsche Marks immediately.
the two sharp points on the ends of the long side. Then flip the
bear market range. The chart is signaUng to traders to sell
the opposite side will mark an exact 0.618 ratio of the original mea-
drop it—very painful.
constructed from one of two materials, stainless steel or aluminum.
published on the subject. Just one last note about a divider: don't
ask to purchase a tool called a proportional divider These tools are
nacci relationships, please refer to one of the many books already
a paper chart? Call an engineering or drafting supply store, and
surement within 0.1 mm. To use the tool for calculating other Fibo-
What do traders use to easily calculate the Fibonacci ratios on
Do not buy the stainless steel variety as they are excessively heavy.
Point 7. Point 7 is an oscillator high under the resistance
proportional divider to avoid the headache.
defining a bull market at the 40 level.
the store does not know what they are talking about. Just call it a
pulls back. The RSI tries to hold support on the lower range
from their local engineering or drafting supply store will find that
points 6 and 7. The dollar has a small rally and immediately
financial industry. But many traders who try to purchase a PRC
Focus on the oscillator movement exactly half way between
tool is also called a precision ratio compass (PRG) by many in the
subtly tests the 30 zone before the dollar begins to rebound.
its age as it is stamped 'Alvin—West Germany.' The exact same
this oscillator squeezes the RSI back over the 30 level and
instrument being distributed in the United States). Mine shows
The tight back-and-fill chop that forms in the market after
company called Alvin (at least 'Alvin' is the name stamped on the
has fallen through this zone and become extremely oversold.
One brand that is particularly easy to handle is made by a German
at 30 to be tested after a minor rebound whenever the market
ask to buy a proportional divider, as this is an engineering tool.
market support zone. The market will allow the support level
printed on a full page. Most art supply stores will be clueless if you
6 except that it is an extreme low that occurs below the bear
that is too small for use with most computer screens or for charts
point 6 in Figure 1.5. There is nothing very notable at point
length. There are several financial firms selling a y^z-inch divider
Point 6. Back to the DMK/$ chart and the oscillator low at
You will also want to buy a divider that is at least eight inches in
was subtle as it developed between points 6 and 7. If you miss
the length of the tool. To use it, measure the distance by opening
has tried to develop support at this zone, but the first attempt
grid line. Both the 10 and GS line will be just off-center relative to
for a bull market. This is actually the second time the market
'Golden Section,' which is the same ratio as a divider with a '10'
Point 8. The oscillator is clearly at the support zone reserved
Some models will have a grid line labeled 'GS' which stands for
point 8.
screw, which also serves to keep the two prongs of the tool together.
also significant. The dollar fails and moves the RSI down to
the moveable prongs to 10 on the scale called circles. Tighten the
zone for a bear market. The M pattern in the indicator is
This is how to use it in a very fast summary: Set the grid line on
18
•
D I S P E L L I N G SOME COMMON B E L I E F S ABOUT INDICATORS
this transitional signal that the dollar is developing a major trend reversal, move along the chart to point 9.
Oscillators Do Not Travel between 0 and 100 • 1 9
surprised at the depth of information available to us. Let's keep digging deeper.
directly below line 2 are developing under the 60 level. This could
40 and 90 levels.
between 40 and 70+, denoting a bull market. The oscillator highs
dollar remains in an uptrend, and the RSI travels between the
within the context of a trending bull market. The RSI is traveling
for a market correction within a trending bull market. The
and the oscillator lows show that each correction in this stock was
Point 10 shows the RSI once again holding the support zone
weekly chart for Caterpillar. The stock is clearly in a bull market,
Point 10. Now the bull trend is in force with confirmation.
point will apply to individual stock charts. Figure 1.6 shows the
maintained, and a decline unfolds in the dollar from point 9.
the examples discussed so far. All that has been defined up to this
bear market resistance line. The breakout is only briefly
About now the stock trader may be feeling somewhat left out of
Point 9. The dollar rallies and the RSI breaks above the
be a warning that would have to be watched closely for a transition
described at length in the prior figures. However, it should be asked
from points 8 to 10. Between points 8 and 9 there is an oscillator
danger zone. Individual stocks will abide by the range guidelines
of breaking the discussion about the trend transition that develops
into a bear market if a pullback breaks below the 40 level from this
There is one subtlety that can be pointed out now without risk
peak in the middle of these points that develops exactly at the 60 level. Follow the oscillator down from this pivot high marked with
Caterpillar Inc
Corn-Weekly
an arrow. The oscillator low that follows the peak at the 60 level occurs exactly at a 0.382 retracement of the bear market range. Use your proportional divider to see that the distance from the RSI pivot low relative to the support zone for a bear market is an exact 0.618 measurement. Because we know that the space under the RSI pivot to the support zone is exactly 0.618, we know without further measurement that the oscillator has found support at the 0.382 retracement level from the 60 level. Lost on this point? It will be discussed again later in greater detail. The important points to grasp from Figure 1.5 are the trend reversal signals. As the chapters build concepts, you will begin to focus on different levels of detail within the same indicator. We have only begun to discuss technical indicators, and I suspect many might already be
I !i
T
97
r 98
Figure 1.6 Aerodynamic Investments Inc., ® 1996-2011, www.aeroinvest.com
20
•
D I S P E L L I N G SOMK C O M M O N B E L I E F S ABOUT INDICATORS
R everaqes
Soft D r i n k s -Monthly ,,.,•••1.1.' „-•'••'
Oscillators Do Not Travel between 0 and 100 • 2 1
E l e c t Co/SP-Monthly -4000.00 -3000.00 •2000.00
•1000.00 ,'1 r''''''
,1-'',,.'''''
v' RSI •80.00
•60.00
-40.00
1 85
14
Figure
1
86
1 87
1
88
—1 89
1
90
r— 91
93
92
1
1
1
94
1
95
i 96
1
97
~i
8 3
77
1
1
1
1
78 79 80 81
1 1 1 1 1 1 1 1—T 1 82 83 84 85 86 87 88 89 90 91
1
1
1
1
1
1
r
92 93 94 95 96 97 98
Figure 1.8 Aerodynamic Investments Inc., © 1996-2011, www.aeroinvest.com
1.7
if synthetic indices can be expected to follow these same ranges when they themselves are never actually traded?
by the Cash S&P Index. The data are in a downtrend, which is inter-
all the characteristics discussed previously for a bear market.
it is heavily weighted toward Coca-Cola stock. Regardless of the
broken as the sector underperforms the Cash S&P 500 and displays
be thinking that this particular sector is itself more like a stock as
ing bear market. The critical support zone at 45 to 55 is continually
is ample warning in 1988 that a major rally will follow. You might
The RSI at points 2, 3, and 4 show characteristic pivots for a trend-
declines only to the support zone within the 40 to 50 range. There
have implied that the sector was outperforming the S&P 500 Index.
will apply. The sector consolidates in 1988 and 1994, but the RSI
S&P 500 Index. Had the data shown an uptrend, the chart would
(Soft Drinks). Regardless of the sector evaluated, the range rules
preted as a severe underperformance for this sector relative to the
Figure 1.7 shows a monthly chart for the S&P sector Beverages
S&P sector's compositional weighting of stocks, these trend ranges will still occur.
About now you might see that the range rules clearly apply to various financial markets, but you can rightfully ask now if similar
have been falling in a market action that is similar in appearance to
bar chart created by dividing the S&P sector for Electric Companies
offers the daily chart for the U.S. Treasury 30-Year Yields. U.S. rates
performance between two markets. In Figure 1.8 we have a monthly
the weekly chart for the stock Novell. For comparison, Figure .9b
will create a synthetic market by developing a bar chart of relative
formations will occur within various time horizons. Figure 1.9(3 is
So let's test this premise in a tougher market environment. We
2 2 • D I S P E L L I N G SOME COMMON B E L I E F S ABOUT INDICATORS
Nkq-Weekly
22.000 •18.000
•14.000
•10.000
Oscillators Do Not Travel between 0 and 100 * 2 3
Novell. Clearly there is no relationship between these two charts, and none is implied except that both markets are presently in bear markets. While the markets are different and the time horizons displayed vary, the ranges that the RSI oscillator is traveling is the same for both markets. While the boundaries may be very slightly different as bond rates use 25 to 35 as their RSI support zone, when the stock establishes 30 to 35 as the support zone, the guidelines first introduced at the start of this chapter remain in effect. The
RSI w / o Zones 59.68
resistance zone at 55 to 65 marks trouble for both markets. The market action we have not discussed up to this point is a sideways or range-bound consolidation. The monthly chart for Bethlehem Steel covers a period of 15 years in Figure 1.10. Over this period of time Bethlehem Steel has been range bound in a listFigure 1.9a
less drift. It has been tradable, however. The oscillator peaks are all
TYXY-Daily
T reasury
at the 55 to 65 resistance zone, and the key oscillator lows display
Yield
Figure 1.10
Figure 1.9b
24 • D I S P E L L I N G SOME COMMON B E L I E F S ABOUT INDICATORS
the characteristic support zone for a bear market. The transition from this extremely long consolidation will be no different from
Oscillators Do hlot Travel between 0 and 100 • 25
Du F'ont F I n e N R m n u r s r o m - W e R k l y -75.000
what was discussed in detail for D M K / $ .
65.000
l'l'''''Haf'l(l f
Has there been a nagging question in your mind as the series of
-55.000
charts for various markets and time horizons has been discussed?
-45.000
'What period is being used for RSI in all these charts? Are we see•35.000
ing different time periods in order tofitthe oscillator to the example to create an ideal range?' No. Every chart regardless of market or time horizon has displayed an RSI with a 14-period interval. Every
RSI w/0 Zones 71.07 •80.00
single one of them. Now let me ask you, 'What would be wrong with using these range rules to establish the correct time period?' Nothing,
•60.00
except you will later discover that the RSI has a specific price projec-
40.00
tion capability that is best applied to a 14-period RSI. For this reason
—r-
96
all of these charts use the same interval within the RSI formula. We can conclude this chapter on an upbeat note with Figure 1.11.
Figure 1.11 A e r o d y n a m i c I n v e s t m e n t s Inc., © 1996-2011 www.aeroinvest.com
The weekly D u Pont stock is the image to keep in mind for an ideal bull market. The 75 to 85 range has marked the tops while the 40 to 50 support zone has marked all the pullbacks within the developing bull market. The chart also shows an immediate buy signal at the time of this screen capture. A new price high is indicated as long as the 55 to 65 level is exceeded and the 40 level is not broken. However, a rally will then lead to a pattern of great risk as we will see when we discuss wave interpretations much later in this book.
as a lone wolf. We can read probability in a global environment; the key is to always bundle and measure cash flows internally and externally that influence a single market. This will be discussed in Chapter 12 to add to this original discussion about RSI range movement. The method does not change, but we have become more sophisticated in how we do what we do successfully
Indicators require us to develop a choreographed balance of different methods. We still have much to explore together. As a final revised note D u Pont has not returned to the price highs of 1998 as of 2011. RSI established a resistance level near 66 for ten years in a monthly chart. For the first time RSI has moved above this horizontal line of resistance, but its sector sees the move
T
chapter 2 DOMINANT TRADING C Y C L E S A R E NOT T I M E SYMMETRICAL
he first e d i t i o n described for the reader the standard conventions of using fixed period cycles. I was unaware at the t i m e
that I was b e g i n n i n g a personal journey to examine cycle m e t h ods i n great depth. M y first attempt to write about cycles looks so antiquated to me now. T h e chapter included the use of expanding Fibonacci ratios to forecast the t i m i n g of possible market price pivots. T h e charts demonstrated h o w poorly these conventional m e t h ods worked i n a trading environment. T h e methods described were not comprehensive and the discussion itself had insufficient depth. Some methods of cycle analysis may be appropriate for strategy analysis and other applications are more suited to t r a d i n g signals. M a r k e t analysts and traders have very different needs and a trader must be consciously aware w h e n the use of one m e t h o d may be more appropriate than another. T h e differences arise out of people's sensitivity to risk and their need for t i m i n g precision. 27
28
• D I S P E L L I N G S O M E C O M M O N B E L I E F S ABOUT INDICATORS
W h a t do we even mean by a cycle? T h e Foundation for the Study of Cycles (foundationforthestudyofcycles.org) does n o t reference
Dominant Trading Cycles Are Not Time Symmetrical
• 29
1834 to 1900. M o r e follows about Benner shortly, but this particular cycle had a 44:1 win-to-loss ratio.
and global trade disruptions.
fixed interval can be found. Assuredly, as soon as the 'perfect' rhyth-
nize the connection o f food shortages leading to political instability
say cycles are impeccably precise, but some still believe the perfect
we trade agricultural grains or financial markets, we should recog-
regularity cannot be the result o f chance. N o credible source w i l l
scientists believe changes the yields o f our agricultural produce. I f
the r h y t h m persists over a long span of t i m e that we recognize this
solar storms make a difference to the earth's magnetic field w h i c h
tuation that repeats over time with reasonable regularity. It is not u n t i l
we care? T h e relentless c h u r n i n g o f solar gases and the resulting
historical data. W e easily miss the fact that a cycle is a rhythmic fluc-
to the n u m b e r o f sunspots and solar flares f r o m the sun. W h y do
software vendor often limits our w i n d o w o f view to only ten years o f
These events, as measured by proton radiation, have a correlation
w i t h these rudimentary tools i n our hands. B u t at the same t i m e the
t a i n i n g nitrate-rich layers that reconstruct a history o f past events.
tool offered i n most analysis programs. Therefore most of us begin
on an 11-year cycle. W e measure this cycle i n glacial ice cores con-
Cycle tools w i t h fixed periods are easier to code and often the only
along w i t h other signs of solar magnetic activity, waxes and wanes
assumptions f r o m the programmers who code our market software.
Consider the f o l l o w i n g example where the n u m b e r o f sunspots,
fixed periods i n any definition describing a cycle. We obtain these
markets on b o t h the price axis, y axis, and x axis that defines time.
1876. Figure 2.0 is Benner's Nine-Year Cycle i n Pig-iron Prices f r o m
accompany decay. B u t many do not realize this occurs as well i n
T h i s is a cycle discovered by Samuel Benner w h i c h he published i n
out i n the sun for a m o n t h and it is clear that contraction changes
fluctuation.
once again is also a cycle that has a repeatable r h y t h m i c
easier to see i n nature t h a n i n market swings. Just leave an orange
series i m p l y i n g a spread progression o f 8, 9, and 10 that repeats at 8
cycle means expansion and contraction attributes and those are
fluctuations can be more t h a n just a fixed interval. T h e n u m b e r
sion and contraction characteristics w i t h i n markets. A growth-decay
become the market price high. We use cycles incorrectly. R h y t h m i c
Fixed period cycles fail to adjust to natural r h y t h m s o f expan-
mic period is identified, it w i l l change its t i m i n g and even invert to
T h e greater span o f t i m e one is studying, the harder it is to see the expansion and contraction aspects of the cycle. A 400-year chart o f 1
J
^ .
.,
f
1840
Y 1650
—W-rvrNr 1860
V */
V
ij
Y
si
G o l d may appear to display a fixed period cycle. B u t w h e n a quarterly chart (three months w i t h i n one bar) is expanded to detail the data i n a daily presentation, the cycle t i m i n g that fits major price
-7-:>:—l-rt
1870
1880
1890
1900
lows i n the long h o r i z o n c o u l d be o f f significantly. Cycles offer us bigger picture and thereby help us to interpret the signals w i t h i n
Source: Benner's Prophecies, 1875
a tool to sharpen our awareness o f the r h y t h m i c undulations i n a
Figure 2.0
Benner 9Yr Cycle in Pig-iron Prices 1834-1900
30 • DISPELLING SOME COMMON BELIEFS ABOUT INDICATORS
Dominant Trading Cycles Are Not Time Symmetrical • 31
exacerbate the market decline into those projected market lows.
the Nikkei showed the conventional use of cycles. Read this section
multiple cycle lows that bottom in a close proximity of time will
1997/1998 the illustrations and dialogue describing the charts of
juxtaposition. We know that cycles have a cumulative effect and
our other technical methods. When this book was first written in
by an 85-period cycle in Figure 2.2. Both the 89- and 85-period
ing experience and research.
cycle in Figures 2.1 and 2.2, the 89-period cycle has been replaced
tion versus the changes that followed after a decade of further trad-
While the monthly Nikkei Index displays the same 52-period
without changes so it can be used as a comparison of cycle conven-
likely be confounded by this June decline i f they thought that the
the first quarter of 2006 when these two cycles bottom in a close
Traders who used the 85-period cycle were not wrong but would
problems again for the Japanese market around the end of 2005 and
cycle accurately warns of the decline unfolding now in June 1998.
ally accurate. These two independent cycles also project extreme
that a decline would unfold into January 1998. The 89-period
cycles are measurable, predictable, and for some markets exception-
toms for these cycle periods. The 85-period cycle accurately warned
A 52- and 89-period cycle on the chart argues the case well that
chart, which makes more recent market lows the third cycle bot-
the worid today: the monthly chart of the Japanese Nikkei Index.
cycles are anchored from lows just out of view to the left of the
Figure 2.1 is one of the most cyclically symmetrical markets in
SNIY-Monthly
NIKIvEl -Month
Figure 2.1 Nikkei 225 Index Source: TradeStation. © TradeStation Technologies
SNIY-Monthly
NIKKEI -Month
Figure 2.2 Nikkei 225 Index Source: TradeStation. © TradeStation Technologies, Inc.
32
• D I S P E L L I N G S O M E C O M M O N B E L I E F S ABOUT INDICATORS
Dominant Trading Cycles Are Not Time Symmetrical
• 33
may have an even greater impact.
ferent as the cycle lows are not as close together. A n interpretation
series is certainly of value, but conceptually the Fibonacci sequence
to 2005, and the character of the decline itself could be very dif-
quest to recognize what has immediate significance. The number
the 89-period to that of an 85-period cycle, the forecast shifts back
matical series of numbers does offer an essential component in our
be a major problem for this market. By shifting the preference from
alone will not provide us with the definitive answer, this mathe-
cycle in conjunction with the 52-period warns that early 2006 will
which to determine a major market bottom. While Fibonacci ratios
Nikkei had estabhshed a bottom six months earUer. The 89-period
and 2.618 and that they will directly impact your profit and loss.
are ignored and the periods defined today are not later adjusted for
much tied to the ratios 0.146, 0.236, 0.382, 0.500, 0.618, 1.00,1.618,
in 1998. However, both periods will be wrong if new price data
discussion for proof, know that your financial livelihood will be very
the other; we can see from these charts both had predictive value
can accept the bottom line of his theorem without the background
89 in these cycles? In reality, one period is not more correct than
an abbreviation of 'lilius Bonacci.' If you are new to Fibonacci but
can be changed so easily by just a minor period adjustment of 85 to
did you know Fibonacci was not his name? That was a nickname,
use if both the nearby market interpretation and extended views
cian, Leonard of Pisa, better known as Leonardo Fibonacci. But
that bottom together in Figure 2.1. What is the correct period to
not escape an encounter with the remarkable Italian mathemati-
comparison to the sharp spike bottom implied by the 2006 cycles
As a professional trader and student of the markets, one can-
could be made that an orderly decline in 2005 would unfold in
the markets' natural growth and decay attributes that will develop between now and 2005.
Why? These ratios have been around for a very long time. Leonardo Fibonacci wrote about them in his book. Liber Abaci, in
the number preceding it. Between alternate numbers, the higher
give us a higher probability than a six-month window of time from
number in the sequence over five is approximately 1.618 times
cycles on a monthly chart are correct. We need something that will
0.618, or 61.8 percent, of the number that follows it. Any adjacent
It is not good enough to tell a trader that both the 85- and 89-period
fact that any given number over three in the series is approximately
'What is the most important cycle within the market I trade now?'
ages and within indicators. The ratios themselves come from the
tions will not provide us with the definitive answer when we ask,
specific numbers will be of value to you as periods for moving aver-
bottom line for this methodology up front: Fibonacci cycle projec-
1, 1, 2, 3, 5, 8, 13, 21, 34, 55, 89, 144, 233, 377, and so forth. These
heading into a discussion about Fibonacci ratios. Let me offer the
the second, the second to the third, and you develop the sequence:
phrase growth and decay is used, most traders will assume we are
exactly 0.618. To create the number series, add the first number to
decay attributes within the markets for a moment. As soon as the
Pyramid of Gizeh. The slant edge of the Great Pyramid is almost
will address this particular low shortly, but let's stick with growth-
was in use around 4700 BC when the Egyptians built the Great
missed altogether by the cycles using periods of 52, 85, and 89. We
1202, but the number sequence from which these ratios are derived
In Figure 2.2 the major low that developed in July 1995 was
34
• D I S P E L L I N G S O M E COMMON B E L I E F S ABOUT INDICATORS
will be approximately 2.618 times the first, and the lower number
SNIY Weekly
Dominant Trading Cycles Are Not Time Symmetrical • 35
NIKKEI INDKX
Weekly
between alternates will be approximately 0.382. The really interesting aspect about these ratios is that they are found in anything that has a growth and decay developmental cycle. Plants, animals, vegetables, minerals, and—you've guessed it—markets. As all living entities do exhibit these exact same expansion and contraction ratios, Leonardo's mathematical theorem is viewed as a law of nature. It should not be surprising that these ratios are frequently respected within the markets. Markets expand and contract and so abide by the same law. If you want to know more, head to the Internet and start surfing. You will undoubtedly discover the original question concerning rabbit population growth and see that Leonardo Fibonacci and Darth Vader shared the same style of dress. But let's move on to see how the Fibonacci number sequence applies to cycles in the markets. In Figure 2.3 we are still viewing the Japanese Nikkei Index, but
Figure 2.3 Nikkei 225 Index Source: TradeStation. © TradeStation Technologies, Inc.
projection that denotes a market high. The Fibonacci cycle at point
or strength of a signal will be is the heart of the issue for a trader.
ratio. These lines are called Fibonacci time cycles. At points 1 and 3
reversal at point 6. The question of knowing what the magnitude
lines. Each vertical timeline plotted is derived from a Fibonacci
cycle was also correct—it just did not amount to much of a trend
The 1995 low marks the starting anchor for the vertical series of
the symmetrical cycle bottom was stronger, but the Fibonacci time
shows vertical broken lines that are numbered one through seven.
conflict. Which cycle method will be proved right? In hindsight
lows near the end of 1993 and the start of 1997. The same chart also
accompanies the price bottom in early 1997. That just adds a lot of
cycle is charted that displays corresponding cycle bottoms at market
6 immediately precedes a cycle low in the symmetrical cycle that
we have dropped down to a weekly chart. A symmetrical fixed-period
that failed. The work was notflawed—itwas just that the firecracker,
But take a closer look at point 6. That is a significant Fibonacci cycle
Many of us have witnessed an in-depth Fibonacci time cycle analysis
market trend reversal so assume a reversal from the present trend.
marked in Figure 2.3. The additional information just becomes noise.
they are approaching a market high or low. True, a pivot implies a
nacci time cycle from that low would not coincide with any other cycle
nacci time cycles that will be troubling for traders—not knowing if
the market bottom near the close of 1993 (not on the chart), a Fibo-
points preceding a market pullback. That is the thing about Fibo-
If we plot a second Fibonacci time cycle with the starting point at
the line bisects a market low. Points 2, 4, 5, 6, and 7 all mark pivot
36 • D I S P E L L I N G SOME COMMON B E L I E F S ABOUT INDICATORS
after having been ignited, d i d not produce any consequential reac-
Dominant Trading Cycles Are Not Time Symmetrical
SNIY-Monthly
• 37
NIKKEI-Montlily
tion. A l l the right ingredients seemed to be present, but then n o t h i n g happened—a d u d , or worse a reaction opposite to our expectations. M u l t i p l e Fibonacci cycles projected f r o m numerous market lows reveal confluence points where different ratios overiap one another or form very tight cluster groups. T h e cumulative effect when cycles o f varying periods or Fibonacci frequencies overlap to form a clustered group should mean a high-probability pivot and trend reversal for the market. W h i l e this is true for Fibonacci retracements derived f r o m price, it is far less accurate for Fibonacci time projections. T h e inaccuracies a n d weaknesses we a l l experience as traders w i t h b o t h symmetrical and Fibonacci cycle analysis may be caused by an oversight. T h i s oversight is i g n o r i n g the basic premise that n o t h i n g is static; therefore, rescaling for cycles f o r m i n g w i t h i n m u l tiple t i m e horizons cannot be defined f r o m linear or symmetrical periods. Figure 2.4 returns to the m o n t h l y bar chart for the N i k k e i
Figure 2.4 Nikkei 225 Index Source: TradeStation. © TradeStation Technologies, Inc.
Index. T h e same t w o cycles displayed i n Figure 2.2 are repeated
We w i l l need to add one more cycle to capture this low.
period i n this chart that captures the market l o w that was missed i n
unidentified by any o f the cycle periods selected up to this point.
cycle has now been added, and the heavier line helps locate the new
all significant. W e still have the price low at point X that remains
i n this chart showing 52 and 89 periods. I n Figure 2.4, a 44-period
tions and we then ignore information that may not fit our definitions.
lows that coincide w i t h the cycle bottoms clustering at p o i n t 4 are
as we c o n d i t i o n ourselves to look for patterns that fit our assump-
k n o w w h i c h cycle period is most i m p o r t a n t , but clearly the market
established trends? T h a t is an important observation many overlook
cycles are b e g i n n i n g to f o r m a cluster now near point 4. W e do not
1986. W i l l the future cycles be lows or mark continuations w i t h i n
period is also slightly early for the January 1998 b o t t o m where three
w i t h 56 periods o n the heavier line marks t h e start o f a rally i n
cycle b o t t o m at point 3 is late for the price low, but the same cycle
ing to suggest a low for another two years. I n addition, the cycle
ity a n d contribute to the m e l t d o w n drop into t h e 1990 low. T h e
1995. W e now have a chart w i t h lots o f cycles up to 1998 and n o t h -
At p o i n t 2 b o t h the 4 4 - and 89-period cycles have a close p r o x i m -
is a 56-period cycle. I t marks the i m p o r t a n t low that occurred i n
the price low that coincides at the cycle b o t t o m marked at p o i n t I .
T h e cycle that is now highlighted by a heavier line i n Figure 2.5
January 1998. T h e 44-period cycle was selected as it also captures
38 • DISPELLING SOME COMMON BELIEFS ABOUT INDICATORS
SNIY-Monthly
N I K l v E I -Monthly
Dominant Trading Cycles Are Not Time Symmetrical
SNIY-Monthly
• 39
N I K K E I -Montl.ly
of the tight cluster of cycles that will mark a capitulation bottom.
they are also lows of tremendous significance. Point 1 shows you the
of numerous symmetrical cycle bottoms. As for point 6, it is the end
only do the resulting Fibonacci cycles fall entirely on market lows but
to become the most significant of all the cycles preceding this cluster
tered at the start of the chart, to project the Fibonacci cycles. Not
within our cycles plotted in this chart. Point 5 marks what promises
are not using a price low, but the symmetrical cycle bottoms clus-
to it. Point 4 defines a major price low that is not even portrayed
Fibonacci series extend forward in time from this anchor point. You
late, but it is much earlier than the symmetrical cycle that is closest
ratios between the two cycle bottoms plotted in 1986. Then let the
There is more on this in the revised Chapter 4.) Point 2 is two periods
cycle in a different way. Anchor the tight starting cluster of Fibonacci
tools give better accuracy when projected from the first retracement.
charts. But now look at what happens when you use the Fibonacci
that becomes a failure to resume the prior trend. Many technical
Figure 2.6 has all the cycles we have been building on in the prior
Figure 2.6 Nikkei 225 Index Source: TradeStation. © TradeStation Technologies, Inc.
Figure 2.5 Nikkei 225 Index Source: TradeStation. © TradeStation Technologies, Inc.
basically frothed back and forth and the cycle marked the start of a
the new trend. The new trend often begins with the first retracement
Figure 2.6 a major bottom? No. The market had bottomed and
prices at price swing extremes end the prior trend, but do not start
What have I learned since 1998? Was the next major cycle in
anchor point to start this projection. (A note added in 2011: Market
40
• D I S P E L L I N G S O M E C O M M O N B E L I E F S ABOUT INDICATORS
Dominant Trading Cycles Are Not Time Symmetrical
• 41
markets. W h i l e his cycle work is i m p o r t a n t , it is his thought process
A cycle can mark where market acceleration w i l l begin.
he examined that we fail to consider today i n our m o d e r n global
cycle has completed, but prices w i l l be well o f f a price h i g h or low.
cyclical chart work that he developed, b u t i n t h e diverse subjects
strong rally. T h i s is extremely i m p o r t a n t as markets can wait u n t i l a
Two
k i n d s o f cycles were d e m o n s t r a t e d : s y m m e t r i c a l a n d
Fibonacci rhythms w i t h i n the N i k k e i Index. W h a t I d i d not k n o w
that is even more valuable. T h e t h i r d great master was W i l l i a m Delbert G a n n . His l i f e t i m e cal axes o f a market chart. His brilliance was i n the utilization o f
desired i n f o r m a t i o n sought.
work was the study o f cycles o n the horizontal, diagonal, and verti-
t h e n was that the m e t h o d demonstrated was incorrect for t h e
movement. I w i l l shortly show you a demonstration o f this.
t h i n k box ready to be called u p o n to solve a problem. As experience
his work to surmise social changes, outcomes o f wars, and market
t i o n o f well-researched tools that we accumulate i n an intellectual
the father. W D . G a n n was a brilliant analyst because he c o u l d use
mastery of technical analysis is not the end goal, but only a collec-
books apply John Gann's methods—the methods o f the son n o t
we hyperfocus o n just the h u n t for a trading signal. W e forget that
strong views for and against his work because most people w r i t i n g
to our o w n detriment. O u r t h i n k i n g becomes very shallow w h e n
G a n n k n e w h o w to u t i l i z e geometric space i n t i m e . T h e r e are
cycle work there is brilliance and creativity that we have abandoned
geometry, harmonics, a n d astronomical t i m e cycles. Simply p u t ,
Cycles have taught me that w i t h i n the historical body o f fixed
matures it gets easier to recognize the tools needed for any given market puzzle developing.
Consider some o f the fixed period cycle work of Edward D e w e y Figure 2.7 is M r . Dewey's study o f major real estate activity i n the
hand-drawn chart, so please excuse the quality o f the reproduction.
a chart from Benner's book. His charts contained Fibonacci relation-
ing the line wider so it w o u l d reproduce more easily. T h i s is an old
forecast on market booms and panics i n N o r t h America. Figure 2.8 is
added by the author.' I then exaggerated M r . Dewey's cycle by mak-
Prophecies i n 1875. T h i s book was the first study w i t h an extended
compiled by M r . Roy W e n z l i c k . A n WA year periodicity has been
could be heading. T h e first was Samuel Benner who wrote Benner's
activity i n the U n i t e d States, 1795-1939, after adjustment for trend,
we have been and used this i n f o r m a t i o n to see where their future
states, 'Figure 8. M a j o r Real Estate Activity. A n index of real estate
United States. T h e y all used technical analysis to t h i n k about where
United States f r o m 1795 to 1939. T h e description under the chart
There were three great cycle analyst masters who resided w i t h i n the
ships that had never been documented before i n his work.
W h e n M r . Dewey wrote 'after adjustment for t r e n d ' he was tell-
each t i m e the one-period crosses the longer period it is a value o f
(www.mta.org). M r . Dewey's brilliance was n o t just i n the body o f
the longer and plot the result as a line. I t becomes an oscillator as
safe keeping o f the M a r k e t Technicians Association i n N e w York
longer period. Subtract the value of the shorter period average f r o m
for the Study o f Cycles. T h i s l i f e t i m e body o f work is now i n the
to create a one-period m o v i n g average and x-period average w i t h a
E d w a r d R. Dewey, the founder and president o f the F o u n d a t i o n
ing us the data was detrended. A simple m e t h o d o f detrending is
T h e second master was b r i l l i a n t i n his study o f fixed cycles:
42 • D I S P E L L I N G SOME COMMON B E L I E F S ABOUT INDICATORS
Dominant Trading Cycles Are Not Time Symmetrical * 43
To extend the Dewey real estate chart in Figure 2.7 to current times we need to find a common denominator. The interval of 18'/3 years is 55 thirds (18 * 3 = 54 and add 1). Using 1938 from near the last cycle low (1938 * 3 = 5,814, add 110 (two cycles), then divide by 3 = 1974%. This marks an interesting start of a tremendous DJIA bull market. If we add four cycles (220 thirds) to the year 1938, the cycle falls in the year 2011'/3. But you may think that 'the real estate crash came in 2008'. Good point, but that Figure 2.7
Major Real E s t a t e Activity (U.S.A. 1795-1939)
was the start of the crash, we are still in it, and we don't have a this cycle of Mr. Dewey's with the Gann cycle discussed shortly.
Cycles, 1987, Edward R. Dewey
bottom as this is written in March of 2011. You need to combine
Source:
Cycles Classic Library Collection, Foundation for the Study of
Mr. Dewey's cycle work began with data from 1795 and remains American business cycle.
value and the value is negative when it is less than the longer period
valuable today to show where the inflection points are within the
zero. When the shorter value is above the longer it will be a positive average. A real-time application will be found in Chapter 8 where a comparison is made between the RSI and this simple oscillator when it becomes more useful in strong trends. Three great masters of cycle analysis have been named. How would these cycle gurus have fared in today's markets?
Most of our vendors cut off our window of view after 10 years. It is difficult and expensive to purchase the historical data and then it often takes custom software to even see the data files in our procession. It is little wonder so few of us make the effort to connect the cycle work of our predecessors to current market data. The only way this will ever change is for you to communicate your needs to your vendor. A soft real estate market means an economic downturn. Knowing this rhythm in real estate sharpens our awareness of the market risks. We knew a housing bubble was blooming with an eminent collapse as its resolution as that is historically how all bubbles burst. But we did not know as traders when the collapse would happen— or did we? As traders we do not establish a position based on this data, but we know and immediately recognize the significance of a
Figure 2.8 Source:
Pig-iron Market Panics 1800-1991
Benner Prophecies, 1875
trigger as it happens and can use our better trading tools with the early recognition that a major new trend is in force. Such analysis
44 • DISPELLING SOME COMMON BELIEFS ABOUT INDICATORS
Dominant Trading Cycles Are Not Time Symmetrical • 45
in one another.
and emotional. We can act because we are prepared with a prede-
the country, destroyed the credit and confidence which men had
prepares us to act decisively when others around us are confused termined game plan in place.
The bank-note circulation began to exceed the total specie in
contraction of the currency began to be felt, multitudes of banks
I have highlighted in bold text.
when the culmination in general business had been reached, and
sion. Please take special notice of the statements Benner wrote that
specie from the country in large quantities, and in the year 1819,
sive that I will reprint a portion of it before we continue our discus-
issues of the United States Bank, and of the local banks, drove
of Future Ups and Downs in Prices is so rare and extremely expen-
banks were projected in various parts of the Union. The united
in his Fibonacci cycle work of 1875. His book, Benner's Prophecies
bank mania had reached its height; more than two hundred new
also referenced real estate prices and commercial booms and panics
the country in the years 1815, '16, and '17, and in the year 1818, the
Mr. Benner's work came well before Mr. Dewey's. Yet Benner
and individuals were broken. The panic producing a disastrous
spread ruin far and wide over the land.
prediction in this book, of a commercial revolution and financial
merchants in this country and others engaged in business, and
financial panics, can be predicted with much certainty; and the
revulsion in trade, caused the failure of nine-tenths of all the
Commercial Revulsions in this country, which are attended with
crisis in 1891 is based upon the inevitable cycle which is ever true to the laws of trade, as affected and ruled by the operations of the laws of natural causes. The panic of 1873 was a commercial revolution; our paper
Two-thirds of the real estate passed from the hands of the owners to their creditors. A banker, in a letter to the Secretary of State, in 1830, describes the times as follows;
will be long recollected.
in this crisis.
stances, property, and industry of every district of the United States
Silver), and banks only suspended currency payments for a time
'The disasters of 1819 which seriously affected the circum-
money was not based upon specie (specie refers to Gold and
among the manufactures and trades in all branches of business.'
1800, coin constituted the bulk of the circulation; after this year
1827 was disastrous and alarming; 1828 was characterized by failures
within that space of time. During all this time, and up to the year
institutions in 1826. The scarcity of money among the trades in
of the United States, nor did any country ever flourish more
1825; there was great revulsion among the banks and other monied
no people en)oyed more happiness or prosperity than the people
spring of 1822; numerous and very extensive failures took place in
the adoption of the Federal Constitution in 1787 to the year 1798,
A sudden and pressing scarcity of money prevailed in the
In the Report of Finances for 1854 and 1855, it is stated that
' according to circumstances until 1834, a year of extreme dullness
tree, they have withered and impaired the healthful condition of
After the year 1828 business continued to be depressed, vibrating
the banks came, and all things became changed, like the Upas
46
• D I S P E L L I N G SOME COMMON B E L I E F S ABOUT INDICATORS
Dominant Trading Cycles Are Not Time Symmetrical
*
a crashing panic, and reaction in all trades, manufactures, railroads,
paper money led to the establishment of new banks.
must come, and in September 1873, we had the culmination—
also correspondingly issued bank notes—each increased issue of
ing was astonishing in the world in the years 1871, 72; but the end
the banks increased their reserves of species (gold and silver), they
and manufacture flourished and was prosperous; our railroad build-
the imports of gold and silver increased to an enormous extent; as
commercial activity was the order of the day, all branches of business
increased very fast, business revised, and in the year 1835 and '36,
country continued to be semi-prosperous until 1870, after which year
in all branches of trade; after which our stock of precious metals
The State banks that had numbered i n 1830 only three hundred and twenty-nine, with a capital of one hundred and ten millions,
47
and industries, which is still going on, and we have not reached hard pan.
commercial revulsion of 1819, we find it was eighteen years to the
hundred and ninety millions.
magnitude compared with other panics. Commencing with the
to seven hundred and eighty-eight, with a capital paid i n of two
years, stand out upon the pages of history of this country i n their
ary, 1837, to six hundred and twenty-four, or, including branches,
The panics of 1819, '37, '57, and '73, during this period of
increased, according to the treasury report, by the first of Janu-
us with all its train of woes.
panic of 1857 was on the same downward grade that had character-
teen years will end in 1891, when the next panic will burst upon
panic year to another. The general direction of business after the
return in the same order; the present cycle consisting of eigh-
History repeats itself with marvelous accuracy in detail from one
panics fifty-four years in their order to make a revolution, or to
the second year in our panic cycles, and is eighteen years from 1819.
the commencement of the repetition of the same order. It takes
consequence of a ruinous fall i n prices. This year of reaction makes
pleted in 1857, and the cycle of sixteen years ending i n 1873, was
of merchants, manufacturers, and all classes engaged i n trade, i n
and twenty years and repeat. The cycle of twenty years was com-
commercial revulsion throughout the country, involving the fortunes
to the crisis of 1873—making the order of cycles sixteen, eighteen,
1837, a suspension of specie payments by all the banks, and a general
crisis of 1837, twenty years to the crisis of 1857; and sixteen years
Mark the result and culmination; a panic! In the month of May,
ized the times after the panic of 1819 and 1837, until all business
his forecast. His book was the first o f its k i n d .
cial revulsion and panic in money, the catastrophe would have been
entered numerous reprints. Figure 2.9 is the cycle he used to make
let me observe here, that i f then had been the time for a commer-
and h i g h correctly i n his book. T h a t is why his book survived and
and up to the year 1865, when a temporary reaction set in. Reader
ner plus or m i n u s a year he d i d very well and h i t the cycle l o w
again improved, and was very active during the war of the rebellion
So how d i d Benner do w i t h his forecast for 1891? A l l o w i n g Ben-
had culminated i n depression in the year 1861, after which trade
ended i n a panic bust where housing prices collapsed. N e w York's
cycle was not then complete. And the commerce and trade of the
I n 1837, 1857, 1873, and 1893 a N e w York residential b o o m
the most deplorable national calamity upon record. However, the
48 • DISPELLING SOME C O M M O N BELIEFS ABOUT INDICATORS
Dominant Trading Cycles Are Not Time Symmetrical
• 49
1999 to early 2000 top led to a panic that pushed the early Internet companies out of grace. Many fell from dot-coms to 'dot gones.' What we are learning in this example is how to use cycle analyFigure 2.9
Index of I n t e r n a t i o n a l Battles f r o m 550-1957
Source: Cycles Classic Library Collection, Foundation for t h e Study o f Cycles, 1987, Edward R. Dewey
sis to give us a sense of where we stand today within an historical cycle of significance. That is far more valuable than the hunt for a single isolated entry point to establish a new trade. Do you want
research. One easy way is to scan the papers from a year before a
September 1836 were selling for $50 in April 1837.
I ' l l leave you to answer that one as it is important to do the
into a depression, lots in the West 100s which cost $480 a piece in
dient that always contributed to the cycle panics of the past?
York Mayor Philip Hone. When banks failed and the economy fell
concerning the valuation of paper currencies. Was this an ingre-
land that sold for $100 to $150 rose to $2,000, according to ex-New
in reserves of gold and silver and the changes in public confidence
were used to supplement historical data for these studies. In 1826
economic panic? Notice how Benner takes notice of the decline
thing for us to do? The New York Times has papers on file that
tant questions, such as, how did these events build into a major
history recorded entirely by electronic means really the smartest
bigger picture is invaluable. From these cycles we can ask impor-
ing cycles is through newspaper advertising. Is having our current
in 1996. Knowing where we are now in the context of a much
1892 to 1893. One of the ways we can research economic hous-
hiding? That is why my first discussion of cycles missed the mark
national economy recorded in the years of 1837, 1857, 1873, and
a fish, or do you want to know where the entire school of fish is
housing busts were caused by recessions (or 'panics') i n the
predictable.
in American business history, Benner described the year 1819 as a
will show you there are similarities in human patterns that can be
Although we are taught as students that this was the worst period
ings, the editorials, the news itself all paint a sentiment pattern that
of 1929 to 1930 was not the worst depression in American history.
market bust sets in. The ads, the real estate listings and job offer-
The Great American Depression that followed the NYSE crash
period when nine-tenths of all businesses failed.
1 have a school in North Carolina that is dedicated to teaching
being right. Most people shrug their shoulders and state that they
this cycle. The cycle mapping human sentiment was spot on and the
so they have to think in ways that will increase their probability of
developed, but not a real estate panic one might try to associate with
turn their technique into a puzzle they had never considered before
to current times, the cycle falls on 1999. In hindsight we know a panic
trader brings to the school, regardless of culture or technique, I will
of 1891 as the start of a panic contraction period and add 54 years up
tunity to learn. Don't run away from the unknown. Whatever the
years, that is, one-third of Benner s cycle. If we take Benner s forecast
is showing how a question or irregularity can become an oppor-
to panic, as Dewey also identified a 55-period quarterly cycle, or I8/3
advanced traders. One of the most important things I demonstrate
It is interesting that Benner stated that it takes 54 years from panic
50
• D I S P E L L I N G S O M E C O M M O N B E L I E F S ABOUT INDICATORS
Dominant Trading Cycles Are Not Time Symmetrical
• 51
the N i k k e i price data is superficial i f the discussion goes no deeper.
that it is correct w h e n they don't really get it themselves.
increase the probability o f our work. L o o k i n g how cycles fall w i t h i n
T h e i r worst response is to take someone else's answer and assume
a broader range o f markets that provide us w i t h balance that w i l l
and want to r u n into the u n k n o w n to better their understanding.
ultimate triggers, asking ourselves questions helps us to focus on
don't k n o w the answer. T h e best traders recognize the o p p o r t u n i t y
statistical study. But first you need to have the data and lots of it.
can make. T h e race is on. But the manufacturing industry for any
F i n d i n g i f one set of data influences another set of data requires a
education is key i n America as we have to make products no one else
the changes i n the earth's magnetic field w h i c h has a similar cycle?
show that manufacturing was the key. It remains so today. T h a t means
to an 11-year cycle. Is one o f the factors influencing weather patterns
recovery after a major business cycle contraction?' B o t h state and
and solar storm activity is 11 years. Agricultural yields are also tied
i n g i n current times. W e also need to ask, 'what contributed to the
T h e cycle interval is 10.8 years. W e k n o w the h a l f cycle for sunspot
about. T h e i r chart work should make us t h i n k about the risk bloom-
study of an Index of International Battles f r o m 550 t h r o u g h 1957.
cycle of consumer sentiment that Dewey and Benner both warned us
Consider f u r t h e r the work o f Dewey. Figure 2.9 shows Dewey's
T h e concerns for a diminishing confidence i n our currency f o r m a
country is tied to the ability to sell its goods elsewhere and that means
F i g u r e 2.10 is Dewey's l o n g - t e r m c h a r t for E u r o p e a n W h e a t
i n this discussion.
correlation, you better realize that G F I is based on a revenue stream
prices. I t is interesting h o w this cycle p e r i o d keeps reappearing
the gold stocks G F I and N E M and find they have a slight out-of-step
Beveridge w h o discovered there was a 54-year cycle i n W h e a t
they need not spend time on analyzing currencies. But i f you analyze
prices f r o m 1500
you need an opinion on currencies. Too many stock traders believe
to 1869. T h i s data was c o l l e c t e d by L o r d
ity i n ingenious ways. H e t h o u g h t outside the box of convention.
you want to be or not. T h i s is not about an analyst becoming a funda-
Dewey studied cycles that mapped h u m a n sentiment and activ-
i n South African Rand currency You are a currency trader whether mentalist. It is only showing you that we must t h i n k about what we are seeing i n our charts, indicators, cycle work, and intermarket rela-
< Trind'
• '
r-i—r-i—r-i
i - • r-r-r
r-'
tionships i n order to do exceptionally well i n this global environment. Should we not be asking ourselves, ' h o w often has the world seen a war break out after a financial collapse?' Is there a relationship i n t i m i n g between the collapse and political instability? Was it a loss o f confidence i n currency that triggered the wars, a drastic change i n the cost and availability of food that led to political instability, or debt that c o u l d not be sustained due to more benefactors t h a n worker bees paying into the social programs? Whatever the
Figure 2.10 European Wheat Prices 1500-1869 Source: Cycles Classic Library Collection, Foundation for the Study of Cycles, 1987, Edward R. Dewey
52 • D I S P E L L I N G SOME COMMON B E L I E F S ABOUT INDICATORS
Dominant Trading Cycles Are Not Time Symmetrical
* 53
suspect was due to concerns about the growing war i n Europe.
ing a cycle, we easily forget that the real purpose is to increase our
the 1941 release was a more open and unveiled dialogue that I
within a single market. If we stop thinking about why we are study-
of the 1941 edition that they are hand corrected. I n addition,
miss i n our own work today i f we only focus on a trading signal
Company, Inc.) i n distribution today. There are so few copies
This is part of the brilhance of his work and what we can easily
probability within a complex multitiered global environment. Trade signals do not occur i n isolation any more.
Gann wrote a segment on page 310 titled 'Why Hitler W i l l Lose the War'. Keep i n mind that this student of markets was making
which he must trade and live. Gann wrote:
easier times membership declines. Dewey was not the only one to
he used this knowledge to develop a theory of the bigger picture in
In economic times of hardship church membership increases; in
fought remain true today. While Gann was a technician of markets,
Was Dewey studying the growth of the Presbyterian church? No.
United States to enter World War U. His views on why wars are
church members in the United States from 1826 through 1948.
a forecast before the Japanese attack on Pearl Harbor caused the
In Figure 2.11 we see Dewey's 9-year cycle in new Presbyterian
study economic cycles through various habits and trends within the general public. The third cycle analyst from the United States is perhaps the
Before going into the details of why Hider will lose the war, it is important to analyze for what the war is being fought. We read much
is being waged over, but when the war is over and the smoke clears
from those i n his revised 1951 edition (Lambert-Gann Publishing
of the seas. This is not a true picture or a true fact of what the war
ities (W.D. Gann & Son, Inc., 1941), describes different charts
that the war is being fought to preserve democracy and the freedom
first edition of his book. How to Make Profits Trading in Commod-
about the fight against Nazism, Communism and Socialism, and
least understood. Few students of Gann's work realize that the
CENTi
1
I
1
,
1
,
,
1
,
,
,
,.
away, the world and the public will know for what the war was fought. This war is no different than any other war. It is being fought for commercial supremacy to determine WHO is to dominate the commerce of the world in the future. Germany wants to dictate
fr 1
t*40
'
—'
prices and dominate the commerce of the world, and England wishes to do the same. Russia would like also to do the same thing. It now appears that the New Dealers in the United States —J
11
• laeO -L
1
1
NMO
1
1
l»00
11 _
JI
-
mo
JI
1, L
would like to dictate everything to the world. England has always
»40
Figure 2.11 Growth of the Presbyterian Church in America (1826-1948)
Source:
Cycles Classic Library Collection, Foundation for the Study of
Cycles, 1987, Edward R. Dewey
fought to maintain freedom of the seas and control the commerce of the wodd, and England knows that it is a part of wisdom to work with the United States, and with Russia to determine
5 4 • D I S P E L L I N G S O M E COMMON B E L I E F S ABOUT INDICATORS
Dominant Trading Cycles Are Not Time Symmetrical • 5 5
will form a pattern of distribution and go nowhere? Yes, there are
trade of the world.
to timing market entry, exits, and can we be warned when a market
United States and England dominate the seas they will rule the
cycles demonstrated in my first edition? Is there a better approach
War because they do not have control of the sea. As long as the
Did I find a solution to the problems inherent to all fixed period
the trade relations after the war is over. Germany will lose the
Comparing TIME PERIODS of the past, Hitler's campaign into Russia has not made the progress that Napoleon made in the same
better tools for trading and we can be warned when to stay away from a market. This alone is invaluable for risk assessment.
with calling today's market turmoil?
ten the day he made the advance into Russia. No chain is stronger
some of our current market strife, but how would Gann have fared
is wrong! It is this writer's opinion that Hitler's 'Waterloo' was writ-
value in the longer term as well. Both Benner and Dewey called
with the pace of Napoleon with his horses and mules. Something
tion to the problem of timing trades, but his work is of particular
war and moving rapidly. Yet the big Blitzkreig has failed to keep up
In my opinion the work of W D . Gann has provided a better solu-
number of days. Hitler has every improved machine for waging
in Gann's writings. Mathematical Formula for Market Predictions:
and Germany will lose the war.
144. Gann describes the importance of the Square of Twelve (144)
the weak point, and that weak point will be the undoing of Hitler,
as the 56-year cycle, or the 'Great Cycle,' because it is based on
another smarter. The other nations fighting against Hitler will find
Gann devoted a tremendous amount of time to what he described
than its weakest link. No man has ever been so smart that there isn't
The Master Mathematical Price, Time, and Trend Calculator. The
techniques applied in recent markets.
true. In his 1952 edition, in circulation today, Gann removed
targets. But first you will want a credible example of his analysis
showed the strength of his research techniques, had been proven
had a chance to look at Gann's Wheel to extract price and time
occurred before any of these statements about the war, which
Great Cycle will be discussed further in Chapter 9 when we have
Gann's 1941 release of How to Make Profits in Commodities
asked to write Chapter 5. It was written in 2006 and the book was
when he revised the first edition.
Thinking from the World's Top Minds' (Bloomberg Press). I was
see he could prove the exact same cycles applied a decade later
edited by David Keller, 'Breakthroughs in Technical Analysis: New
changed some of the chart references. Why? Gann wanted us to
Gann's approach to market timing can be found in the book
not only the commentary with his opinion on the war, but he
in current markets very well. This particular cycle of Gann's was
environment in which they are operating.
approaching on the horizon. So far Gann's cycle has performed
ers or analysts far more aware and knowledgeable about the trading
ones.' I tried to warn readers in this book, as this was a horrific cycle
are difficult to recognize in shorter cycles. Longer cycles make trad-
'Gann's methods warn that the years 2008 to 2012 will be difficult
me that they are of greatest value in the longer term. Human events
released in early 2007. O n page 104 you will find the following,
The more I study cycles of fixed periods the clearer it becomes to
56
• D I S P E L L I N G S O M E C O M M O N B E L I E F S ABOUT INDICATORS
Dominant Trading Cycles Are Not Time Symmetrical
• 57
before risking my own bank account and likely you do too.
failures, political unrest, and food panics stretching across 100 years
examined by chi-square tests and more. I want mathematical proof
correlated to times of great turmoil with wars, business and bank
or diminish the strength of a cycle beat. But when a strong cycle
and easier to interpret.
the cumulative sum of numerous cycles all working to enhance
other technical methods I use seem to become far more effective
1960. Fixed cycle enthusiasts will describe the beat of a cycle as
something will happen. Just give me the when and suddenly all the
The truth is, even the 54-year cycle can disappear as it did in
of history. As a trader it does not bother me that I don't know why
that nature does not depend on fixed cycles.
next bear market will unravel global markets again? Yes. The same
with an offset interval from when it was last seen. The answer is
still living through them now, can Gann analysis warn us when the
reappears, this description does not answer why the period begins
But if 2008 to 2012 were foretold to be rough years and we are
cycle and confluence of supporting cycles point to 2019 to 2021.
Gann was a master of finding confluence targets on all three axes.
The concept of confluence price target zones will be discussed in
others? If there is a market leader do we focus on the timing of
grasp the concept of confluence time targets along the vertical axis.
whether the two dates are connected. Is one market leading all the
of a confluence target zone on the horizontal price axis, it is easier to
2012 that this bear market is not over. Therefore, we need to think
(geometry), and vertical (time) axes. If you understand the concept
begin with, one must accept from Gann's cycle warning of 2008 to
He worked to identify objectives on the horizontal (price), diagonal
However, knowing a target date doesn't end the research. To
that market, such as China, or do we focus on our own markets?
Chapter 6 and is the focus of my book, Fibonacci Analysis (Bloom-
natural law sciences.
lead you to study mundane astrology. But even this is mathematically
the changing intervals between cycle beats can be explained by
astrology. Eventually you will start to make connections that will
The truth is that this is also found along the time axis as well and
are a beginner, focus on astronomy (the science), and stay away from
between targets is irregular because markets expand and contract.
natural law or, in today's language, the study of astronomy. If you
fixed interval between each target zone. The separation or spread
natural law to describe the movement of planets. It was the study of
These target zones do not form a grid on the horizontal axis with a
that books written in the late 1800s and early 1900s used the term
ferent Fibonacci ratios derived from several different price ranges.
subject that is not worthy of print. What people fail to recognize is
different Fibonacci ratios cluster tightly together. They must be dif-
and the laws of vibration. There is a pile of hoopla out there on this
berg Press). Confluence price targets are narrow price ranges where
Gann's time analysis utilizes cycles borrowed from natural law
I use in making up my annual forecasts.' In Gann's book. Tunnel
in natal astrology either. The data from the NASA website can be
p. 116): 'The most important thing of all is the Time factor, which
ogy when you read the phrase 'natural law' and I haven't any interest
following in one of his early works. Truth of the Stock Tape (1923,
is falling; sell all. What you are likely thinking about is natal astrol-
nary testing to develop cycles for time analysis. Gann states the
horoscopes or write an emotionally charged newsletter that the sky
Figure 2.12 shows the China Shanghai Composite with prelimi-
different from what is used to prepare a newspaper column with daily
58 • D I S P E L L I N G SOME COMMON B E L I E F S ABOUT INDICATORS
WBQHT
I
Chapter | 3
-O.OOOOtiaMS -0.0OD2U46IO 0-8002331055 -0.0ODI32451V 0.0061 G9t» I 0C00M67O3I 0.0002415771 0-000203*031 -0,0002M7227 TRANS (»MD)
~3
lO.13.20Ce 17:38MO SV
Figure 2.12
C h i n a S h a n g l i a i C o m p o s i t e w i t h T i m e A n a l y s i s Study,
CHOOSING A N D A D J U S T I N G PERIOD S E T U P FOR O S C I L L A T O R S
A e r o d y n a m i c I n v e s t m e n t s Inc., © 1996-2011, Daily Market Report, www.aeroinvest.com
Source:
Market Trader Gold, Alphee Lavoie and Sergey Tarasov
Through the Air, he states through his fictitious character Robert, ' I have determined the major and minor time factors which repeat in the history of nations, men and markets' (p. 70). The specific event being tested in Figure 2.12 is called a retrograde cycle. We will take an introductory look at a few of Gann's
A
s the Stochastics indicator is one of the more widely used studies in the industry, it is a good place to begin a discussion about
how to set up initial periods. As mentioned in the first chapter, quote
vendors use the same default periods for their studies. Stochastics is generally one of the first studies that novice technical traders will
methods in Chapter 9. While Benner, Dewey, and Gann all had different ways of charting cycles and focused on some that at first appear similar, they had in common something even more valuable; they were thinkers. These men used their cycle work to forecast and time the sentiment changes that impact on their business environment based on the
add to their market data, allowing professional traders to use this knowledge to their advantage for short-horizon market moves. The S&P 500 futures market can be graphed in three-minute time intervals. But clearly this would be ridiculous when the market is moving 20 S&P points in a minute. Markets sometimes adopt entirely different personalities and knowing the technical tool to
rhythmic patterns of the past.
use for a given environment is essential. In flash crash environments I personally do not use normalized oscillators at all that track
Endnotes
between 0 and 100. A simple detrended oscillator from two simple
Dewy, E.R., with Mandino, O., Foundation of the Study of Cycles, New York:
moving averages will keep the trade on and take it off with the best
Hawthorn Books, Inc., 1971. 59
60
•
D I S P E L L I N G S O M E C O M M O N B E L I E F S ABOUT INDICATORS
Choosing and Adjusting Period Setup for Oscillators
* 61
charts on down to the shorter time intervals. I strongly recommend
normalized oscillators does not change.
only after extensive analysis of longer time horizons from monthly
periods 1 and 9 as an example. But the need for and application of
fast waterfall declines. Go down in time rather than longer. It applies
timing. Take a look at the spread between two simple averages of
consider using in the S&P are 100 by 3 and 50 by 3. Why those par-
mind that the two-minute chart is used in emergency situations like
use candlesticks or consider Point-and-Figure charts. Intervals to
Bond markets I rarely drop below 20-minutes. Also, please keep in
data format. If you need to look at your data in a different format,
short time intervals i f the price swings are not as clearly defined.
an analysis method, your wave structure will de distorted in a tick
a 60-minute or daily bar chart. Don't drop down to these extremely
over a fixed unit of time. If you apply the Elliott Wave Principle, as
on the time and price axis and lead you to believe that the chart was
of trades into a single bar as opposed to plotting the trades that occur
12,1998. The data are so well defined that I could remove the labels
that you do not use tick charts as you will be forcing a fixed number
Figure 3.1 is a two-minute chart of the September S&P on June
ticular periods? These setup periods for point and figure charts are RSPU8-2 m i n 06/12/98 C=111250 +1050 -111200
popular with the floor traders in the S&P pit. Even though volume has now switched to electronic trading, old habits still have an influence. Which leads us right back to Figure 3.1 because, if our market
our own position entry or exit, we should monitor that element i f
-110400
can be influenced by a known entity that could shift the timing for
-110800
at all possible. If there is a favored analytic method used by a large -110000
group of traders like a specific period for point and figure charts, -109600
candlestick charts, or simply a common error used by inexperienced traders in mass, that knowledge can be used to your advantage.
stochastic - Slow
9:58
mi
11:??
1^:1
g
_ „
-100.00
The two-minute S&P chart in Figure 3.1 was deliberately set up
taneous global village. Our ability to trade any market is becoming
-20.00
tries with less experienced professional traders are part of the instan-
-40.00
monitored. The 'Stochastics Default Club' remains because coun-
•60.00
with the vendor's default periods so that the retail traders could be
-80.00
easier as exchanges blend together in an effort to survive. What is
U2 m
1^:54
fascinating is that the mass psychology remains the same. I might
Figure 3.1 Aerodynamic Investments Inc., ® 1996-2011, www.aeroinvest.com Source: TradeStation. © TradeStation Technologies
see a reason in a very thin market to delay my order i f the timing is moments away from the Stochastics Default Club. There clearly
62
• D I S P E L L I N G SOME COMMON B E L I E F S ABOUT INDICATORS
Choosing and Adjusting Period Setup for Oscillators • 6 3
do better. It is a promise kept, but we miss you, George.
the retail sector most likely entering the market. The P in the same
the Default Club in seminars so people would push themselves to
interest. In Figure 3.1, a few signals are marked to show an R for
setup Stochastics. He made me promise I would not stop abusing
a Stochastics study that is deliberately set up incorrectly can be of
he referenced my definition for the trader that never learned how to
suggest that something could be out of sync. But then looking at
Sadly George Lane has passed and he used to make me laugh when
tern and momentum indicator combination in my own charts to
long before basic survival instincts motivate us to find a better way
have to be other concerns present first, such as an Elhott wave pat-
chart denotes where professionals are using the retail trader's volume
very forgiving if the period used is just slightly off the mark.
or a W bottom in the indicator like the chart displays at R2, they will
rect cycle to use than one might suspect. However, Stochastics is
the novice trading group three signals in a row and offers divergence
cycles are not symmetrical, and it will be harder to identify the cor-
what they often see demonstrated on the Internet. If Stochastics gives
Club'? Well, forming a new club will be harder to do as dominant
They are doing just what their user's manual told them to do and
right period? Won't we be establishing a 'Modified Stochastics
the 20 line and will sell as the indicator crosses below the 80 level.
will happen when everyone uses the same method to define the
Default Club will enter on cue as the indicator crosses up through
How do we go about finding the right period to use? And what
to establish more favorable entries for themselves. The Stochastics
for defining the correct periodicity for %D is to study the time
to boldly step in front of an oncoming freight train. Splat. It isn't too
In Figure 3.2 the weekly DMK/$ chart is offered. The method
runaway fast market shortly after your indicators gave you permission
period used for the primary oscillator %D, different results will occur.
like the emotional swings of being trapped on the wrong side of a
gave a smoother sine wave than the Fast Stochastics. By varying the
started. Maybe it is just a rite of passage. There is no experience quite
along with %D-Slow as a confirming indicator. The Slow Stochastics
to be disrespectful, as I was once a member myself when I first
%D. George Lane used a three-line oscillator by plotting %K, %D,
chart. Coining the phrase 'Stochastics Default Club' is not meant
primary and %K in a shorter interval to offer a leading indicator to
would be entering orders from entirely different zones from the same
relationship. Stochastics is a two-line oscillator that uses %D as the
different time horizons. Add the range rules for oscillators, and we
tion to develop his overbought-oversold oscillator, which shows this
we are making our trading decisions from different periods and from
bar chart or a monthly time period. George Lane used this observa-
have had an opportunity to trigger a reverse signal. That is because
This concept holds true whether we are trading from a two-minute
their orders will likely be entered before the Default Club charts
to accumulate ever closer to the extreme highs of the daily ranges.
not let the Stochastics Default Club off the hook graciously because
the daily range. Conversely, as price increases, the daily closes tend
three-wave corrective move. The experienced traders generally will
daily closes tend to accumulate ever closer to their extreme lows of
weaker market positions with size (large position size), producing a
Stochastics is based on the observation that, as price decreases, the
generally come out in larger numbers and squeeze out some of the
64
• D I S P E L L I N G S O M E C O M M O N B E L I E F S ABOUT INDICATORS
f X U S D M M U I -Wookly
D M K . ? - WV.Aly
Choosing and Adjusting Period Setup for Oscillators • 6 5
would be used for %D. But how do we know that the 118-period cycle is the correct one to use? In Chapter 2 we developed a chart that needed four cycles just to mark some of the more significant market lows. As we are never going to stumble on the ideal cycle to use in most cases, here is a useful way to evaluate the period you selected. The compressed x axis can now be expanded so that the data appears normal on the screen. Then add the Stochastics study to the chart using a 59-period interval. While the results in Figure 3.3 are correct, I do not like the results from this period because there is insufficient movement in the indicator between zero and one hundred, and the oscillator does not display the range rules discussed in Chapter 1. So the first efforts have established a period
Figure 3.2
that is too long. I f I repeat the process of arbitrarily plotting a bestfit cycle and taking a look at the period results in the Stochastics
horizon in the chart from which you are making trading decisions
formula, it could take several hours to find something I really liked.
and determine the best cycle for that chart. The data i n this chart have been compressed so that as much data as possible can be seen on the screen for this weekly time horizon. We need to see only
1X1 I S )MMUI W o o k l y
l l v $ - VV.,-cklj-
price lows so the temporary distortion of the data is inconsequential. In this chart a cycle of 118 is identified. This cycle is close, but
L
it does not fit the price lows exactly, for reasons discussed in the preceding chapter. Therefore, bias your cycle placement so that the cycle bottoms align most accurately with the price lows in the most recent data, and let the price lows furthest to the left of the chart fall slightly out of phase. Most will establish cycles from left to right and let the price lows closest to the current bar fall out of sync. That just doesn't make much sense.
A *
/
•19000 •18000 •17000 •16000 •15000

¥^
Stoch Slow Custom
A /IT V
/
AA
•14000 •100.00 •80 00 •60.00 •40.00
/
Use the periodicity that is half of the cycle length you identified. As a 118-period cycle is acknowledged in the weekly chart, a 59-period
•20.00 '93
'94
'95
'96
'97
'98
Figure 3.3
66 • D I S P E L L I N G SOME COMMON B E L I E F S ABOUT INDICATORS
Choosing and Adjusting Period Setup for Oscillators
• 67
part of the optimization process will still be of value to us.
as the period to use.
to use Stochastics in such a manner, the profitability test results as
So let's use the computer's smarts to see what it can come up with
in Chapter 1 to mark the upper and lower levels for a trending bull
into a single period to use as a guide or starting point.
in Chapter I are now present. The four boundary lines discussed
situation we are consciously trying to curve fit the historical data
right period has been identified. How? The range rules discussed
or specialized interpretative ways are not recommended. But in this
results in Figure 3.5 should immediately offer confirmation that the
Curve-fitting technical indicators that are used in unconventional
ahead and add a Stochastics study using a 36 period for % D . The
of 10 to 50 in increments of one unit to identify a better period.
cal range in the database. Having passed the first test, we can go
short. Instead, let the computer run through an optimization range
this is tested when the data are compressed to show the full histori-
know that the 59-period is too long and the default of 10 will be too
be identified. The computer may have found a better cycle, and
Stochastics as an indicator to be plotted in a chart. Regardless, we
chart. After setting a 72-period cycle, comparable price lows can
over System, but their default value is different when you select
that a 72-period cycle will be dominant in this weekly D M K / $
TradeStation uses a default value of 10 in their Stochastics Cross-
will be double the Stochastics period. The computer is indicating
optimize a period when you want to edit the default value. Omega
value to use for % D . Working backward, we know that the cycle
In Figure 3.4 the Omega TradeStation software has the ability to
The computer optimization suggests a period of 36 is the best
Professional charting products all have an optimization feature.
The optimization feature for the Omega TradeStation Stochastics system will use the buy-sell signals generated when the Stochasn
D M K , $ - Weekly
tics oscillators cross through the 80 and 20 levels. As we do not want
hXUSDh Format System: Stochastic Crossover 06J19/9S C= JProperties
000 ^^000
Edit Input Input Name: Input Type: Start [To' E Ofitimize
LENGTH Numeric Simple Stop [so' Function Wizard..
xj Verified:
0
No
0
lnc|l
0
Set default..
Stoch Slow C
tnM. Figure 3.5
Figure 3.4
68 • D I S P E L L I N G SOME COMMON B E L I E F S ABOUT INDICATORS
Choosing and Adjusting Period Setup for Oscillators
• 69
high is not the time to sell. It rarely is as this indicator will first top at
a bear market.
the Composite Index. Like the previous example, the momentum
the lower range for a bull market and the upper resistance zone for
futures. A vertical line on October 31 shows a momentum high in
and bear market have been reduced in this chart to show only two:
The price data show that the computer gave us buy-sell signals when the indicator crossed through the 20 and 80 levels. This is not how George Lane advises us to use his indicator, and we can see that these signals offer poor results. He advises trading only with the
the top of wave iii of 3 implying more to follow. Patience is needed to learn the characteristic swing action of any oscillator you favor to use. TradeStatJon Chirt Analyst..
i8'ji£-_
_
USZll - 240 min U140 25/32 3/32 0.07% 6=140
trend when using this indicator—that is, to buy or sell when permission is obtained from divergence analysis and when the 'permission' signal is accompanied on lower volume. However, when the correct period is identified for the Stochastics Oscillator, it will contain trend information by traveling within the ranges that were detailed in the Chapter 1. While the oscillator character between RSI and Stochastics will be different, the signals to buy or sell will abide by the same range rules. The Stochastics indicator in Figure 3.5 has been marked to show the buy-sell signals generated when you apply range rules. The signals are occurring in the direction of the market's trend. Every point marked with a buy or sell signal was discussed in the first chapter with the exception of three. There are two buy signals along a trend line drawn on the oscillator that occurred in 1996. The addition of trend lines and moving averages is of great value and will be discussed in greater detail separately. The other signal
eb RSI avgi (14,9,45) 51.27
55.83
is the Stochastics peak that has the word 'Elliott' above it. Permission to sell is not given by Stochastics just because it is at an extreme high. However, the rally that unfolds from the low identified with a dotted line to the high is a distinctfive-waveadvance that someone using the Elliott Wave Principle along with Stochastics would have known to act upon, or to at least unwind, or scale back a portion of a long dollar position. Figure 3.6 shows the December 2011 T-Bond
2 ^
Ott
19/7
19/13
lO/H ' 18/25
NU
tl/t
Figure 3.6 A e r o d y n a m i c I n v e s t m e n t s Inc., © 1 9 9 6 - 2 0 1 1 , Daily Market R e p o r t , w w w . a e r o i n v e s t . c o m Source: TradeStation. © TradeStation Technologies
7 0
•
DISPELLING S O M E C O M M O N B E L I E F S ABOUT INDICATORS
After discussing how to identify the correct % D periodicity for Stochastics, you might ask, 'When should the period be changed?' It is generally recognized that a shorter period can be used to anticipate a signal. It will also give you false signals in the form of noise. The best
I t ; ; i > l l 8 88 06/15/98 C = 1 0 8 4 5 0
mm -2800
1 IjitlniU^^ J
'
I
T
Choosing and Adjusting Period Setup for Oscillators
1
* 7 1
R S P U 8 - 6 0 min
tmmi
114000
|M m
0 = 1 0 8 4 5 0 -2800
111000 110000 |j
way to handle volatile markets is to use two windows with the data
f 1'
displayed in a ratio of 4:1. For example, a 240-minute chart next to
1'
CMB c o m p o s i t e
^
110000 109000 ^
a 60-minute chart will help you filter noise from your indicators and improve the timing. I personally do not use Stochastics as I favor the
75 QO
RSIMvgs
ranges in RSI that develop and the Composite index to increase prob-
II
'
109000
ability. But I am aware of what the Stochastics trader is likely thinking. Early in this discussion the question was asked, ' W i l l we not
sax
fm
m.
, ,Mi
m
sin
Source: TradeStation. © TradeStation Technologies
same procedure of using half the cycle length?' First, people will
Figure 3.7 A e r o d y n a m i c I n v e s t m e n t s Inc., © 1 9 9 6 - 2 0 1 1 , www.aeroinvest.com
be creating a new Modified Stochastics Club if everyone uses the find different cycles, but second, you will want to make one last
5/21
5/29
6/04
6/10
ever, in the 88-minute chart the same data points are directly below
is also the easiest way to catch that an adjustment is required.
near points 1,2, and 3 are all exceeding the nearby averages. How-
will find it is not only an effective way to make an adjustment but it
ing averages appear to be incorrectly defined, as the price highs
ing the bar chart. I have not encountered others that do this. You
used to create the bar charts. In the 60-minute bar chart, the mov-
'tweak.' What you will want to change is the time period for creat-
at a very difficult decision point for a trader.
In both charts two oscillators are plotted. The bottom oscillator is
are easy to interpret and are right on their mark when the market is
ratio 4:1. Therefore 88- to 22-minutes would be the better pairing.)
exact same indicators in the 88-minute bar chart trigger signals that
old dogs with nothing to prove, one becomes more humble. Use the
Y, and Z in the 60-minute chart are not very helpful. However, the
displayed 88-minute and 60-minute; when you become one of the
also occurs in the oscillators: the oscillator pivot points marked atX,
ute September S&P 500 bar chart. (I should mention here that I
the moving average at points 1, 2, and 3. Now take a look at what
Figure 3.7 is a comparison between an 88-minute and 60-min-
eral weeks ago, and the key pivots were still working on the left-hand
The only difference between these two windows is the time interval
the screen. In this case, the 88-minutc period of time was set up sev-
plotted with them, and two averages are plotted with the price data.
my fixed-interval averages on prices in the oldest data on the left of
composite of two studies. Both oscillators have two moving averages
down in front of a chart, the first thing I check is the relationship of
developed called the Composite index as it is a formula that is a
How did I know 88 minutes would work at point 3? When I sit
a 14-period RSI, and the middle oscillator is a custom formula I
72 • DISPELLING SOME COMMON BELIEFS ABOUT INDICATORS
side of the screen. Therefore, I would not make a change. As the cur-

Choosing and Adjusting Period Setup for Oscillators • 73
SPU8-13 min (includes Globex session)
•113500
made that if the signals are true on the left, then I should be able to
•114500
rent data coming into the chart is the most critical, the assumption is
•114000 •113500 •113000
•110500
missing its mark. Change the time period of the bar chart. I look at
•111500
the right. In a two-minute chart, the far left or oldest data is clearly
•112500
see accurate signals form in the real-time data entering the screen on
the old points to see if a three-minute chart is a better fit.. maybe a one-minute bar chart. What surprises me most is that I nearly always
•112500 •112000 -111500 •111000
•109500 -110500 RSI 40.94 20,00 80,00
land right on or within a minute of a Fibonacci number. Keep an
RSI 4946 20,00 80,00 *
eye on the old data as it relates to its moving averages. This is really important: consider only the price data and moving averages; never try to adjust the indicators plotted below the data. Let the indica-
'slob'
'^;i'2' '
'M'
•5hy
90,00
^ *
•90,00
•30,00
•30,00
-60,00
•60,00
•eh'i' 'M
'6/06
'6/09
'6/10
Source: TradeStation. © TradeStation Technologies
book created with unconventional time periods, this is the method
Figure 3.8 A e r o d y n a m i c I n v e s t m e n t s Inc., © 1996-2011, www.aerolnvest.com
tors fall wherever they will. When you encounter bar charts in this
•
'
by which they were derived. It is very rare that I change the favored the 13-minute chart forces the RSI to be pressed into the same 80
within intraday periods, and I rarely exceed 240 minutes.
peak with divergence near the 80 level. The same price high i n
periods for specific markets. Clearly, I can use this adjustment only
Some markets such as currencies require longer trading pairings
vals. I n this example the 88-minute chart is forming a second
with longer time horizons such as monthly-weekly chart pairings.
find just one pattern for confirmation between the two time inter-
stops the same time the order is entered. This concept works well
When you look for a market top, use this chart combination to
or buy into within a specific time period. Ensure that you enter
500 futures market and charted i n 88- and 13-minute intervals.
not interpreted in any other way except to find the extremes to sell
ure 3.8 is an 11-period RSI calculated from the September S&P
until you are interested in using it to buy the market. The RSI is
and ignore all other squiggles the indicator may produce. Fig-
of the screen; in fact, don't even look at this screen combination
up a condition so as to look for a specific oscillator formation
any reason. Ignore the fact that the oscillators roll to the bottom
lators do not travel between 0 and 100. In this case we w i l l set
to exit a position as I don't hesitate when a target is realized for
0 and 100. In the first chapter a strong case was made that oscil-
method will help the timing of entering a position. It is not used
period is deliberately used to move an oscillator freely between
the last trigger after everything else you monitor is i n place. This
periodicity of an oscillator. I n the following example, a shorter
level boundary. That's all there is to this signal i f you need only
We need to cover times when you w i l l want to change the
74 • D I S P E L L I N G SOME COMMON B E L I E F S ABOUT INDICATORS
Figure 3.9 Aerodynamic Investments inc., © 1996-2011, Daily Market Report, www.aeroinvest.com Source: TradeStation. © TradeStation Technologies
Choosing and Adjusting Period Setup for Oscillators
• 75
of real-time data. Various u t i l i t y software packages f u n c t i o n differently, b u t they all allow a later evaluation o f the exact screen that was used to enter the trade. W e w i l l discuss some surprising
have missed such an obvious market signal w h e n you looked back?
buy signal for this t i m e h o r i z o n .
on indicators and m o v i n g averages. Ever wondered how you c o u l d
a 360-minute and a n 89-minute chart that completes a f u l l sell-
observations derived f r o m this m e t h o d w h e n we address indicators
more suitable for that market. Figure 3.9 shows the Yen/$ applying
M a n y of the short-horizon chart examples t h r o u g h o u t this book were actual real-time t r a d i n g screens that I took a quick snapshot
It is very possible that it was not there i n real t i m e . H o w do you set the periodicity for other indicators? T h e R S I
capability that we w i l l discuss separately.
position was established or closed. ( I still do this today.) I n this case
w i l l be set at 14 periods because i t has a specific price projection
view of i n order to capture the analytics and prices moments after a
plotted as a histogram that w i l l detrend the differential movement
entire screen is recorded w i t h o u t interfering w i t h t h e collection
w i n d o w below the price data. T h e spread of the two averages can be
as quickly. By just the press o f the p r i n t key o n m y computer, the
simply two m o v i n g averages, one slow, one fast, plotted i n a separate
w o u l d mean that the market w o u l d rally sharply and reverse just
needs to be addressed by p l o t t i n g t w o separate studies. M A C D is
pattern that we w i l l look at m u c h later called an expanded flat. I t
very smooth. Smoothed indicators, however, have a severe lag that
T h e advance f r o m the b o t t o m was t h o u g h t to be an E l l i o t t wave
M A C D is favored by position traders as the indicator's travel is
I was selling into the 1,104 level based on the charts i n Figure 3.10.
76 • DISPELLING SOME COMMON BELIEFS ABOUT INDICATORS
Choosing and Adjusting Period Setup for Oscillators • 77
ods near 6-19-9 and 13-26-9, while the bull market may need two
it easier for a trader to see a precise crossover point.
markets. So the bear market studies may need two studies with peri-
faster moving average crosses through the slower average, making
two MACD studies using slower moving average pairings in rising
of the two averages. The histogram will cross the zero line as the
a lecture given by Gerald Appel at one of the Telerate Technical
using a faster moving average pairing for declining markets and
I learned about the need for two MACD studies while attending
advance, you may need four MACD studies: two MACD studies
moving average pairings for your particular market and time horizon.
default like 13-26-9. As markets tend to correct faster than they
study for MACD and then to do your homework to find the best
be near 6-19-9, and the longer study may use periods closer to the
tive of this discussion is simply to caution you not to use a single
the study, Gerald Appel. A short moving average pairing might
requires some specialized handling, as do all oscillators. The objec-
of 12-26-9 are offered as a starting guideline by the originator of
prices. As you can see from this discussion, the smoothed oscillator
ture sell signals, so two studies are needed. The default periods
mine trend, Gerald Appel tends to favor a 50-day moving average on
for instance, will generally develop late buy signals and prema-
is present so that the correct study combination is applied. To deter-
Figure 3.11. If a single study is used, the MACD in a downtrend,
tions that will serve you best. It becomes critical to know what trend
As a minimum, you will need two MACD studies as illustrated in
You will have to do your own testing to find the pairing combina-
oscillators' travel. A single MACD study should never be used alone.
studies that use 5-24-8 for one, and 5-34-5 or 19-39-9 for the other.
The MACD requires three periods that serve to smooth the
Analysis Seminars offered each year in the 1980s and early 1990s. 'XU;SI)MMl)l-Daily 06/17/98 C = 1 7 8 5 8 - 1 1 6 III0»AVB1 line
DMK,$
D.iily
In my own work I need methods that ultimately offer noncorrelated signals from varying techniques; I do not use the MACD because
17974
my Composite oscillator added to the RSI are better indicators for my needs. This doesn't mean that I view the MACD unfavorably. Rather, the specialized applications I favor in other oscillators are not greatly enhanced if the efforts are duplicated by adding the MACD. But the point here is that considerable time was spent in evaluating the merits of a different formula; then a conscious decision was made. Although the MACD study does not suit my purposes, the underlying premise behind the construction of the MACD indicator and its use for multiple studies will be advantaV'^'V'V'^''Sd'W'b'
'98- V
'M
W
•U 'J
geous to many readers. So don't skip over a technical method just because you may not have an immediate application for it.
Figure 3.11
78 • DISPELLING SOME COMMON BELIEFS ABOUT INDICATORS
SI'Y
mm
CASH SSP 500
Chapter 4 DOMINANT T R E N D L I N E S A R E NOT A L W A Y S FROM E X T R E M E P R I C E HIGHS OR LOWS Figure 3.12 A e r o d y n a m i c I n v e s t m e n t s Inc., © 1996-2011, www.aeroinvest.com Source: TradeStation. © TradeStation Technologies
As the lecture about the MACD progressed, discussing the use of multiple studies and periods, it led to an idea that produced an experimental chart formation. In Figure 3.12 an RSI is plotted with multiple periods. Because it looks like a wave, the study variation is named the 'RSI wave.' However, I found that adding two simple moving averages on top of RSI or Stochastics to be far more valuable than this experimental oscillator display.
W
e all have preexisting mental patterns that will dictate just how constrained our capacity is to learn new ideas or expand
our visual perception beyond the conventional. Our ability to see the critical trend lines in prices or within our indicators will be affected by how strongly we hold onto our preconditioning. This chapter requires active participation from you in order to gain the most from it. Please grab a pencil and a straightedge before we move on, and resist the temptation to turn the page to see the solutions for the different problems. By not looking at the solution, you will consciously learn how you work within boundaries. How you attempt to find a solution is as important as the solution itself. Let s begin with a visual puzzle that you may have seen previously. Even if you have seen this puzzle before, do try to draw an answer as 79
and that others have designed for us. T h e solution is i n Figure 4.2.
lines that you draw, but you may not retrace its path.
requires that we get o u t o f the boxes that we create for ourselves
pencil or R E T R A C I N G a line. You may cross over one o f the four
leaving the center square untouched. T h e key to solving the puzzle
your
by drawing four straight C O N T I N U O U S lines without lifting
try to create a square and attempt to circumscribe it w i t h four lines,
tions carefully. Here is your task: You must connect all nine squares
minds o f this configuration o f nine squares is that we immediately
puzzle is i n Figure 4.1, and it displays nine squares. Read the instruc-
g r i d lines or n o t w i l l make l i t t l e difference. T h e i m p a c t i n o u r
it w i l l be the visual solution to a chart problem that follows. T h e first
Dominant Trend Lines Are Not Always from Extreme Price Highs or Lows • 8 1
80
• D I S P E L L I N G S O M E C O M M O N B E L I E F S ABOUT INDICATORS
for another one? G o o d .
screen a l l serve to precondition us. W h e t h e r you view charts w i t h
pivot that is the solution to the market's current movement. Ready
puter screen. T h e chart's border a n d t h e frame o f the computer
the left to see i f older data just out of view m i g h t contain the price
T h e puzzle is displayed o n paper i n the same dimensions as a c o m -
markets, it at no time states that we may not scroll our data screen to
approximates what we are a l l actually staring at for hours o n end.
grid lines or stay w i t h i n any boundary. W i t h i n the 'rules' of trading
N o t only does i t make the puzzle harder, b u t i t also more closely
instructions state that you had to stay w i t h i n the area marked w i t h
w i t h i n a box that is again surrounded by an outer perimeter border.
tern. Read the directions closely a second t i m e . A t no t i m e d i d the
is i n t e n t i o n a l , a n d i t was also deliberate to frame t h e g r i d lines
a line to be drawn beyond the left boundary, beyond the grid pat-
T h e addition of grid lines i n the background of the nine squares
You may feel this is an u n f a i r quiz because t h e solution requires
T u r n the page and complete the puzzle before reading further.
The Nine Squares
The Nine Squares
Figure 4.2
Figure 4.1
82 • DISPELLING SOME COMMON BELIEFS ABOUT INDICATORS
Dominant Trend Lines Are Not Always from Extreme Price Highs or Lows • 83
for the nine squares. Now try to connect the dots.
you will be asked to connect the nine dots. The dots are labeled 1
in the DJIA chart will be very similar to the pattern that was drawn
trial Average. Just like the puzzle to connect the nine squares, now
opposite page. The four trend lines that will connect all nine points
In Figure 4.3 we have a 120-minute chart of the Dow Jones Indus-
and 9 originates from the intersection of the trend lines and not the
If you get frustrated, look at the solution for the nine squares on the
price high and that the critical trend that connects points 4, 5, 7,
page to see Figure 4.4 until you have attempted to find a solution.
ing that two major trend lines bisect one another just left of the
data out of view to draw the four lines this time. Do not turn the
solution for connecting all nine price lows and highs requires see-
all nine dots using only FOUR trend lines. You will not need any
that always started or ended at the price high in this chart. The
that have been numbered in the chart. Here is your task: Connect
You probably had a tough time if you tried to draw trend lines
to 9, and the dots, of course, are the extreme price highs or lows
DJIY-240 min 56/17/93 C=88294 +1642
6
:
2123
4/06
4/17
4/29
92000
92000
90000
90000
91000
91000
5
5/11
86000
86000
87000
87000
88000
88000
89000
89000
5/21 Figure 4.4
Figure 4.3
The reality? There was a pig in the road—a 'Hog'!
market tests the trend line a third time, it is considered a confirmed
duced a string of emotions, artificial images, and false assumptions.
ing a line that connects two price highs or lows, and that when the
a single-lane bridge. That driver's one single-word warning had pro-
price high. We have been told that a trend line is created by draw-
Dominant Trend Lines Are Not Always from Extreme Price Highs or Lows » 85
84 • DISPELLING SOME COMMON BELIEFS ABOUT INDICATORS
lines, but this preconditioning may prevent us from seeing more
the wrong path and guide us in the wrong direction.
dition us to connect extreme highs and lows easily to form trend
tations of elementary geometry. Preconditioning can lead us down
trend lines as a starting point only. Training and experience con-
It is not wrong, but markets do not operate entirely within the limi-
We need to use the guidelines we have been given concerning
trend. Hogwash! Erase that programming because it is misleading.
with new labels in Figure 4.5.
trates just how preconditioning will influence our ability to make
some of the critical points in this same DJIA chart now marked
first told in my hook Aerodynamic Trading} This is a story that illus-
subtle chart symmetries and formations. Let's take a closer look at
let me digress a moment and share with you a funny story that was
meaningful information that will allow us to act sooner from more
There isn't a trader among us who doesn't enjoy a good story. So
decisions.
From point 1, draw a support trend line to connect points 3, 4,
lines bisect may mark the timing of a market turn.
hogged, more than my fair share. Was it my New York plates that
both directions. Always extend trend lines as the point where trend
driving. I was clearly on my side of the road, and I had not taken, or
and then 8 to 11, then extending the trend lines as far as possible in
drove by, and I became very upset that he seemed so angry with my
of the price high. The apex was formed by connecting points 1 to 5
gestures with his arm out the window and yelled, 'HOG!' He quickly
trend line of greatest importance originates from the apex just left
coming toward me, around that turn. As we passed, the driver made
4, and 5. Point 7 becomes the key to this puzzle as the descending
Near the last bend in the road before the bridge, another car was
in the next bar is on the trend line that defines support for points 3,
Driving along, I approached a narrow bridge on the winding road.
spike drops through the trend line, but the market low that follows
in the north where the Appalachian foothills lead to the mountains.
the high or low that immediately follows the spike. At point 2 the
between New York and Georgia, but Georgia is beautiful, especially
experience, the spikes are better left ignored in favor of connecting
recently relocated from New York. There is quite a culture difference
with the spike or key reversal at point 2. More often than not, in my
gia. The Centennial Olympic Games were approaching, and I had
The important aspect of this trend line is knowing how to work
a narrow country road through the mountains of northern Geor-
and 5, and then mark a resistance level at point 6 prior to a decline.
The story begins on a hot summer's day as I was driving along
trend lines bisect because that point will frequently project
in my life! The immense porker was standing right in the middle of
market support or resistance, but also to find the location where
the bridge. I narrowly missed hitting the largest hog I had ever seen
want to extend trend lines forward to find a price level for future
driver? With no more time to think, I made the last sharp turn toward
I will repeat myself to emphasize this point. Not only do you
prompted him to use the familiar gestures of a Manhattan taxicab
86
•
D I S P E L L I N G S O M E C O M M O N B E L I E F S ABOUT INDICATORS
DJIY-240 min
4/06
2123
C=88294
06/17/98
+1642
7,
Dominant Trend Lines Are Not Always from Extreme Price Highs or Lows * 87
reacts to touching the trend lines because the information is very straightforward to interpret. Frequently traders overlook what happens in the market when 92000
two trend lines cross over one another. In Figure 4.6 there are seven trend lines that cross over at 10 dif91000
ferent points within the chart. Let's start by looking at point a where trend lines 1 and 2 cross. A trend line must be tested by the market
90000
a third time in order to use an intersection point as a warning that a market turn may develop. Trend lines 1 and 2 have both been
89000
tested three times so the crossover at point a was a valid alarm to watch for a possible trend change. When trend lines 1 and 4 cross
88000
at point b, the market is testing trend line 4 the third time, and the crossover can also be used as a possible timing signal for a reversal.
87000
86000 4/17
4/29
5/11
5/21
6/03
Trend lines 1 and 3 cross at point c and would have been useful in
S[> Daily
S & P 500 F u t m - e s - D a i l y
6/15
Figure 4.5
the timing of a market turn before it has occurred. In Figure 4.6 a daily chart of the S&P 500 futures market shows trend lines that have been extended forward as far right as possible. Each trend line is numbered at its point of origin. A l l the trend lines i n this chart, with the exception of trend line 4, illustrate the prior discussion in that it can be seen that trend lines should starjtjrom the price extreme that is behind the key reversal or spike rather than from the actual price low or high. Study the origins of these trend lines carefully. O n your own, you should evaluate how the market
Figure 4.6 Source: TradeStation. © TradeStation Teclinoiogies, Inc.
method. I should add that I use this method only i n longer horizon
provide us w i t h the actual signal to buy or sell.
forecast. Use other indicators and methods i n conjunction w i t h this
T h e operative word here is near this point, as our other tools must
gested to be a time estimate for a possible market t u r n , not a precise
alerting a trader to watch for a trend change near this t i m e period.
Dominant Trend Lines Are Not Always from Extreme Price Highs or Lows • 8 9
88
• D I S P E L L I N G S O M E C O M M O N B E L I E F S ABOUT INDICATORS
charts w h e n I become too fixated on the shorter horizon view.
erately drawn incorrectly to illustrate where the higher probability
to warn me to back up and take a closer look at the longer horizon
w i l l wonder why it has been omitted. Trend line 3 has been delib-
charts. It is the two-by-four across the head that I need sometimes
trend lines 3 and 4 cross. T h e crossover is not labeled, and someone
that helps me keep my perspective w h e n I view shorter time horizon
to examine our other indicators very closely. Just past intersection c,
charts, such as daily, weekly, and m o n t h l y bar charts. It is this tool
T h e intersection of two trend lines is a wake-up call that we need
a trend line. L o o k what happens i n Figure 4.7 w h e n the price h i g h
incorrectly drawn or it is one that d i d not work very well. W h i l e the
key reversals and spikes should be ignored w h e n you start to draw
section at point d is very premature. Either this intersection has been
market h i g h . However, eadier it was stated that i n my experience
upward to the price h i g h i n December 1997 marked X. T h e inter-
trend l i n e 3 is at the price h i g h just one bar b e h i n d the actual
3 intersect. T h e intersection is marked w i t h an X. M o v e your eye
able t i m i n g signal i n this daily chart. I n Figure 4.6 the o r i g i n of
this intersection and move to point d. A t point d trend lines 2 and
Let's go back to trend line 3 w h i c h produced the only question-
point is located f r o m w h i c h to draw a trend line. So please skip past
intersection is close to a pivot i n the market as a price h i g h does follow shortly, there is a problem w i t h trend line 3 that we w i l l look at
HHV
.'•JOO F i i t i
separately i n a moment. A t intersection e trend lines 4 and 6 cross.
•115000
Trend line 6 is being tested for the t h i r d time by the market and can be used as a t i m i n g signal n o w as well as a support level. M o v i n g along to intersection f where trend lines 5 and 6 cross w i l l emphasize that the crossover is a t i m i n g signal for a market pivot and not just a support or resistance price objective. T h e market has a m i n o r setback that corresponds to point f and is of value only as a t i m i n g signal as the market does not use these trend lines as a support level. Point g is not a t i m i n g signal as the price levels near g have been used to establish trend line 7. However, point i where trend line 7 -75000
crosses 6 is an extremely important t i m e projection for a possible trend reversal. T h e final point on this chart h is where trend lines 4 and 5 cross. T h i s point illustrates why the intersection point is sug-
Figure 4.7 Source: TradeStation. © TradeStation Technologies
90
• D I S P E L L I N G SOME COMMON B E L I E F S ABOUT INDICATORS
Dominant Trend Lines Are Not Always from Extreme Price Highs or Lows • 91
we will extend the trend lines as far forward as possible on oscillators.
failure patterns that denote a possible trend change.
ish divergence with price. As recommended for trend lines on prices,
pivots are consistent for what we already know them to be: market
trend lines from oscillator peaks or troughs that form bullish or bear-
the rally in 1998. Draw trend lines so that key reversals and spike
lator peaks or troughs that mean something to us. Therefore, draw
marks one of the more important market turns as it was the start of
extremes. Here is the difference: Draw trend lines only from oscil-
very significant change. Now the crossover of trend lines 2 and 3
trend lines more often than just a conventional line that connects
when trend lines 2 and 3 cross. The new intersection becomes a
lators. The end result will show that the oscillator will respect these
4.8. However, changing the origin of trend line 3 alters the timing
another way that will produce targets of greater significance for oscil-
not been touched in Figure 4.7 from where it was placed in Figure
areas of support. However, as with trend lines on price data, there is
a key reversal is: a market failure. The angle for trend line 2 has
to define areas of resistance or oscillator lows connected to define
itself is a market failure above the trend line. That is exactly what
Oscillators, like price data, can have extreme highs connected
behind the key reversal is used to start the trend line. The key reversal
4 through 10 are all associated with bullish divergences. The first and
indicators so that they can be included.
when the corresponding closing price makes a lower low. Trend lines
quizzes will be offered. But first we need to discuss trend lines on
bullish divergences with price. The oscillator makes a higher low
been drawn using your own mind's eye, so know that a few more
divergences with the closing-price bars. Trend lines 1, 2, and 3 are
rob you of the opportunity to evaluate where the lines might have
lines are drawn by using two oscillator lows or highs that formed
for you to try. Trend lines that you see drawn in a finished chart will
Figure 4.8 shows a daily chart for U.S. Bond futures. All the trend
At the end of this chapter there will be a few more chart quizzes
second oscillator peaks form lower highs, but corresponding closing prices are higher. While Figure 4.8 uses this method of drawing trend lines on the RSI, the method can be applied to any oscillator. Before we move away from Figure 4.8, it should be noted that the scale used to plot RSI or any oscillator is important. The range RSI plotted in Figure 4.8 is altered from the standard default of 0 to 100. If the normal range for an oscillator is 0 to 100, vendors will always plot the indicator using this maximum scale. You want to see the indicator as large as possible. So if the range is within a bull market, you might change the default to a range of 25 to 90. In a bear market the scale might be 15 to 70. You never want to limit Figure 4.8
the details available for interpretation by compressing the scale of
what would seem impossible to the oscillator.
computer screen's height. Otherwise, you are likely to be working
locked while a market produces a meltdown or ballistic rally against
allow for an oscillator to be viewed in at least 50 percent of the
oscillator at an extreme oversold or overbought position can appear
your indicator. In addition, the computer system you use should
Dominant Trend Lines Are Not Always from Extreme Price Highs or Lows • 93
92
• D I S P E L L I N G SOME COMMON B E L I E F S ABOUT INDICATORS
Index a horizontal trend line has been drawn to connect the oscilla-
There is limited value in viewing indicators in this manner.
that have been labeled with the letters a through k. In the Composite
people who want to know only when an indicator is at an extreme.
RSI using a 14-period interval. There are several key oscillator lows
squashed indicator at the bottom of the page. These charts are for
below zero or above 100. Plotted below the Composite Index is the
be said for published chart books that add for our convenience a
that has not been normalized. This oscillator is capable of traveling
ner would be that neither had become a flat line! The same can
separately, but for now it is important to know that it is a formula
one can hope to interpret from either chart displayed in this man-
The Composite Index is a custom formula that will be discussed
differentiate an electroencephalogram versus the RSI. The most
Dollars is displayed. Below the price data is the Composite Index.
the bottom of the screen. On such a scale, it would be difficult to
In Figure 4.9 an 80-minute bar chart of the Japanese Yen per U.S.
from a tickertape-size indicator that fits in a small narrow band at
on the RSI using points f and h. Why was the line drawn at points f
on oscillators. There are two kinds of oscillator: those that have
tor lows at points a, b, and c. A similar horizontal trend line is drawn
Let's move forward and discuss using horizontal trend lines been normalized and will confine the travel of the indicator to stay within a range such as 0 to 100, and those whose formulas allow
FXUSJYMUL-80 min
movement to any extreme. There is great value in using one of each type of oscillator to compensate for a rather serious problem inherent with normalized oscillators such as the RSI and Stochastics. The following question illustrates the problem with normalized oscillators: 'An atom always moves one-half the distance from its current location toward a fixed object. How many times will it have to move from a distance of 1 mile presently to reach the fixed object?' As the atom moves only one-half the distance, it will not reach the fixed object until the diameter of its own size represents more than one-half of the remaining distance. The rate of travel toward the fixed object will become extremely slow the closer the atom gets as the distance covered is smaller and smaller. This is how normalized indicators function, and it is why a normalized
Figure 4.9 Aerodynamic Investments Inc., © 1996-2011, www.aeroinvest.com
Source: TradeStation. © TradeStation Technologies
RSI does not warn that conditions changed u n t i l after the fact at
I n the case of interpreting the RSI, one could establish a long dol-
that the market drop c o u l d have been technically forecasted. T h e
the market used as its major support level w i l l nearly always be tested.
for a rally. I n t e r v e n h o n t h e n occurs to support the Y e n , showing
the RSI makes a m i n o r violation of this trend line. A trend line that
is capable of advancing, prices fail to exceed m i n i m u m objectives
range for the oscillator w i t h i n present market conditions. A t point g
begins to advance rapidly beyond p o i n t e and, t h o u g h the oscillator
number of oscillator lows. T h e line is then viewed as the n o r m a l
true for u p t r e n d i n g markets. I n Figure 4.9 the Composite Index
horizontal line needs to be drawn at the level that shows the greatest
u p w a r d as a market correchon develops, and the opposite w i l l be
tion, we are using this horizontal trend line i n a different way. T h e
I n market downtrends, oscillators w i l l travel rapidly
the oscillator.
far left, the RSI tested the same level as f and h once before. I n addi-
you are on the wrong side of the market if you are positioned
and h and not f r o m the low at point g for the RSI? I f you look to the
Dominant Trend Lines Are Not Always from Extreme Price Highs or Lows • 9 5
94
• D I S P E L L I N G S O M E C O M M O N B E L I E F S ABOUT INDICATORS
lar position at point h as the old range is being tested. Keep i n m i n d
with
p o i n t k, w h i c h is too late.
visible. T h e same should be done for oscillator peaks.
then pulls back as the Yen strengthens (chart prices decline). W h i l e
levels, and the old horizontal trend line w o u l d once again become
W h a t follows f r o m point d is an advance, and the Composite Index
the market drop to new extremes, i t is frequently to these prior
changing.
Index to point d, which warns that the market conditions are
the life of the market or contract, the level w i l l be recorded. Should
trend line. T h e same market move, however, moved the Composite
off the screen to the left, it w i l l no longer be i n view. However, over
RSI. T h e low the RSI makes at i can be interpreted as a test of this
the chart. I f the prior range resumes and this oscillator low scrolls
T h i s is not a buy-sell signal. N o w move your eye to point i o n the
cator, a horizontal trend line should be d r a w n and m a i n t a i n e d o n
that closed at current levels would have the same oscillator position.
I f the oscillator l o w at point d'sa new extreme move for the i n d i -
that oscillators plot prices only on the close, and a sharp spike d o w n
to m a n u a l d r a f t i n g is that the oscillator scale versus t i m e w i l l be
e, the character of the market should be carefully observed.
the indicator should be plotted by h a n d . T h e reason for resorting
may be changing, and once the prior support level is tested at point
ability of trend lines forecasting market turns f r o m an oscillator,
Instead, the Composite Index has warned that market conditions
less so for oscillators unless a change is made. To increase the prob-
correction i n a slightly overextended, but secure, uptrending market.
be used as an indication of the t i m i n g of a market reversal. T h i s is
the second test at point /' could easily be interpreted as a completed
onstrated how the intersection of trend lines drawn o n price could
once again tests the trend line at point /. I f only the RSI were viewed,
O n e last c o m m e n t about trend lines on oscillators: it was d e m -
the Composite Index forms a W , or double bottom, at point e, the RSI
240 characters f r o m side to side w i l l only have 24 lines f r o m top to
oscillator
never a 1:1 ratio for their grid lines. A computer w i t h approximately
price data f o l l o w i n g proportionately. I f y o u pick u p o n l y a single
linear. C o m p u t e r screens are not linear. T h e x axis and y axis are
I n this situation the oscillator advances f r o m p o i n t e w i t h o u t the
n o t i o n f r o m this book, let it be the f o l l o w i n g : When an advances
or declines
disproportionately
to the markets'
movement.
b o t t o m . T h i s w i l l vary w i t h screen resolution. I f the effort is made
9 6 • D I S P E L L I N G SOME COMMON B E L I E F S ABOUT INDICATORS
Dominant Trend Lines Are Not Always from Extreme Price Highs or Lows • 97
once a trend line is started, the angle of the line is set by using key
tance than those drawn on price or indicator by a computer.
for this chart. There are numerous small arrows to show you that,
that intersecting trend lines will have greater accuracy and impor-
Figure 4.11 displays the trend lines that I would have favored
to chart oscillators by hand for longer horizon work, it will be found
a price target from the oscillator. Clearly, if the oscillator could
which you can test your own eye.
level. When we address price objectives, we will reverse engineer
value to the trader than another. There will be two charts with
under the RSI pivot is a time projection that has become a support
lines, but one method of drawing a line could prove to be of greater
to minimize errors that would occur in the most recent data. The T
this chapter. There are no wrong or right answers concerning trend
points that fall in the middle or midsection of the chart in an effort
The time has come to test some of the concepts discussed in
Figure 4.10 is the German DAX Index. Find these trend lines:
decline to point T, it would be of value to know beforehand at what market price the oscillator would realize point T. That is a topic we
• The four major trend lines on price that show why the double top in prices has occurred in the most recent data • Two significant levels of support for this same market with which to identify the nearby objectives for a pullback • In the RSI, the trend lines that mark the levels of greatest current interest for this indicator
will approach at another time. Let's move on to the last trend line quiz and discuss Figure 4.12, displaying a 20-minute bar chart for the September U.S. Bond market. This time you have the Composite Index and the RSI plotted against price. We have discussed drawing trend lines on oscillators that diverge with prices, but trend lines can also be drawn from
DAX-D.-iily
W
• 'A • • 'H
'V '
'J' • • • y
• • 'S' • • 'Q- • • 'N
b' • •••98 •
'F
W
• 'A
'
M '
T
M '
X
V
r.KK.MAN
DAXINDKX
• ^ • • ' V ' V ' V ' ' ' 6 ' ' ' N ' ' ' D 9 8 ' V ' W ' V ••W'l)'
Figure 4.11
Figure 4.10
98
• D I S P E L L I N G S O M E C O M M O N B E L I E F S ABOUT INDICATORS
USDU8-20 mm
SEPTEMBER US BONDS
Dominant Trend Lines Are Not Always from Extreme Price Highs or Lows • 99
toward? I f you k n o w the m o v i n g average that is most likely to be the target, you w i l l be able to calculate a price objective beforehand or recognize the price level once the indicator has declined. T h e Composite Index graph has an indicator peak marked w i t h a black arrow. T h e peak forms underneath
the
point where two moving averages cross over. This is a major signal w h i c h w i l l be discussed i n the next chapter. T h e trend line you draw on the Composite Index should bisect this important peak. (The RSI i n this case is used to f i n d the divergences that form in the Composite Index. Y o u w i l l find that the trend lines on the RSI w i l l be less informative as to the preceding questions.) I n Figure 4.13, t r e n d l i n e A marks the reason that prices have Figure 4.12 A e r o d y n a m i c Investments Inc., © 1996-2011, www.aerolnvest.com Source: TradeStation. ® TradeStation Technologies
stalled at current levels. T r e n d lines 1 and 2 o n the Composite Ur.DU8-2() min
SEPTEMBER VS BONDS
divergences between indicators. I n Figure 4.12 a few key b u l l i s h and bearish divergences between the oscillators are already marked SO that you may be introduced to this concept. As a short-horizon trader, you have two critical questions to answer: 1. W h a t trend line is present w i t h i n the price data to have caused the bond market high i n the last bar to stall? T h e market has already exceeded the nearby m o v i n g average displayed on price; visually you w i l l want to see why the market has not advanced further (marked w i t h a black arrow on the price scale). 2. I f the market pulls back f r o m this trend line that caused the market to stall, what m o v i n g average i n the Composite Index, marked w i t h arrows A and B , w i l l the oscillator likely decline
'6/03
'
'6/05
'
'6/09
'
'6/11
'
'6/15
'
'6/17
' '
'6/19
Figure 4.13
100 • DISPELLING SOME COMMON BELIEFS ABOUT INDICATORS
oscillator are drawn from divergences between the RSI and the
Dominant Trend Lines Are Not Always from Extreme Price Highs or Lows • 101
PC - Doily L:750 i/t - 1 2/8 -0,17% e:74t A:7S0 t/» OiTSZ 2/1 Hi:757 2/1 L«i747 4/8
Composite Index. Trend line 1 is derived from a major bullish divergence formation and, projected forward, it actually identifies the oscillator peak of greatest value to start trend line 3. As trend line 3 originates from a peak that respects trend line 1 and crosses the peak marked with the black arrow, it is moving average B that would offer a price objective for bonds. Both the Price and Oscillator windows have included moving averages. In Figure 4.12 the RSI displays only a single moving average so that the RSI can be more easily seen. In Figure 4.13 both averages calculated from the RSI are displayed. Using averages on indicators is extremely valuable and will be the topic for discussion in Chapter 5. Before moving away from trend lines, we need to take a look
Figure 4.14
duplicate the first line, your task is simply to drag the new line,
price data.
increased when you create a parallel line. If your software can
identify the most important angles to establish channels within the
However, the value of this first angled line is significantly
at a specific application derived from gaps. Gaps actually help us
then drag the copy to a new location.
became so useful that fast estimates could be created by utilizing
copy the first line, just draw a line directly on top of the first and
parallel channels through markets with frequent gaps. The angles
not touch the first line at any time. If you do not have a feature to
often I observed a curious correlation to Gann angles that tracked
being very careful not to change the angle, into a new location. Do
The more experience I acquired with Gann analysis, the more
the gaps alone.
Figure 4.15 demonstrates the new parallel line that was set by
the next major area of support that follows. This is no accident as
angle can be for future levels of price support or resistance.
that the same angle that located support at A3 in Figure 4.14 defines
is extended right. Figure 4.14 shows you at point A3 what value this
support level of the secondary retracement at B3. It is interesting
determined by connecting the bottom of each gap. Then the line
see that one was created from the other. The new line locates the
differentiation. The angle of the line drawn at points A l and A2 is
both lines to the right into infinity, but it is easier now for you to
in most books. For this discussion a gap is a gap without further
it is created by duplicating the first. I really should have extended
next open. Breakaway, running, and exhaustion gaps are discussed
decline in this market. The line is the same length as the first as
grain markets often form gaps from a session closing price to the
using only one point at Bl. Bl was selected as it begins a significant
Figure 4.14 is a daily chart of the Corn futures market. The
102
• D I S P E L L I N G SOME COMMON B E L I E F S ABOUT INDICATORS
Dominant Trend Lines Are Not Always from Extreme Price Highs or Lows • 103
K
- Daily L=7S0 4/1 - 1 2/8 - 0 X 7 %
t:749
MTSt
* / » 0 : 7 5 2 2/S
Hi: 757 2/« UJ747 4/»
ognize that a significant entry level has been realized.
environment. When my software does not offer real Gann tools,
The decline into C5 is accurate and important for a trader to rec-
I have witnessed this relationship for several years in a real-time
Source: TradeStation. © TradeStation Technologies
Source: TradeStation. © TradeStation Technologies
Figure 4.16 A e r o d y n a m i c I n v e s t m e n t s inc., © 1 9 9 6 - 2 0 1 1 , Daily Market Report, w w w . a e r o i n v e s t . c o m
Figure 4.15 A e r o d y n a m i c I n v e s t m e n t s Inc., © 1 9 9 6 - 2 0 1 1 , www.aeroinvest.com
Contents described within books are guarded by book publishers'
this technique is a useful visual guide to estimate where a Gann confluence area may fall within the chart.
contracts. So let me add that the point at C5, a diagonal target, is
Fibonacci technique to create a support or resistance grid through
C5 was a move in the future of when the original angle was drawn.
the pivot. The same problem exists for markets declining. Use the
created by using the gap above C4; therefore, the market decline to
practice because there is often no support of any kind just under
before launching into a strong rally. Recall that the first line was
use trailing stops under prior swing extremes. This is a very risky
extends to October 2010 where it shows C4 is tested two full days
such pivots often have voids under them. It is common practice to
early in the data set when gaps are not being connected. The line
the left of C4. This would be a foolish place to enter a stop because
where the market challenges the line as both support and resistance
bought Corn at C5 a stop does not go under the October low just to
set by using points CI, C2, and C3. Whenever possible use points
detailed in my book, Fibonacci Analysis (Bloomberg Press). If you
it remains parallel to the first. In Figure 4.16 the new third line is
price axis. The method to calculate Fibonacci confluence targets is
parallel lines. Duplicate one of the lines again and be careful that
likely confirmed by a Fibonacci confluence target on the horizontal
Do not stop after you have created a simple channel with two
104
• D I S P E L L I N G S O M E C O M M O N B E L I E F S ABOUT INDICATORS
Dominant Trend Lines Are Not Always from Extreme Price Highs or Lows • 1 0 5
to h o l d the market, fails, and defines a key reversal between B2
than conventions of t r a i l i n g stops under price pivot extremes.
tant is to use the top of the gap at B l . W h a t follows is that B2 tries
it measurable. T h i s is a m u c h better approach to risk management
only one p o i n t is needed to anchor the new line. T h e most impor-
fied. Your risk exposure is to the next target zone, thereby m a k i n g
N o w create a duplicate line. As you are not changing its angle,
the market data and p u t stops under the confluence zone identi-
metric relationships between gaps.
by using the price l o w at A l and t h e n t r u n c a t i n g the small key
T h e n B3 marks the top of another gap showing you there are geo-
futures. A l t h r o u g h to A 2 is a consolidation that can be connected
extremely i m p o r t a n t and is often h o w key reversals are defined.
i n Figure 4.17. T h i s remains a demonstration using the daily C o r n
and A5. T h e market i m m e d i a t e l y closes back above B2. T h i s is
T h e final illustration creates the first line using A l , A 2 , and A3
levels at C2 and C3 a year later. M o s t w o u l d use convention and
and the gap start at A 6 all fall o n this same angled line.
the price h i g h to the left of C I . T h e same angle defines resistance
one gap. T h e line is then extended to the right and points A4, A 5 ,
because it is the end of a clear five-wave E l l i o t t wave pattern f r o m
then set at the gap A3. T h i s is useful w h e n the market only displays
'A'. T h e l i n e was projected forward using o n l y one p o i n t at C I
useful i n methods that utilize geometry analysis. T h e line angle is
T h e final line f r o m C I is again just a duplicate of the first line
reversal at A 2 . T r u n c a t i n g key reversal directional signals is very
connect the h i g h to the left of C I and the pivot h i g h just above B l . T h i s w o u l d have been a deadly error because the line w o u l d cross at p o i n t A 6 and a trader m i g h t be lured into t h i n k i n g it was a resistance area to sell into after the drop f r o m B3. Instead they w o u l d short just at the b o t t o m of the A 6 gap. C o n n e c t i n g highs and connecting extreme lows offers a definition of market trend, but does not identify the most i m p o r t a n t geometric proportions developing w i t h i n a market.
Figure 4.17 Aerodynamic Investments Inc., © 1996-2011, Dally Market Report, www.aeroinvest.com Source: TradeStation. © TradeStation Technologies
chapter S I G N A L S FROM MOVING AVERAGES ARE F R E Q U E N T L Y A B S E N T IN REAL-TIME CHARTS
4i 1 1 ow d i d I miss such an obvious signal like that one?' Chances I
I are that you d i d not miss the signal; it may not have been
present w h e n you were considering the trade. M a n y traders suspect that signals may have appeared differently to t h e m i n real t i m e , but few take the t i m e to really explore the character of the indicators f r o m w h i c h they trade. W h a t do I mean by the 'character of an indicator'? T h e character of an indicator is how it moves across the computer screen relative to other indicators or data. T h e character is also the psychological change we experience because the c o m puter changes the y-axis scale to display for us a range defined by the m a x i m u m h i g h and low w i t h i n a fixed number of bars. As new data enter on the right side of our screen and old data scroll o f f on
107 •
and improves timing. I recall the day the DJIA fell nearly 1,000
lead to a new y-axis scale to accommodate the data.
use 22 and 88-minutes together as a 1 to 4 ratio filters false signals
the left, new extreme highs or lows in price or an indicator may
Signals from Moving Averages Are Frequently Absent in Real-Time Charts • 109
108 • DISPELLING SOME COMMON BELIEFS ABOUT INDICATORS
cepts, though market ranges have changed dramatically.
as possible toward the most current data. It will be important to
levels. Just recognize please these examples remain to to their con-
indicators. Scroll forward from the first bar in your database as fast
was trading intraday off the monthly chart support and resistance
view historical data. Scrolling offers an interesting means to observe
points and recovered much of the loss in a single day. Incredibly, I
Technical analysis software allows traders to scroll backward to
tics, MACD, or for that matter any indicator that has a smoothed
the changes that occur in the most current bar.
the same indicator peak or low viewed at a later time. Slow Stochas-
the still pictures that are normally viewed as static charts except for
appearance of the indicator at time 1. The T2, or time 2, label is
bar charts for the same market. In a sense you will be animating
are looking at the exact same indicator and market. T l shows the
subset or micropicture for what will develop in longer time-horizon
in the left window and T2 on the right. In each figure pairing, you
ter you will see in your indicators in a 15-minute bar chart will be a
All three examples have an oscillator high or low that is marked TI
view a 15-minute chart that has 500 bars in its history. The charac-
intraday bar charts for different markets that remain anonymous.
many new data points as possible for this exercise. As an example,
time grid mark on the x axis. Figures 5.2 and 5.3 are much longer
gained from a monthly chart as the objective is to scroll through as
three-minute S&P bar chart. A good point of reference is the 11:06
data entering the screen on the right. Clearly, not much will be
Figures 5.1, 5.2, and 5.3 are all intraday charts. Figure 5.1 is a
scroll forward and watch the movement that occurs nearest the new
By not looking at specifics and animating our indicators, just for the purposes of observation, we are able to see that the undulations of our indicators may dramatically change their relative positions and scale as new data are added to the right-hand side of the computer screen. Traders must know if the indicators they trade from rescale or operate within fixed boundaries. Do the indicators we use as graph additions on prices or other indicators shift their positions over time? As mentioned eariier, I often take a snapshot or freeze-frame picture of a trading screen just after an order has been entered. The
f
Jw 7 /
T2
purpose is to allow a later evaluation of why the trigger was pulled at that precise moment to enter or exit the trade. These figures are extremely short horizon as this discussion was appropriate for the markets in 1998. However, the figures remain true to the principles being discussed. Markets in 2011 require longer intraday periods. I
P
1 mllllllll /
'IIP'
lliii.,
12/24
11:06 1:00
11:06
Figure 5.1
110
• DISPELLING SOME COMMON BELIEFS ABOUT INDICATORS
Signals from Moving Averages Are Frequently Absent in Real-Time Charts * 111
variable in its formula calculated by applying a moving average, is capable of experiencing the dramatic changes demonstrated in Figures 5.1 through 5.3. The degree to which you will see such a dramatic shift will depend on the following: 1. Consider the formula of the indicator and the type of moving average that is used. Does it use a simple or exponential average, for example? The impact on the current period will be affected by the type of weighting applied to the number of bars being averaged. The smoothing technique selected will also determine how significant it will be for the current bar when extreme ranges drop out of the moving average calculation. 2. The time period or number of bars to be averaged remains the same, but the time interval is volatile. Also, the fewer the elements averaged, the greater will be the displacement that occurs. Short moving averages are extremely dynamic. 3. The size of the window you have chosen to display an indicator will define the number of bars used to chart the y-axis scale. In Figures 5.1 through 5.3, the charts are all narrow-width windows displaying a limited number of bars at any one time. This was deliberately done to exaggerate the distortion possible with indicators plotted from intraday data on a computer using Microsoft's Windows 95/98 platform. As soon as we are given the ability to define our own dimensions for a chart's size, our natural tendency is to overdo our adjustments. If you like your trading screens to display a mosaic of numerous small windows at one time, just beware that you may want to enlarge a single smaller window to full screen size before making a final judgment.
112
• D I S P E L L I N G S O M E C O M M O N B E L I E F S ABOUT INDICATORS
Signals from Moving Averages Are Frequently Absent in Real-Time Charts • 1 1 3
same time interval charted.
peak to the far left of T l has now scrolled off the screen and the com-
low of a prior oscillator extreme recorded a few weeks earlier for the
from T l in Figure 5.1 is that the histogram indicator with the highest
manner. T l was in fact an oscillator at an extreme low equal to the
under resistance. In real time it was not. The reason T2 is so different
So now T l is at least two bars behind. The M A C D operates in this
an interpretation in hindsight that the oscillator was peaking directly
histogram is the differential of the two moving averages on price.
scale and has moved directly under the moving averages, allowing
was not the case at T l , and what complicates the matter is that the
interval T2, the exact same oscillator peak appears much larger in
ideal support was tested by an oversold oscillator in hindsight. This
ages charted above the oscillator at time interval T l . However, at time
The averages near T2 have been recalculated, and it appears that
In Figure 5.1 the oscillator peak is nowhere near the moving aver-
pattern or a price projection system to this oscillator extreme, and we
time environments.
weeks is a major piece of technical information. Add an Elliott wave
averages and make allowances in their interpretation of them in real-
currently at an extreme position that has not been seen in over two
cognizant of the dynamic character of shorter period simple moving
averages are ignored. An oscillator tracking intraday data that are
able to displacement. They are of great value, but a trader should be
be different. In this situation only the histogram is viewed and the
common Fibonacci numbers of five and eight are extremely vulner-
Sometimes the same indicator is used, but the interpretation can
for simple moving averages. Simple moving average periods using the
ignore those that are known to perform pooriy in a different climate.
ing forward calculations as the oldest data drops out of the formulas
tors that have been expressly tested for extreme situations only and
look-back period for the averages no longer uses extremes in its roll-
y axis. When markets are in extreme conditions, I will use indica-
ages relative to the oscillator histogram. The change occurs when the
couple of weeks earlier would reappear as the computer rescales the
There is also a dramatic change between the T l and T2 moving aver-
port for oscillators. In this case an 'extreme' trend line marked a
two moving averages on prices that are charted in the top window.)
historical displacement extremes of maximum resistance and sup-
in all three of the figures is the spread or differential between the
Recall the discussion about using a horizontal trend line to mark
puter has rescaled the y axis to plot the histogram. (The histogram
situations, a trader might have sold at the precise low of T l just prior
their formulas, and they do not know that the market is in a freefall.
you did not know how this indicator performed in different market
indicators for T l are actually off by one bar because of the nature of
Such an indicator move would likely produce a new price low. I f
tion of the indicator until the next bar forward. In Figure 5.2 the
test the prior support lows that defined the range for the last week.
indicators do not see current bars as they cannot calculate the posi-
would also be known that the oscillator would most likely decline to
indicators—use the closing price to calculate the indicator. Some
two weeks ago and had entered a new range for the indicator, it
appeared in real time. Stochastics, RSI, and M A C D — i n fact, most
rebound. As the oscillator had declined to the extremes made over
bar at T l records the oscillator and moving averages as they
know a signal to buy exists at T l that could lead to a sharp slingshot
In Figure 5.2 a sharp market decline develops. The most recent
market.
to this type of trap. It will also be a high-risk scenario for the Elliott
be able to forecast the future travel of their indicators and the
to a short squeeze rally. The breakout trader is the most vulnerable
Signals from Moving Averages Are Frequently Absent in Real-Time Charts * 1 1 5
114
• D I S P E L L I N G S O M E C O M M O N B E L I E F S ABOUT INDICATORS
to pull the trigger with confidence because the signal in real time
offer confirmation that the larger trend will resume.
in fact shift upward into a permanent position will allow a trader
be discussed in the price projection section of this book which will
was being monitored. The knowledge that the moving average will
rules defined for oscillators. There are oscillator patterns that will
the price data remains much higher than the moving average that
such a rebound is a countertrend rally in addition to the range
decide to buy when the market has declined to an objective but
stops. There are specific oscillator patterns that will warn us that
new data have been added to the screen. As an example, we may
that has a minimal capital exposure because we can use tighter
we know they will appear perfectly aligned once 10 bars or so of
least allowing a market position to be entered with the larger trend
we must knowingly jump in front of the indicator signals because
can be applied, allowing trades in the market from both sides or at
part of the historical record constantly building. In some cases,
Once market direction is anticipated, price projection methods
tions once the most recent bar moves to the left and becomes a
pate the next indicator signals before they even appear on the screen.
tors on other indicators, are dynamic and will change their posi-
high-probability patterns that develop in oscillators, we can antici-
It is important to recognize that indicators on prices, or indica-
wave trader who relies on price patterns alone. By understanding the
placement.
seem hard to believe that T l and T2 are identical indicators that
especially if they use moving averages to define initial or trailing stop
jump up and move to just under the oscillator at point T2. It may
so weekly position traders are not excluded from this discussion,
extreme to the left of T l scrolls out of view, the moving averages
weekly charts, the capital exposure in question will be far greater,
compared to the new price lows that develop. Once the oscillator
can be correctly interpreted. While the shift will appear minor in
In Figure 5.3 the oscillator at T l shows bullish divergence
are being viewed at a later time. This demonstration does not discredit technical
analysis;
We must know how our indicators are constructed, the impact of our trading time horizon on these indicators, the scale that is nor-
tor changes become finely woven into the fabric of our intuition.
evaluations. Most analysts are unaware of the dynamic nature
It is only after endless hours of careful observation that these indica-
differences between real-time trade decisions and postmortem
in normal and extreme market movements in real-time conditions.
extremely short horizon trading is even more vulnerable to these
mal for charting an indicator, and then, how the indicator responds
it does, however, discredit some display setups and warns that
after extensive training and preparation. Intuitive knowing is the
calculates and incorporates current data into their formulas will
direct knowing without conscious use of reasoning can come only
the indicators used and understands how their quote vendor
some traders are blessed to have had since birth? I think not, as a
the analyst or trader who does have intimate knowledge about
Is intuition a gut feeling, some sixth sense, or inner wisdom that
that surrounds the current bar on our trading screens. However,
116
• D I S P E L L I N G S O M E C O M M O N B E L I E F S ABOUT INDICATORS
Signals from Moving Averages Are Frequently Absent in Real-Time Charts • 1 1 7
software.
tion about indicator dynamics is new to you, it will open the door to
tics will require some custom setup changes to most quote vendors'
of the indicators displayed in the first three figures. If this informa-
Adding moving averages to traditional plots for the RSI or Stochas-
'Sell.' Some readers will be disturbed by the dramatic displacement
Stochastics, the RSI, and the custom oscillator Composite Index.
The easiest part of our job is to pick up the phone and say 'Buy' or
cussion will focus on applying moving averages to three oscillators:
elaborate study that what we learn becomes effortless and automated.
New Commodity Trading Systems and Methods. For now, our dis-
minimize the results. We need to know our indicators through such
try, a good reference can be found in Perry Kaufman's book, The
the critical timing of making the trade will be destroyed and will
understanding of the different types of averages used in our indus-
worry about how an indicator might change with the passing of time,
has already been well traveled. But if you need to reinforce your
of times until their actions became intuitive. As traders, if we have to
tives to closing prices. I have elected not to cover ground here that
hit at the same time?' Both sports legends had to practice thousands
kinds of moving averages and comparisons of how to use alterna-
process.' Yogi Berra, baseball legend, stated, 'How can you think and
discussions elsewhere in published books that detail the different
happens after the ball hits my hands, it might screw up the whole
and discuss using moving averages profitably. There are numerous
Player Joe Montana once stated, 'If I ever stopped to think about what
will be dynamic in real-time conditions, we can now move forward
the luxury of time to think. Three-time Super Bowl Most Valuable
With the knowledge that moving averages on oscillators or prices
essential key for success in fast-paced environments. We do not have
unknown variables X, Y, and Z within the first lines that state
questions will be answered over time.
ventions. The periods that work best for you should replace the
tor questions before the markets answer them for you. Either way, the
convert these equations for you to fit their system's charting con-
about your own indicators in real-time scenarios. Answer the indica-
ing averages to Omega's TradeStation. Other vendors will gladly
the questions this discussion will undoubtedly leave behind with you
In Figure 5.4 are the PowerEditor formulas required to add mov-
many hours of hard work that must be done. You will need to resolve
look at the results and interpretation.
accustomed to viewing. This will train and prepare us for those
you know how the charts can be constructed, let's take a detailed
respond to market movements and conditions that we may not be
fixed period of 14 for reasons that will be addressed later. Now that
do not trade so that we can experience how our indicators will
L E N G T H ( X ) , PERIOD(Y), and PERIOD2(Z). The RSI will use a
It is also helpful to apply indicators to markets that we normally
ply offers more flexibility. Both of these industry leaders add two
cial market as it did in the 1980s.
it is smoothed by a moving average. This chart alternative sim-
which now acts more like a commodity market rather than a finan-
averages. When a Slow Stochastics is used, or % D is smoothed,
the recent transition that occurred in the 1990s in the S&P 500,
lar is charted with a Fast Stochastics study that includes two moving
begins to adopt some other market's action. A n example would be
In Figure 5.5 the Weekly German Deutsche Mark per U.S. Dol-
occasions when our principal market changes its character and
118 • DISPELLING SOME COMMON BELIEFS ABOUT INDICATORS
Signals from Moving Averages Are Frequently Absent in Real-Time Charts
119
FXUSDMMUL'Weckly
Indicator: Stochastics+Avgs Input: Length(X),PERI0D(Y).PERI0D2(Z]; ploti (FastK(Length),'FastK'l; plot2(FastD(Length),'FastD'l; plot3(Average(FastD(Length),PERIOD),'Plot3'|; plot4(Average(FastDlLength),PERIOD2),'Plot4'l;
Indicator: RSI+Avgs Input: LENGTH(14),PERIOD(Y),PERIOD2(Z); Ploti (RSIlCloseXENGTHj/'Plotl 'I: Plot2(Averagel(RSIlClDseXENGTH)),PERIOD);'Plot2'l; Plot3|XAverageHRSI(Close,LENGTH)I,PERIOD2),'Plot3'1;
Figure 5.4 Aerodynamic Investments Inc., © 1996-2011, www.aeroinvest.com Source: TradeStation. © TradeStation Technologies
Figure 5.5 Source: TradeStation. © TradeStation Technologies
would form i f the oscillator failed to break above the longer aver-
difference is that at point 2 the Stochastics indicator is staying above
may otherwise be easily overlooked.
indicator breaks above the longer average and fails. The important
dous clarification and depth in the oscillators' interpretation that
age. At point 2 the moving average is tested by Stochastics after the
moving averages to their oscillators because doing so adds tremen-
ing a stronger signal than that which is present in this chart.
is bearish i f the indicator fails at this intersection. In real time the
would have been slightly higher than the short-period average, giv-
ing average is crossing down through the longer period, the signal
tions from the real-time chart position, and the Stochastics indicator
moving averages are crossing over one another. I f the shorter mov-
moving averages will shift upward into their permanent chart posi-
where the Stochastics study has rolled up to test the point where two
tor was below the short-period average. As we know, short-period
moving average. We begin the chart evaluation at point I , which is
the shorter average, and in previous attempts the Stochastics indica-
In Figure 5.5 the Stochastics study has a short- and longer period
shorter moving average would most likely not have crossed the lon-
Point 3 shows a double top i n the indicator under the longer
average. Is Stochastics really under the average at point 3? The
average would be accurate, and one can easily see that an intersection
as a double top provided the pattern forms under the trending
averages. The sharp angle of descent in the shorter period moving
moving average. A n M pattern in the indicator can be interpreted
ger period average, but Stochastics would still be failing under the
the same, but the shift in hindsight could have been anticipated.
point 4, as the dollar does in this chart. The oscillator then travels
the %D would have been testing the longer average. The signal is
jump from a trampoline. The market accelerates from signals like
relationship would have been important in real time as more likely
quently respond to such formations in the same way as one would
moving average while %K tests the slower moving average. This
4 occurs at an intersection and is a stronger signal. Oscillators fre-
at point 6 that the %D component of Stochastics is above the shorter
as it is soon to intersect the slower moving average. Therefore, point
pulls back a second time to test the longer moving average. Notice
the average. The angle of ascent in the shorter period is important
important signal. The Stochastics indicator then jumps upward and
%K breaks below the moving average, but the slower %D stays above
other than divergences at extremes, they would not be aware of this
viewed as a market failure signal. At point 4 the faster component
because most traders do not read information from the Stochastics
slower component of Stochastics remains under resistance so it is
Signals from Moving Averages Are Frequently Absent in Real-Time Charts • 121
120 • DISPELLING SOME COMMON BELIEFS ABOUT INDICATORS
most of the important patterns that will be seen between moving
waves very often retrace nearly all of the early first waves.
correction within a developing uptrend. Points 1 through 9 detail
retracements early in their transition to a trend change. Second
than those that formed at point 2, the decline is confirmed to be a
trader should know that currency markets commonly form deep
screen lows opposite point 2, but with prices staying much higher
and second waves is developing. Whether you knew this or not, a
rolls under the averages. As the oscillator then breaks sharply to the
the Elliott Wave Principle, a trader would suspect a series of first
At points 8 and 9 a gradual trend shift develops as the indicator
knowledge about the normal character of currency markets. Using
for both averages has been in force for some time without change.
below. This break requires the use of more than one method and a
which is where it would have been in real time as the angle of ascent
where the oscillator stalls briefly near the averages and then breaks
At point 7 Stochastics remains above both the moving averages,
into highs that form an M-pattern top and breaks sharply. Note
prices that developed at the oscillator lows in 1996 up to the price
is overwhelming that the dollar wants to attempt a rally. However,
market has only attained half the distance it will achieve from the
ket, testing the 45 level in this situation. The information at point 5
this pullback and will be discussed later. But in a quick summary the
Stochastics pullback at point 5 is at the support zone for a bull mar-
than the longer period. This is a price projection that accompanies
is a trader's last warning that the trend reversal is now in force. The
differential; that is to say that the shorter period average is higher
lator stays above the longer moving average for the first time, and it
the signal is present because the two moving averages have a positive
levels, which denotes growing market strength. At point 4 the oscil-
chastics pulled back to the 75 level in an uptrend. Confirmation of
higher. Therefore, the indicator is becoming oversold at higher price
that one would have permission to buy the market when the Sto-
position, but the prices on close for these corresponding lows are
W forms near the 75 level, which conforms to an earlier statement
The oscillator low at point 4 is higher than the current Stochastics
that form in Stochastics after the oscillator low in 1996. The third
oscillator lows that develop just prior to point 5 compared to point 4.
averages on a Stochastics study. Note the series of three W bottoms
There is an extremely important juxtaposition in the Stochastics
122
•
D I S P E L L I N G S O M E C O M M O N B E L I E F S ABOUT INDICATORS
n
Signals from Moving Averages Are Frequently Absent in Real-Time Charts • 1 2 3
In Figure 5.4 the true formula I use has been given to you, but Figure 5.6 shows that you may elect to use two exponential averages or two simple averages. It becomes a matter of how you want the real-time displacement to be handled relative to this indicator, which can be determined only if you evaluate the differences on your own. In Figure 5.6 a horizontal line near the 60 level has been added to help your eye see when the RSI is failing at the upper zone for a bear market or using the zone as support in a bull market environment. The same market and chart was used with just the RSI when we discussed detecting trend reversals in Chapter 1. This chart interpretation adds another dimension of detail. Points 1,2, and 3 are all RSI pivots that occur below the moving average. Points
Figure 5.6
4, 5, and 6 are also straight forward, showing the RSI pivots develop-
highs that followed prior to the current buy signal in Stochastics. We will go into more detail on price projections at another time. Let's
ing above the averages. The RSI at point 7 is more difficult to interpret in this chart. Is point 7 a failure to exceed the averages that will lead to a market
take a look at a chart applying the RSI indicator. The weekly DMK/$* chart is again displayed in Figure 5.6, but this time with a 14-period RSI that includes two moving averages. The two averages in this chart use the exact same period, but one is a simple moving average while the other is an exponential moving average. In my own charts I will use one fast simple moving average and one slower exponential moving average. In Figure 5.6 the signals are discussed for the RSI as the indicator relates to a single average. You need to decide if you prefer to use the simple or exponential moving average formula.
decline? This is the same point that was so distinctive and clearly a buy signal for Figure 5.5 when the Stochastics were discussed. So is something wrong with the RSI, or is this a contradiction between oscillator formulas? What is wrong is that the slower exponential RSI is not displayed in this chart but, if it were present, it would show that the RSI was forming a buy signal at the same time as the Stochastics. But the difficulty is not in the omission of an average; it is in making a blatant observation about an oscillator relative to a moving average without giving it further consideration once the first interpretation has been made. How often are we satisfied because we have formed 'an answer' to the puzzle at hand, after which we
*The D M K / $ charts were not removed in the second edition discussions because I believe the Euro will cease to exist and the D M K / $ will return.
confidently move on to something else?
124 • DISPELLING SOME COMMON BELIEFS ABOUT INDICATORS
There is a wealth of information developing at and around point 7 that has been overlooked:
Signals from Moving Averages Are Frequently Absent in Real-Time Charts * 125
from long-time horizon charts right down into shorter time horizons that increases the probability of the signal in the shortest time horizon used to decide when the position should be entered.
1. The two RSI oscillator lows that straddle the high at point 7 form bullish divergences with prices. 2. This divergence occurs above the 40 to 45 support zone defined as the lower range for a bull market. 3. In addition, the price level on close associated with the oscilla-
A similar formation seen in a 60-minute chart may in fact be the last hidden warning of a ballistic rally that has been building within a market for months. The progression of signals that develops from month, week, day, intraday charts, and through the different time horizons we use for trading a market is in essence
anticipated before they develop.
lator lows, or they can develop in the midpoint of travel and
as time-consuming triangles versus sharp quick pullbacks can be
has become support. These signals can occur at extreme oscil-
tions can be detected before they develop and how patterns such
level of great significance and the former area of resistance
rect sequential order. This is how more complex market correc-
oscillator low is a market that has found a former resistance
travel in order to realign all these domino signals into their cor-
that once formed a peak at a price level that later develops an
warning we need to find an alternative path that a market may
divergence oscillator lows surrounding point 7. An oscillator
a domino effect. A few dominos being out of sequence is the
tor peak at point 3 is the same price area that forms the bullish
An example of how this method is applied is offered in Appendix A,
formed in the monthly chart much earlier.
exactly what I do in my analytic work between international markets.
this signal in a weekly chart may be confirming a signal that
of fitting different time horizons together into a sequential order is
horizon that is longer than the present chart. In other words,
the domino puzzle will fit together that brings results. This aspect
developed earlier in the same chart or within the chart time
simple one?' It is this struggle to understand how all the pieces of
extremes because they are frequently confirming signals that
complex correction would have developed instead of a fairly quick
more powerful than the patterns that form at the oscillator
Often I have been asked, 'How could you have possibly known a
become hidden signals. The hidden signals are in fact far
where the Yen/$ has been identified as the key to understanding the
Our markets are interlinked and complex. Companies that strongly
longer horizon chart. It is this transfer and confirmation of signals
and ultimately dictate the short-horizon direction for the S&P 500.
hidden signals are confirmations for an earlier signal located in a
of falling domino signals throughout the global financial complex
rounding point 7 further. It was stated above that frequently these
would be the first trigger signal that would cause the chain reaction
We need to consider this signal that develops in the lows sur-
in fact were ignored for the S&P in favor of believing that the Yen/$
signal is bullish. The pivot high at point 7 is just a stall for time.
next major move for the American S&P 500 Index. The indicators
With or without the slower moving average on the RSI, the
Mark strengthened.
in foreign languages. Without the techniques offered by technical
could lead to a long-term bear market for the dollar as the Deutsche
language to assimilate the masses of fundamental reports printed
fails to push the RSI through the 60 level. Such an oscillator pattern
us all more closely together. Technical analysis offers us a universal
RSI. The danger area for the dollar will come if the next advance
with the expanding global communication highways that are tying
the support zone defined for a bull market, the 40 to 45 area for the
believe each market is an independent entity have not kept pace
Signals from Moving Averages Are Frequently Absent in Real-Time Charts • 127
126 • DISPELLING SOME COMMON BELIEFS ABOUT INDICATORS
Now we are beginning to apply all the techniques discussed at great
nical analysis is growing.
a daily chart for London's Gilt Bond futures market in Figure 5.7.
direction. It is for this reason that the number of people using tech-
the chart examples. Let's correct this imbalance now and focus on
factors in time to make a correct decision about probable market
Up until this juncture, global rates have been largely ignored in
analysis, we could not possibly process all the global fundamental
additional information to the details we obtain from Figure 5.7.)
that forms between points 8 and 9. The oscillator low is below point
trend lines to this same market to see how they might contribute
ing a dollar rally. However, take a look at the oscillator pivot low
oscillators; cycles have been identified. (In Figure 5.8, we will add
a moving average which are easily interpreted as buy signals preced-
length in the first five chapters. Averages are added to prices and
Returning to Figure 5.6, points 8 and 9 show oscillator lows over
8, though at a corresponding higher price. This is confirmation of an uptrend as the oscillator became oversold at a higher price showing market strength. Points 10, I I , and 12 all form oscillator peaks
LGL-Daily
London G I L T B O N D Futures
below the average. The averages again clearly make the indicator easier to interpret. The reentry signal to buy the market does not actually develop until point 13. An exit signal forms at point 14. When an M pattern forms under an average in the oscillator, it is generally viewed a major sell signal; similarly, a W just above an average would be viewed as a buy signal. In this situation the M pattern declines to the 60 level and uses the old resistance zone for oscillators in a bear market as a support zone, adding confidence that the market will remain in an uptrend. At point 15 confirmation develops, and the rally resumes with renewed enthusiasm. Points 15, 16, and 17 all become important considerations for a major trend reversal that was discussed in Chapter 1. The saving grace for the U.S. Dollar is that the oscillator after pivot high 17 leads to
Figure 5.7
128
• DISPELLING SOME C O M M O N BELIEFS ABOUT INDICATORS
LG L-Daily
London G I L T B O N D Futures
Signals from Moving Averages Are Frequently Absent in Real-Time Charts • 1 2 9
RSI remains below the short-period averages, but it declines only to the 45 level in the RSI, indicating a correction in a bull market. The Composite Index has unlimited range between its upper and lower displacement, so range rules will not apply to this indicator normally. At point 3 the RSI is testing the averages, and the Composite Index is maintaining its distance above the shorter moving average. Both signals are correctly bullish for London Bonds. Moving along to point 4 is bearish divergence between the Composite Index and the RSI. The RSI travels under the double line, staying above the moving averages, and it offers no warning. The cycles on price remain up. There are only two warnings present in this chart that a decline will soon follow: the first is the bearish divergence between the RSI and the Composite Index; the second is the price advance from the bar marked with the first cycle low into the high that corresponds to point 4 in the Composite Index, which is a very
Figure 5.8
clear five-wave rally, applying the Elliott Wave Principle. The fact
between indicators that forms at point 5.
and one longer period. Starting at point 1 on the Composite Index,
the first bounce i n the market is foretold by a bullish divergence
with two simple moving averages added to the oscillator: one short
allows both oscillators to alleviate their overbought condition, and
a 14-period interval. The middle indicator is the Composite Index
which the bond market will decline or of its timing. The correction
ferences between simple and exponential weightings. The RSI uses
an extreme, but it offers no further information about the degree to
image. So the two averages in the chart allow you to explore the dif-
multiple noncorrelating methods. In this case, the RSI is clearly at
longer average makes it hard to see the RSI in a black-and-white
we are certain to be blindsided by the market if we do not develop
exponential moving average on the RSI, but in this market the
and experience one can develop with a single indicator or method,
mentioned earlier, I would normally use a longer period for the
nearby. Point 4 is marked to show that no matter how much detail
same period, but one is simple and the other is exponential. As I
that the Composite Index has topped warns that a correction is
Figure 5.7 includes the RSI with two averages that use the exact
that forms in the RSI. Move to the oscillator peak that develops at
leads toward bullish divergence between indicators at point 2. The
moving average, which is a little easier to read than the formation
in the RSI. Bearish divergence at an extreme is a sell signal that
At point 6 the Composite Index fails to exceed the long-period
there is a sharp divergence between the angle of the peaks that form
and 5.8 in one black-and-white chart, the lines are easily differenti-
a cycle low in price that is important coincidental evidence.
it would be difficult to see all the notations drawn in Figures 5.7
gence a second time at point 7 with the RSI. Point 7 coincides with
1,2, and 3 have also become valuable as drawn on the RSI. While
to new lows, but the Composite Index does not form a bullish diver-
posite Index have become extremely valuable additions. Trend lines
as it is unable to exceed the moving averages. The RSI then declines
previously overlooked pivots. Trend lines 2, 3, and 5 on the Com-
the buy signal at point 5 is premature. The RSI has a similar warning
not discussed in Figure 5.7 have major support or resistance now at
under the crossover point of two moving averages, which warns that
Figure 5.8 shows that many of the pivots in the oscillators that were
point S in the Composite Index. The Composite Index is directly
Signals from Moving Averages Are Frequently Absent in Real-Time Charts • 131
130 • DISPELLING SOME COMMON BELIEFS ABOUT INDICATORS
Following point 7 in Figure 5.7, both oscillators break above their moving averages before prices are able to exceed their own averages.
ated with systems that use color. Now that we have a strong arsenal of tools from which to define
alone is insufficient.
8. Point 8 in the Composite Index holds above the longer moving
methods, which is necessary because knowing the market direction
the Composite Index and the RSI leads to a pullback toward point
market direction, we are ready to discuss various price projection
This is a lead that should not be missed. Minor divergence between
average. At point 9 the trend in the RSI is not broken, but it leads to a push that forms a divergence with the Composite Index. Point 10 is an oscillator low that has broken the averages in the Composite Index. However, note that point 10 is lower than point 8, but the oscillator is at a higher price at point 10. The market has become oversold at a higher price level, denoting strength. The RSI forms a similar pattern between the new low and point 9. The RSI is the first to cross back down through the averages, offering the first warning. At point 11 a bullish divergence develops, and a rally soon follows into point 12. Point 12 is an M pattern in both the RSI and the Composite Index that warns that the cycle low approaching at point 14 will soon be seen. The oscillators then form a W bottom at point 13 which coincides with the cycle low at point 14. The cycle at point 14 is clearly a cycle that could be viewed in a weekly chart, and it is significant that the oscillators form a signal near this same cycle bottom. Now take the same market and add trend lines from the oscillator lows and highs that have formed bullish or bearish divergences.
PART 2
CALCULATING MARKET PRICE OBJECTIVES
chapter ADJUSTING TRADITIONAL FIBONACCI PROJECTIONS FOR HIGHER PROBABILITY TARGETS
W
hen this chapter was written 10 years ago it was a specialization of m i n e at that t i m e . A d d another 10 years o f experience
and the subject became a book released early i n 2006. T h e book is called Fibonacci
Analysis
and it was published by Bloomberg Press.
T h i s chapter should be viewed as a light introduction on the subject that I view as the most critical first step o f any analysis work I do. M y first step i n analyzing any market always begins by identifying the Fibonacci price grid that identifies support and resistance levels. O n l y t h e n do I apply and consider other methods to answer how the market w i l l react to the targets: whether nearby targets w i l l end the trend, i f there w i l l be an i n t e r i m swing, or i f I need to invest t i m e i n evaluating the probability o f m u l t i p l e Fibonacci targets by adding G a n n targets as a confluence test. Oscillators reveal
135
constant t h r o u g h o u t a t r e n d i n g move. T h e early swings m i g h t be
and position leverage w i l l be.
or contract w i t h i n the move. Price s w i n g proportions are rarely
are at c a l l i n g a market reversal, the better your risk management
w h e n the price internals o f a trend begin to proportionally expand
and tell us w h e n to enter the market, h o l d , or exit. T h e better you
a range has to be changed. T h e overly simplistic approach fails
how a market w i l l react to a target. T h e y give us permission to act
Adjusting Traditional Fibonacci Projections for Higher Probability Targets • 1 3 7
136
• CALCULATING MARKET PRICE OBJECTIVES
longest bars within
chart is the single m e t h o d I believe has controlled m y equity curve
precisely
Fibonacci confluence zones develop a grid t h r o u g h o u t a market
character to the price swings that f o r m . We are able to
i f they want to do this for a l o n g t i m e . But u n d e r s t a n d i n g how
as m o m e n t u m dies. Markets have an expansion and contraction
and soon enough everyone learns to pick their battles more wisely
the t e r m i n a t i o n swings may show weakness by b e c o m i n g short
newer traders to recognize. T h e old dog fights leave scars over t i m e
other h a n d , the early swings i n a t r e n d m i g h t be elongated, while
out or trade w i t h smaller size. I t is a valuable f o u r t h o p t i o n for
short and the t e r m i n a t i o n swings m i g h t be exaggerated. O n the
W h e n indicators c o n f l i c t it is a technical signal that means keep
all these years t h r o u g h changing volatility.
how the market is expanding
or contracting
measure
by using
the
the move. A detailed discussion about market
expansion and c o n t r a c t i o n w i l l be f o u n d i n Fibonacci
Analysis,
volatility, or v o l u m e .
low to the market h i g h — t h e n calculate the corresponding price
is t r u e regardless o f t i m e h o r i z o n , type o f market, price range,
f r o m obvious price extremes—for example, f r o m an extreme price
ical g r i d a market is using to b u i l d f u t u r e price movement. T h i s
easy. Just take a measurement of the distance traveled by a market
milestones i n the internals o f a swing w h i c h define the mathemat-
example, many experienced traders believe Fibonacci analysis is
but it is i m p o r t a n t for y o u to k n o w n o w t h a t every market has
It is clear that many people share a few c o m m o n beliefs. As an
significant pivot points.
protective stops tucked safely below or above these Fibonacci price
the longest and strongest rather t h a n the extremes that mark the
orders at these retracement levels are frequently missed. Worse,
most valuable details lie h i d d e n w i t h i n the internal bars that are
straight forward, b u t i n practice traders soon discover that their l i m i t
they stop l o o k i n g at the internal details w i t h i n a data swing. Yet the
0.382, 0.50, or 0.618, relative to the total distance. I t sounds very
T h e most c o m m o n error I see w h e n coaching traders is to f i n d
levels that represent a retracement equal to the Fibonacci ratios
this m e t h o d is applied correctly.
highs and lows. T h i s m e t h o d of using price extremes to define
a stronger feel about how your w o r k i n g screens w i l l appear when
retracements must always be calculated f r o m the extreme price
end of this chapter w i t h actual trading screens to help you develop
A n o t h e r c o m m o n m i s c o n c e p t i o n is t h i n k i n g that F i b o n a c c i
range to be subdivided. You w i l l f i n d four different markets at the
be too simplistic for the reality of the markets they trade or analyze.
waves and any bar that begins a strong move to mark the start of the
Traders fail to consider that their view of Fibonacci analysis m i g h t
directional patterns like key reversals. Always use the start of third
greater risk by having to chase a market move or be left out entirely.
O f t e n the best ranges to subdivide require t r u n c a t i n g market
objectives are washed o u t of the market, forcing a trader to accept
138
• CALCULATING MARKET PRICE OBJECTIVES
Adjusting Traditional Fibonacci Projections for Higher Probability Targets • 1 3 9
a deep retracement it is a warning the trend reversal is far more
the 0.382 retracement calculation were left behind when a sharp
aware of the longest bar within the up trend. I f it is exceeded i n
The traders who made this calculation and entered buy orders at
tell us the larger trend was losing strength. You should also be
price objective for the correction at R, the retracement target.
do not want to see a 0.618 retracement. A full retracement would
extreme price high and low marked on the chart to project the
observation is that bonds are in a bull market in this chart and we
Most traders will calculate the Fibonacci retracements from the
rebound resumed i n the direction of the larger trend. The first
Figure 6.1 is a weekly chart of the T-Bond futures market.
significant. The chart in Figure 6.2 shows the Fibonacci retracements when a change is made to use the strong start of the rally after the key reversal i n the secondary pullback. The price projection range ended at 'R'. The resulting subdivisions contain the 0.382 target that R becomes the exact level for the bond market reversal. In Figure 6.3 is a chart of weekly Crude O i l futures. It gives me a chance to show how a market respects the confluence or clustered areas of Fibonacci ratios. The price pivot just under $120 level is stopped by the multiple cluster of ratios 0.618 and 0.50 at the same price. The subdivided ranges both began from the gray line that runs across to the $84 area on the y axis. The endings of both ranges are the last two top lines within the chart data. Under this area is another confluence zone at $110. The market does pullback, but it is clear from the oscillators that the larger rally was incomplete. Therefore on the next attempt the market challenges the target for major resistance under $120 and fails. The fact the market uses the lower Fibonacci ratios from this resistance method is a way to confirm the accuracy of your selection to start the range. The price high in 2011 was used again after the failure to make a new calculation. The 2011 high to the low of the range used to identify that high risk top, was again Figure 6.1
subdivided into the ratios of 0.382, 0.50, and 0.618. Notice how
140 • CALCULATING MARKET PRICE OBJECTIVES
' TradeStation Chart Analysi..
-
Adjusting Traditional Fibonacci Projections for Higher Probability Targets • 141
n'X
eUS - Weekly U141 26/32 26/32 0.58% Bsl41 25/ .. 144 141 26/32
0.382. 0.500.61^
il3«
120
R
r
112
t 2
irt • cha>mf,9siU 0 26.69 71.37 77.53
Figure 6.3 Aerodynamic Investments Inc., © 1996-2011, Daily Market Report, www.aeroinvest.com Source: TradeStation. © TradeStation Technologies
the market uses these levels as resistance in the down trend back
Figure 6.2 Aerodynamic Investments Inc., © 1996-2011, www.aeroi nvest.com Source: TradeStation. © TradeStation Technologies
towards the lows. If the market had pulled back and then exceeded the resistance levels identified at the 2011 high, new and higher levels of resistance
142
• CALCULATING M A R K E T P R I C E O B J E C T I V E S
Adjusting Traditional Fibonacci Projections for Higher Probability Targets •
143
being a beginner.
same low that started the earlier range selection. T h e n end the
the h i g h . It becomes your only option and you w i l l never get past
w o u l d have to be k n o w n . H o w this is done is by using the exact
see how these methods apply to any t i m e horizon and any market.
Fibonacci clusters.
futures. It w i l l look very similar to the weekly O i l chart to help you
T h e new ranges covering a wider price range w i l l create the new
T h e chart i n Figure 6.4 is a 240-minute intraday view of G o l d
new range on strong bars above those already used i n this chart.
only f o r m w h e n different Fibonacci ratios cluster together at, or very
Fibonacci ratios make a confluence target and the ratios M U S T be
near 1750. D o you see it? It was mentioned that confluence zones
v i d i n g a larger range do not make a confluence target zone. O n l y
However, i n Figure 6.4 is an error i n an i m p l i e d confluence zone
Important. T h e endings of a range and a Fibonacci ratio subdi-
different ones. As example 0.618 and a 0.50 is a target. C h e c k the default Fibonacci ratios that your vendor uses as the
TradeStation Chart Analysis - « GCP) 240 min (CO. eeC(P) - Z « min U1792.80 1.70
_ n.x
0.09% B:17»2,70 A:17M.8« 0:1795.80 Hi:179» ..
start-up as they are often wrong. T h e y should be set at 0.382, 0.500, and
0.618. A n y t i m e y o u add, subtract, m u l t i p l y , or d i v i d e a
F i b o n a c c i ratio y o u w i l l get another F i b o n a c c i ratio. T h e r e fore creating ratios f r o m m u l t i p l e ranges w i l l define levels that are also a d d i t i o n a l Fibonacci values. You w i l l only need three i n your setup. Beware that several professional charting products o n the market still l i m i t the trader f r o m using Fibonacci calculations correctly and keep the user f r o m m o v i n g beyond elementary applications. As an example, C Q G for W i n d o w s draw their Fibonacci ratios i n such a manner that the grid w i l l project backward toward the oldest data w h e n you start f r o m a price low and move the mouse upward toward a price h i g h w h e n a resistance target is required. I f you are calculating support or resistance the subdivisions need to move toward the y axis so they align w i t h the current bar. T h i s allows you to study internal price pivots as they relate to the Fibonacci grid being drawn. I n up trends, w h e n support levels are needed, always start f r o m the price h i g h and move d o w n toward successive price lows. I f you start f r o m the b o t t o m you are locked into p i c k i n g
Figure 6.4 Aerodynamic Investments Inc., ® 1996-2011, Daily Market Report, www.aeroinvest.com Source: TradeStation. © TradeStation Technologies
144
• CALCULATING MARKET PRICE OBJECTIVES
Adjusting Traditional Fibonacci Projections for Higher Probability Targets • 1 4 5
near, a market price. You cannot consider the end of a range as a Hne that satisfies the criteria needed to reveal a major area of resistance. With that clarification a confluence zone does develop at 1750 when a range higher than the ones seen in this chart is added. So I will use the 1750 area in this discussion as major resistance. We see the Gold market respect the target by developing a decline that again breaks below 1700. When the market falls through 1700 do not think the lines under the market are to serve as support levels. While they may on occasion serve both needs, this is not how to use these calculations. To determine support the process must be repeated. To create support you start at a bar high and work down through successive strong price bar lows. Only then do you reveal the true support levels that become major objectives. You w i l l need to read my book Fibonacci Analysis
to learn
how to do more advanced work such as using the target zones themselves to create future price swings beyond a second or third swing relationship. It indeed can be done and you can work backwards from future targets as if they were already on the screen to verify targets for the nearby market swing have been correctly defined.
»/7 •>»
' mi'
»Ai
tkj
ott
iM
18A3
imt
uks'
N»»
Figure 6.5 Source: TradeStation. © TradeStation Technologies
We have discussed only one method of calculating price
the 138.2% relationship of the height from the first swing mark a
next method.
of Fibonacci ratios are obtained. We see the market near 1750 use
we will want to use the price extremes for swing projections in this
and then added to the point selected at price 'C'. From 'C' a series
the discussion about ignoring spikes for retracement calculations,
rise of this market swing. The height is what is actual measured
calculate Fibonacci price expansion swing projections. Contrary to
price objective will be made. Points a' and 'b' are the diagonal or
objectives using Fibonacci ratios. Another common practice is to
us widely used by swing traders. The catch is vendors create the
chart as we discussed in Figure 6.4. However this time a different
respected resistance price target. There is a catch with this tool that
Figure 6.5 continues with the same 240-minute Gold futures
of price label to the corresponding oscillator position.
creates a significant error i n markets that trade i n higher price scales.
wave interpretations, but an effort is made to show the juxtaposition
measurement and project a percentage rather than actual price. It
Adjusting Traditional Fibonacci Projections for Higher Probability Targets • 1 4 7
146
• CALCULATING MARKET PRICE OBJECTIVES
signal a market b o t t o m . W h e n a new price l o w develops to the
different ranges to create the targets of greatest significance.
sponding to this price l o w d i d not display b u l l i s h divergence or
you started i n the right direction. T h e key is to use multiples f r o m
declined to a 0.382 retracement target, the oscillator lows corre-
needs considerable more development to do it justice. But it w i l l get
ket low at 3' was not the f i n a l bottom? W h i l e the price l o w at 3'
and add it to the pivot. T h i s is actually a very brief introduction and
H o w w o u l d the n o n - E l l i o t t trader have k n o w n that the mar-
To avoid this problem I use boxes or the Fibonacci retracement tool
far right tests the support level where t w o m o v i n g averages cross
correspond i n the indicators. T h e indicators are n o t b e i n g given
T h e Composite Index corresponding to the price low at 2 on the
t i o n has also been transferred f r o m the price data to the pivots that
of wave 2 d o w n , w h i c h is to the right of the oscillator divergences.
A M E X C o m p u t e r Technology Index. T h e E l l i o t t wave interpreta-
buy signal. T h e safest buy signal for this market is at the b o t t o m
corresponding m o v i n g averages w i t h the daily price data for the
b o t t o m of wave 5', the indicators diverge and offer the first visible
Figure 6.6 displays m y Composite Index and the RSI w i t h their
over one another. W e have discussed i n previous chapters how this A C Y - Daily
AMEX C O M P U T E E T E C H N O L O G Y
indicator f o r m a t i o n frequently leads to a strong market move. You may want to come back to this chart after we have discussed the E W P i n Chapter 10. T h e RSI range rules for identifying trends that we covered i n Chapter 1 are i m p o r t a n t i n Figure 6.6. D o not overlook h o w the market reacts to the extreme oscillator t r e n d line highs. I t is i m p o r t a n t to p o i n t out that the market reacts and frequently changes direction w h e n the Composite Index and RSI indicators diverge f r o m one another. T h e Composite Index w i l l be discussed f u r t h e r i n Chapter 12, b u t this divergence between the RSI and Composite Index is one of the p r i m a r y reasons for the form u l a . T h e Composite Index warns w h e n the RSI w i l l fail to detect a market reversal before the failure. I f you are familiar w i t h the E W P , notice that waves i i i , 5, and i i i all top at the same Composite oscillator extreme. T h i s is an impor-
Figure 6.6 Aerodynamic Investments Inc., © 1996-2011, Daily Market Report, www.aeroinvest.com Source: TradeStation. © TradeStation Technologies
tant observation to carry into Chapter 10 w h e n we develop wave structure w i t h oscillators to guide us.
148
• CALCULATING MARKET PRICE OBJECTIVES
Adjusting Traditional Fibonacci Projections for Higher Probability Targets • 1 4 9
first advance.
swing objectives for a market decline. N o w we w i l l determine the
price ratios 0.618, 1.00, a n d 1.618 for the next rally relative to the
tions. I n Figure 6.5 we discussed h o w to calculate t h e Fibonacci
nal grid w i l l become our template for c a l c u l a f i n g the F i b o n a c c i
screen i f your software allows you to reposition Fibonacci calcula-
ing a Fibonacci grid f r o m a price l o w and h i g h . T h e first or origi-
ize that you can easily make these projections o n your c o m p u t e r
is k n o w n to both the author and the reader. Let's begin by creat-
(equality w i t h a former move), 1.618, and 2.618. You may n o t real-
for you are being w r i t t e n before the o u t c o m e of the market move
swing projections. T h e most c o m m o n projections are 0.618, 1.00
being created i n a real-time situation. T h e calculations described
Figure 6.5 began the discussion o f h o w to calculate Fibonacci
o f the first. N o w pick u p t h e duplicate or copied Fibonacci g r i d
500 futures contract o n July 9, 1998. T h i s particular example is
culate the first Fibonacci targets. D r a w a second g r i d right o n top
F i g u r e 6.7 is a n 8 8 - m i n u t e bar chart o f the September S & P
Figure 6.7 shows that a strong market s w i n g is selected to cal-
primary objectives for a market poised to make a new historic h i g h .
and move i t , b e i n g careful n o t to change t h e size o f the range. SPU8-88 min
I n this example the Fibonacci g r i d , 'copy 1 , ' shows that the market advanced a n equal distance relative to the first advance. W e can see this relationship into the present h i g h . As the market has already realized this target, we need to f i n d the next higher objective. W h e r e w i l l t h e m a r k e t go? F i g u r e 6.7 shows an X as t h e o r i g i n a l F i b o n a c c i g r i d has another copy overlapping t h e o r i g i nal for a second p r o j e c t i o n . T h e second copy w i l l be added to the first copy to project where the 1.618 price objective is located relative to t h e o r i g i n a l move. T h i s is t h e m e t h o d I referenced i n Figure 6.5 that w o u l d solve t h e F i b o n a c c i expansion tool issue w h e n most vendors add a percentage a n d the result is inaccurate i n higher price scales. Figure 6.8 shows the relocated copy #2 added to the top range o f copy #1. By adding a second copy, we have an exact price objective drawn at the 1.618 relationship. T h e top line i n the second copy is 'V/dd
• '7/07'
at a 2.00, 2.382, 2.5, and 2.618 target. Are we done? T h i s is where most people w i l l stop, but we can do m u c h better. W e can improve
Figure 6.7 Source: TradeStation. © TradeStation Technologies
the price objective accuracy by adding additional price projections
150 • CALCULATING MARKET PRICE OBJECTIVES
Adjusting Traditional Fibonacci Projections for Higher Probability Targets • 151
07/09/98 C=117280 -210
07/09/98 C=117330 -160
SPU8-88 m m
SPU8-88 m i n y #
1.6
2 Position 2
isA^
-118000
^^^M^
Copy#l
Fibonacci Grid
Jj/JVr. x l ) r ^ Ij^'^
Original
l^iU^ko^
Position 1
_ ) ^ 7'^}^
'7/01 '
•119000
•117000
-117000
J90.G0
-90.00
'li^i
•118000
-116000
-116000
-115000
-115000
RSI
RSI
-60.00
-60.00
-30.00
;30.00
'6/29 '
'6/29'
'7/07 '
'7/01'
'7/c/d
• '7/07'
'
Figure 6.9
Figure 6.8
Source: TradeStation. © TradeStation Technologies
Source: TradeStation. © TradeStation Technologies
position 1, then pick it up with the mouse and move it into posior equality price projection identified from the Fibonacci grid at
jections will be made from several meaningful pivot levels.
tion 2. Now this is what becomes extremely important: The 1.00
from different market swing ranges. The calculations or swing proIn Figure 6.9 the price projections remain on the screen from
position 2 (the upper range for position 2) has a confluence zone
have found the price objective of greatest significance.
projection. The next step is to draw a duplicate grid right on top of
you have ovedapping Fibonacci projections that form a cluster, you
will be different Fibonacci ratios than those obtained from the first
and I.50-not 1.618. The 1.618 objective will be too high. When
It begins from a different price low and the range when subdivided
Therefore, the important price objectives are 1.00, the former high,
This is a longer market swing that will create the new calculations.
in position 2 overlaps the 1.50 price target created in Figure 6.8.
swing projection. Start by drawing a Fibonacci grid at position 1.
common with one of the first calculations. The 1.00 projection
the steps developed for Figure 6.8. Now we need to create a second
152
* CALCULATING MARKET PRICE OBJECTIVES
Five days have passed since the explanation for Figure 6.9 was w r i t t e n . I t is n o w July 14, 1998, a n d a breakout rally i n t o new
Adjusting Traditional Fibonacci Projections for Higher Probability Targets • 1 5 3
old ranges. T h e group of clustered Fibonacci ratios fall into place easily, as you w i l l see at the end of the chapter.
objective o f greatest significance. I n Figure 6.10 the 1.50 Fibonacci
of resistance levels between two major targets, i n this case the o l d
ence zones f r o m independent methods overiap, it marks the price
cessfully penetrated the f o r m e r highs. A void space is the absence
that overlap f r o m the different methods. W h e n clusters, or c o n f l u -
the 1.50 relationship w o u l d be the next target i f the market suc-
independent m e t h o d is then compared to identify the major clusters
objective we i d e n t i f i e d f r o m Figure 6.8. W e t h e n i d e n t i f i e d that
are derived f r o m three noncorrelated price projection methods. Each
under the o l d h i g h for several days. T h e o l d h i g h was t h e 1.00
people may not realize is that the support-resistance price targets I use
be seen i n Figure 6.10. T h e chart shows the market consolidated
I have a strong reputation for price projection accuracy. But what
highs has occurred i n the September S & P 500 contract, w h i c h can
h i g h and the 1.500 objective f r o m Figure 6.8 and the 1.00 target (same level) f r o m Figure 6.9. Below the 1.500 objective is a m i n o r resistance level at 1.382. T h e market breaks o u t above the o l d h i g h and produces a fast market c o n d i t i o n right to the 1.382 l i n e that
SPU8 88 min 07/14/98 C=118410
has been extended for your comparison f r o m our original calculations. T h e market then pushes u p to the major objective and again shows respect for our projected target. T h e progression of steps we described i n Figures 6.8 a n d 6.9 w o u l d n o w be repeated as new market pivot levels have been added to our trading screens. W h a t w o u l d have happened i f the market had failed to acknowledge these objectives? I f you are learning, you made a n error. You are likely l o o k i n g at swings and not using strong bars. But i f you d i d the m e t h o d correctly you w i l l f i n d that corrective expanding and contracting triangles w i l l ignore these targets. I t is the only t i m e I have trouble w i t h this m e t h o d and it is so consistent it is a reliable early w a r n i n g that a triangle is f o r m i n g before the swings make it obvious what pattern is developing. T h e reason this occurs is that the market data is rescaling. After a triangle the grid of support or resistance lines may define clustered targets using different pivots. It is not a guess, as the market w i l l be very clear it does not want the
Figure 6.10 Source: TradeStation. © TradeStation Technologies
154
• CALCULATING MARKET PRICE OBJECTIVES
Adjusting Traditional Fibonacci Projections for Higher Probability Targets * 1 5 5
define why the extra effort is w o r t h it.
T h i s intraday target gains significance w h e n the same target levels
now need to cover how to create m u l t i p l e retracement levels and
c o u l d define a m i n o r c o r r e c t i o n i n this market's larger t r e n d .
W e have discussed m u l t i p l e Fibonacci swing projections; we
projection f r o m earlier steps was respected by the market. T h i s
w i t h G o l d and O i l to show you how historical charts help to recall
w i t h i n m o n t h l y and quarterly charts w i l l be sparse.
is to review the historical performance of the methods used. Begin
that develops i n longer h o r i z o n charts, t h o u g h confluence patterns
O n e of the advantages of c a p t u r i n g markets as you trade t h e m
develop i n longer h o r i z o n charts. Always be aware of the target grid
is i n Figure 6.11, w h i c h shows the m o n t h l y and weekly G o l d futures
Fibonacci, G a n n , and the RSI. N o w the three methods I use can
why a trade was entered w h e n and where. T h e first real-time chart
T h e three methods i d e n t i f i e d i n the 1998 first e d i t i o n were
handle any type of market volatility. T h e y are F i b o n a c c i , G a n n , and the E W P . I use the RSI w i t h the C o m p o s i t e Index all the t i m e , b u t not to create price objectives because any price objective m e t h o d I feel must be able to use m o r e t h a n just closing prices. T h i s was p a r t i c u l a r l y demonstrated w h e n the D J I A f e l l nearly 1,000 points i n a single day. T h e Fibonacci confluence zone accurately i d e n t i f i e d the historic support level o f that w i l d day. I was trading intraday off a m o n t h l y chart, b u t the market c o n t i n u e d to respect the F i b o n a c c i c o n f l u e n c e zones. O n l y the t i m e h o r i z o n changed a n d not the methodology. However, I had experienced years earlier, as the volatility increased, that targets derived f r o m RSI were o f l i m i t e d use. We w i l l look at the oscillators I consider to be essential i n later chapters and a few u n c o n v e n t i o n a l applications that increase your p r o b a b i l i t y of being accurate at the c r i t i cal decision point w h e n y o u need to k n o w what w i l l happen at a Fibonacci confluence target. T h e r e is no question that the m e t h o d of greatest importance for m e after Fibonacci and E l l i o t t derived f r o m m o m e n t u m oscillators is G a n n . G a n n is the o n l y m e t h o d I have f o u n d that can warn w h e n a l o n g choppy correchon is about to begin before it happens. B u t adding t i m e analysis w i l l n o t be useful i f your work o n the h o r i z o n t a l axis is weak to start w i t h .
Figure 6.11 Aerodynamic Investments Inc., ® 1996-2011, Dally Market Report, www.aeroinvest.com Source: TradeStation. © TradeStation Technologies
156
• CALCULATING MARKET PRICE OBJECTIVES
market as captured o n January 9, 2009. I n the m o n t h l y chart you w i l l f i n d horizontal lines at 887, 852, 820, 745, and 697. Study the price level at 697 and you w i l l see that there are m u l -
Adjusting Traditional Fibonacci Projections for Higher Probability Targets * 1 5 7
T h e 697 target zone is i d e n t i f i e d as a major area o f support because several Fibonacci ratios cluster together i n this area. T h e 1,004 level was used as the top o f all the ranges selected.
The
at levels P and Q .
i n a box is t h e n subdivided into Fibonacci ratios. Contract l i m i t a -
w i l l always be used. Bottoms of other ranges selected are located
rather than a conventional swing relationship. T h e new projection
ing f r o m 1,004. D o not assume extreme price lows of prior swings
F i b o n a c c i c o n f l u e n c e zone f r o m a F i b o n a c c i c o n f l u e n c e zone
are numerous ranges selected w i t h i n this m o n t h l y chart all start-
m e t h o d . A measured m o v e d is p r o j e c t e d above a n d b e l o w a
I used the top of a second key reversal that fails at 1,004. T h e r e
T h e 1,004 and 1,086 target levels were obtained using a different
ranges. T h e brief spike that runs t h r o u g h 1,004 is truncated and
ing of different ratios develops along the price levels 745 and 852.
lows along lines Q and P show the bottoms of t w o of the selected
tiple Fibonacci ratios overlapping one another. A similar cluster-
ranges end at sequenfial tops of bars. N o f i c e I d i d not state the end
swings have been identified before the first is complete.
lated f r o m ranges that all have a c o m m o n starting low and then the
confluence zone at 745 w i t h an 1,086 target. I t means three market
clustered rafios are the major support levels. Resistance is calcu-
equal. T h e r e is a second box projected f r o m a higher Fibonacci
remove the noise of unnecessary rafios. T h e confluence or areas of
the price differential under the b o t t o m of the box and above it are
zontal line and then delete the Fibonacci calculations entirely to
swing f r o m the l o w to the Fibonacci confluence zone. Therefore
of price support. I often mark the confluence zones w i t h a h o r i -
T h e box is used like an equality
i n the book, Fibonacci
narrow confluence zone of Fibonacci ratios that define the targets
here as to w h y the 697 level was used, b u t you w i l l find i t detailed
T h e horizontal lines across the chart mark an extension f r o m the
tions w i t h Bloomberg prevent me f r o m detailing this method further
Analysis.
of price swings or highs o f enfire trends. Use the internal bars show-
is i m p o r t a n t to mark these accurately as they dictate stop placement
2. Before 1,004 was realized the corrective retracement to 745 or
ing strength; w h e n the confluence zone has a dimension o f w i d t h i t
1. A target was made for the swing up that ended near 1,004.
and you never want stops to fall w i t h i n a confluence zone.
697 c o u l d be made. 3. T h e n a target for the next swing up after the correction c o u l d be made that identified 1,086.
I n the weekly chart of Figure 6.11 it is clear that the 697 target is also the apex o f the triangle i n the previous swing up to the left. T h e market t h e n respected the support target at 697 w h i c h was
the astute reader m i g h t ask: why w o u l d you not buy the 745 target?
oscillators to read how a b o t t o m w i l l f o r m .
t h r o u g h the target zone or tests the same target a second t i m e . But
retracements. T h a t is where I use G a n n analysis, the E W P , and m y
chart. T h e target at 697 is bought when the market passes back up
decline f r o m 1,004 w i l l stop at 745 or 697 by using just Fibonacci
first d r a w n w h e n the key reversal top developed i n the m o n t h l y
I a m often three swings ahead of the market. I do not k n o w i f the
158
• CALCULATING MARKET PRICE OBJECTIVES
Adjusting Traditional Fibonacci Projections for Higher Probability Targets • 1 5 9
plete because it is missing one d o w n w a r d swing. A l l i m p u l s i v e
14, 2008. T h e market has spiked d o w n and back up t h r o u g h the
develops and the detailed chart o f the daily data shows it is i n c o m -
the m o n t h l y a n d daily G o l d charts were captured o n September
just under the 855 confluence zone. T h e f i f t h wave (wave v) then
to see the screens as they appeared then. I n Figure 6.12 we see that
E l l i o t t wave c o u n t w i t h an incomplete decline. Wave iv upends
near 745, a higher support zone, we w o u l d need to go back i n t i m e
745 support zone. However, the daily chart o n the right shows an
To answer the question of why the G o l d market was not bought
waves w i l l subdivide into five-wave structures, so E l l i o t t wave is one reason n o t to buy early. So where is the first possible target for a bottom? I n the m o n t h l y chart the next Fibonacci target is at 726. It is also i m p o r t a n t to notice that the market has yet to react or rebound f r o m support confluence zones i n the m o n t h l y chart. I n the daily G o l d chart the Composite Index is m a k i n g a double bottom but when we cover the oscillator it w i l l be clear this is a low probability for defining a final low as the indicator needs to roll up and over its o w n moving averages. T h a t fits a m i n o r bounce that leads to a last swing down w h i c h causes divergence w i t h the oscillator. T h e next confluence zone under 726.60 is a narrow band where the confluence zone falls w i t h i n the price levels of 697 and 706 as marked o n the chart. T h i s is the target. Why? Because I k n o w the RSI and Composite Index do not develop sustainable bottoms o f f a W formation. T h e y need to b u i l d u p o n divergences and return to a prior oscillator range w h i c h the market creates w i t h i n its o w n chart. T h e range w i l l be a horizontal trend line connecting m o m e n t u m reversals. W e w i l l cover this another t i m e , but the i m p o r t a n t point is that the supporting tools warned the trend remains d o w n and a bounce w i l l only be corrective. To clean u p the screen I must have deleted the 706 p r i c e l i n e s o m e t i m e after September 14, 2008, as i t is n o longer disFigure 6.12 Aerodynamic Investments Inc., © 1996-2011, Daily Market Report, www.aeroinvest.com Source: TradeStation. © TradeStation Technologies
played i n the charts c a p t u r e d o n January 9, 2009.
Remember
these are m y actual trading screens a n d y o u w i l l see the longer
trade longer.
development with possible targets. Wave (4) does not imply the
holding the best fill at the start of a trend. Be patient and you will
of September 14, 2008 shows wave (4). I label the wave under
you can be. There are no bonus dollars either for being the guy
horizon charts do not change all that much. The monthly chart
Adjusting Traditional Fibonacci Projections for Higher Probability Targets • 1 6 1
160
• CALCULATING MARKET PRICE OBJECTIVES
work; figure it out now. Then when your own market adopts that
short horizon trader.
kets you have never seen to see how they work for you. If they don't
work is easily visible next to a trading screen, although I am a
with a hardy laugh; 'Exactly!' he said. 'Use your methods on mar-
also shows you one chart representing some of the long horizon
was to say, 'George, I don't trade Oats.' George was quick to reply
weaves into the interpretation of trend and oscillator position. It
Oats market during an historic parabolic spike. M y first comment
it shows I do use it in all time horizons during trading and it
to me by George Lane. George told me to use my methods on the
bit to reference Elliott structure again, but if this is new to you
I recall one of the best pieces of advice of my career was given
triangle pattern to the left. I know, I am jumping the gun a little
I always need to know entry levels and stop placement before
character, and someday it will, you'll be ready.'
in a market that is going parabolic.
you to pay attention to details and make you more patient to gain
strong trending markets and how to use a multiple price target grid
and close enough estimates, this method of analysis w i l l force
or get out. It was the way to resolve how to establish positions in
side the confluence zone. If you tend to do things in approximate
they both are at extreme displacements that agree, pull the trigger
In rallies I mark the top of the zone as stops must be placed out-
of 1:4. Then use an oscillator on both that is not normalized. When
zone for stop calculations that must be just away from these areas.
secret was to drop down in time and use two charts in a time ratio
the following way. In declines I mark the bottom of the target
It proved to be incredibly valuable advice over the years. The
pulling the trigger. Therefore I use the width of a target zone in
to handle volatility extremes.
that a bigger fish slams your trend as you gloat about how accurate
template screen display as I need all three time intervals available
dollars for unwinding at the exact target. Bank it early in the event
futures in monthly, weekly, and daily charts. This is my standard
took me many novice years to realize that you do not get bonus
of larger trends will be. In Figure 6.13 we have the N Y M E X O i l
having accurate entry levels to leverage into with full positions. It
where gaps will form, and how fast or complex the corrective phases
as the target zone approaches. I make up for getting out early by
spread between these targets can warn how the market can move,
into positions and exit positions early as I unwind with the trend
Fibonacci confluence zones are far more than just a price. The
ence zone away from a target entry zone. So do not guess; I scale
price fall because of the spread between support confluence zones.
aligning price bar ranges. Stops should be at least one full conflu-
it would be corrective. But I also knew it would be a hard deep
which is very repeatable, but which requires you to be exact when
drop that developed in 2006 in O i l . I knew before it began that
be in trouble fast. Keep in mind that this is a very precise method
The drop in Gold in 2008 was similar in character to the
accuracy. The traders who do everything fast and rough guess will
162
• CALCULAIING MARKET PRICE OBJECTIVES
Adjusting Traditional Fibonacci Projections for Higher Probability Targets * 1 6 3
T h i s particular O i l move became the basis o f m y c o n t r i b u t i o n i n Breakthroughs
in Technical Analysis,
edited by D a v i d Keller
(Bloomberg Press). I m e n t i o n these different books so you k n o w where to f i n d more details t h a n a single chapter can offer. Trust me w h e n I say you do not make any money w r i t i n g books. Authors get pennies and 40 percent goes to distributors, w i t h the rest to the publisher. I have to straighten that misconception out as i t sends me crazy. W e cannot design the cover and we cannot set the price a book w i l l be sold at, but the real benefit is intellectual property. I learn w h o is d o i n g what w i t h i n the industry and this helps m e , as the author, grow to new levels o f understanding that I m i g h t not have been able to do on m y o w n . I n F i g u r e 6.13 the weekly chart i d e n t i f i e d t h e 84 p r i c e level as t h e e n d o f a m i d d l e l e g u p i n a large c o r r e c t i o n we E l l i o t t analysts c a l l an E x p a n d e d Flat. T h e five-wave structure that is promised f r o m such a setup fell to major support levels w i t h great accuracy. H e r e is w h a t y o u need to take away f r o m this chart. You w i l l k n o w the character o f the c o r r e c t i o n , a n d k n o w i t is a c o r r e c t i o n w i t h i n a larger t r e n d that w i l l resume, before i t even starts. W e also k n o w f r o m t h e spread o f t h e c o n f l u e n c e zones the character o f the c o r r e c t i o n to c o m e . W h e n t h e zones have w i d e spaces between t h e m w i t h n o or very few single F i b o n a c c i ratios between c o n f l u e n c e zones a crash character w i l l always f o l l o w . T h e 2008 equity crash was k n o w n i n advance because every g l o b a l m a r k e t h a d a s i m i l a r setup t h a t was feeding o f f o f China's g r o w t h : the M i d d l e East, Asia, South A f r i c a , I n d i a , a n d especially Australia. Yet the A m e r i c a n Indexes lagged far b e h i n d . T h e c o n f l u e n c e zones cannot give y o u the t i m i n g , b u t y o u k n o w the character o f the resolution.
164 • C A L C U L A T I N G M A R K E T P R I C E O B J E C T I V E S
Adjusting Traditional Fibonacci Projections for Higher Probability Targets • 165
How do you know it is corrective when the freefall hits? That is where I must lean on other methods—Elliott, Gann, and oscillators as covered in this book. Here in the O i l market you have a clean demonstration of how the support zones will fall within your chart when done correctly. Notice proportional swings through the use of boxes are also added by measuring confluence zones. The daily chart cuts right through a price swing near 4', but the final target is right on at 51.1 will not buy 51, but wait for the first reaction. Figure 6.14 shows O i l eight months later, on September 18, 2007. This chart shows I bought O i l based on the RSI signal. The wave count was complete and the oscillator gave permission to buy when it tested the horizontal range that held the market from 2004 through mid-2006. The O i l contract has rolled over to a new front month several
CMB Composite Q
SO.13 S0.24 67.40
times and the basis spread had changed the actual low in this perpetual contract. How to handle the basis spread differential when contracts roll over is probably a chapter in itself Now is not the best time to step into this tar pit, but we will look at the problem in the Gann discussion later. A note about future contracts when they roll over before I forget: Do not subtract the basis differential from your old Fibonacci confluence zones to estimate the key support or resistance levels in the new futures contract. You must create a whole new grid of confluence zones by selecting ranges specific to the new contract. This is a problem futures traders alone encounter. The best solution is to use Gann contracts and that is why this discussion will be passed over until Chapter 9. In Figure 6.15 the O i l market is captured on March 19, 2008.
n '04
'
'
'
1—'
<
'OS
1 — I
'
<
>
'O*
1— '07
Figure 6.14 A e r o d y n a m i c I n v e s t m e n t s Inc., ® 1996-2011, Daily Market Report, w w w . a e r o i n v e s t . c o m Source: TradeStation. © TradeStation Technologies
The wave counts have been changed to recognize a strong extending
166 • C A L C U L A T I N G M A R K E T P R I C E O B J E C T I V E S
TradeStation Char..
s
i •' _ n
®CL.f(t>l - Monthly NVMEX 1=103.60 -4.90^
_ tiao 0»
TradeStstlon Chart A . ® a . P ( D ) - W««kly NYMEX U103.60
- 4 . « 0 -4.5:
11637
14.04
110.00
V
108,14 100.00% (97.9
f110.00
ifMsn 98.45
i
- iii 11
f,
EESEE ;IVV.VV
Adjusting Traditional Fibonacci Projections for Higher Probability Targets •
167
horizon trades were made throughout the trending move based on shorter horizon charts. A quick study to help you develop a feel of how a chart should look with confluence zones is displayed in Figure 6.16. The daily Nasdaq was captured on February 27, 2007 and this figure shows you that the support confluence zones were all started from the
Figure 6.15 Aerodynamic Investments Inc., ® 1996-2011, Daily Market Report, www.aeroinvest.com
Source: TradeStation. © TradeStation Technologies trend based on the oscillator position that must be considered when developing wave interpretations. The Fibonacci confluence zones are being used to project Fibonacci expansions and at this time 110 was important, though only a top for wave v of 3 of (3). It is important to see the old buy signal in 2007 is still on the chart, as the long horizon target at 110 is the end of the projected move. Then you have to start all over again to calculate the support levels and project the target for a swing up to form wave (5). When confluence zones fall into clean
Source: TradeStation. © TradeStation Teclinologies
from the 50s to the 110 target. This was the big picture as shorter
Figure 6.16 Aerodynamic Investments Inc., © 1996-2011, Daily Market Report, www.aeroinvest.com
clusters these become the best markets to trade. The trade was not
168
• CALCULATING MARKET PRICE OBJECTIVES
same high marked b. This is not the actual price high to the left of the screen. Then the multiple ranges selected end at the levels marked R, Q, and P. The strong freefall the day before stops right
Chapter 7
on the first confluence zone near 1,740. For those readers less familiar with TradeStation, the black highlight price on the y axis of 1,750 is the closing price and not the zone itself Do not overlook the horizontal lines on the Composite Index. That is an historical reference line where the Composite Index defined a new extreme or a horizontal support level. Therefore when both hold support the Fibonacci confluence zone and the indicator become important.
PRICE P R O J E C T I O N S BY REVERSE-ENGINEERING INDICATORS
One Fibonacci zone is marked M I D . It is the mid-point where an equality swing was made i n September of 2007 using the box technique. The swing up was an accurate target that ended wave 'v of 3'. Keep it simple and just look at the top of the second box.
a trading tool. It is a calculation to define a worst case scenario.
over as the market approached the top of the second box.
A
was coming. But the target was exceedingly accurate. The trade was
Because the wave structure was incomplete we again knew more
clarification is needed in this second edition about reverseengineering indicators. This is an analytic method and not
chaos surrounds us.
every price swing pivot was, 'hello, where did you come from and
scenarios. Prepare beforehand because it is more difficult when
once the basics are correctly in place. The question I used to ask of
it is professional analysts who have to guide their firm in extreme
cult chart patterns. Your observation skills will be your best teacher
It might be needed in trading someday, but not every day. Often
Practice with strong trends first and work your way to more diffi-
level that contained both a Fibonacci and Cann Wheel confluence
just the first warning that your bigger picture has begun to change.
some other momentum extreme for y. The level y was the first price
in for something different from what you believed at first, and it is
71 as a 'what-if' scenario to drive the oscillator to point 7 ' and not
mathematical connections; it is an early warning that you might be
I have never been asked why I elected to calculate 'y' in Figure
you expect continues. When it is a lone event it is hard to see the
indeed a real-time calculation and result. However, to my surprise,
the chart. If it is part of the one you are familiar with, the pattern
The example that follows of the USDJPY in Figure 78 was
who's your daddy?' Every pivot has a mathematical family within
zone. Not every confluence zone derived from Cann will overlay the major levels identified from multiple Fibonacci ratios derived • 169
•
170
• CALCULATING MARKET PRICE OBJECTIVES
Price Projections by Reverse-Engineering
Indicators
• 171
dor that offers ASCII text format or CSI/Metastock data format.
overlapping confluence zones. Only then do I play the 'what-if'
application. I suspect the procedure will be similar for any data ven-
my oscillator to see the displacement that correlates with the two
procedure will export real-time FutureSource data files into an Excel
strongest levels that coincide with the Fibonacci analysis. I study
ing a step-by-step demonstration of how to accomplish this task. This
Then I apply Gann as an overlay described in Chapter 9, to find the
your data file, I will get you past this horrendous hurdle by display-
a step in front of it. I create the support grid described in Chapter 6.
reverse-engineering an indicator will be useless if you cannot import
China has been found at this level. So the method that follows has
imported and recognized by Microsoft Excel. As the discussion about
same target, it is as if the market is stating that the Great Wall of
data from the real-time quote system into a format that could be
from different ranges. However, when the two methods identify the
game to see what price is needed in the next bar to drive the oscillator back to that specific displacement. Now you are in better shape to read what was first written without further changes.
The objective of this exercise is to import the data and the indicators displayed in Figure 7.1 into Microsoft Excel for further analysis. Figure 7.1 displays the TradeStation chart of the daily discussed how the moving averages were added to the RSI graph in
bility of correctly interpreting indicators to determine market direc-
Yen/$ with the RSI and two moving averages on the indicator. We
In the first five chapters we discussed ways to increase our probation. Many of the techniques gave us the means to identify precise targets for our indicators. As an example, we know that a trend line drawn from oscillator pivots that form bullish or bearish divergence
KXUSJYMUI
Daily
YENS-Daily
with prices, or another indicator, is a significant area of future support or resistance for that indicator. If we know the chart position toward which an indicator is focused on traveling, then we are left with one unknown variable: the price level required to drive the indicator to that projected chart position. Therefore, we are about to embark on a discussion of reverse-engineering our indicators to define an extreme scenario price objective. Technical analysis software products generally do not give us the ability to reverse-engineer our indicators to solve for an unknown future price, at least not at this time. However, Microsoft Excel does provide us with the means to expand the functionality of our realtime quote systems. It is not difficult to create a chart or project the price within Microsoft Excel; the difficulty is in exporting the raw
'W
•
'A ' •
'M'
• V • ' 'j
isjB Data Window; FXUSJYMUL-Daily rime 110W97 ll/2ira7 11«4ra7 11/25/97 11/26/97
5:58cm 5:58pm 5:58pm 5:58pm 5:58pm
0 12700 12585 12656 12B87 12740
H 12729 12661 12728 12805 12791
L 12580 12489 12637 12635 12674
| C 12585 12636 12685 12725 12713
V 0 0 0 0 0
01 0 0 0 0 0
Plot1 62.73 64J7 66,66 68,16 67,19
Plol2 68,11 67,88 67.70 67,06 67.01
|
|
Plots 58.86 59.11 59.44 59.82 60.14
Figure 7.1 Aerodynamic Investments Inc., ® 1996-2011, www.aeroinvest.com Source: TradeStation. © TradeStation Technologies
172
• CALCULATING MARKET PRICE OBJECTIVES
Prfce Projections by Reverse-Engineering Indicators
*
173
N o spaces or fancy characters—just use letters. Then you must type
the RSI d o w n to p o i n t 2 i n this chart.
if needed, and then name the file w i t h no more t h a n eight letters.
solve for the u n k n o w n variable y, the price that is required to push
may want to use the same computer. Type the drive letter, directory
we k n o w the level the oscillator w i l l most l i k e l y target, we need to
because I use a different computer for m y offline analysis work. You
port f r o m current highs i n anticipation of a significant decline. As
empty box. I a m transferring the file to a floppy disc i n drive A ;
assumption that the RSI w i l l r e t u r n to this i m p o r t a n t level of sup-
T h i s is a very i m p o r t a n t step. A file name has to be entered i n the
low near the 45 level at p o i n t I i n Figure 7 . 1 . W e w i l l make the
Figure 7.2 displays the w i n d o w that w i l l open after the last step.
for the D o l l a r versus the Yen, and the RSI produced an oscillator
the mouse, or select Send to File, and press Enter on the keyboard.
t r e n d i n g b u l l market. T h i s chart clearly shows an u p w a r d t r e n d
is the last o p t i o n o n the b o t t o m of the pop-up m e n u . D o u b l e - c l i c k
we k n o w the major support level is at 40 to 45 for the RSI i n a
that offered Display A l l Data w i l l also display Send to File . . . . T h i s
a previous chapter. F r o m the range rules we discussed i n Chapter 1,
it w i l l not overwrite a file w i t h the same name. TradeStation w i l l
dow. M o v e the mouse cursor to any position w i t h i n the header bar
beep at you that it has transferred the file to the disc, when i n fact
M o s t likely you w i l l not have all the data visible i n the data w i n -
the exact same name, i n this case yen.txt, TradeStation w i l l m e r r i l y
one W o r k P l a n , you have the correct data and indicators displayed.
to m e n t i o n : i f your directory or floppy disc already has a file w i t h
were last viewing. So make sure that i f you have m u l t i p l e charts i n
.txt, and you w i l l be stuck at this step for ever more. A n o t h e r quirk
W i n d o w . T h e data w i n d o w w i l l correspond to the active chart you
to designate a default extension, w h i c h it w i l l do i f you do not type
d o w n menus on the top of the TradeStation screen, and select Data
file because you added this extension. D o not allow the computer
w i n d o w by c l i c k i n g the left mouse b u t t o n o n V i e w f r o m the p u l l -
the extension .txt after your file name. Excel w i l l k n o w this is a text
Figure 7.1 shows a data w i n d o w below the chart. Display this
that has the title Data W i n d o w , and click the right mouse b u t t o n . A small m e n u w i l l appear, and you w i l l need to put a check mark beside Show A l l Data. Press Enter, and the data w i n d o w w i l l now c o n t a i n several new c o l u m n s that include your indicators. O n e severe l i m i t a t i o n to note: TradeStation w i l l transfer only six columns of indicator data. I n Figure 7.1 we have the RSI and t w o averages. T h r e e columns have been used. I f you have more t h a n one indicator, you w i l l likely have to do this procedure a couple of times as it is very easy to exceed the six-column l i m i t of TradeStation.
: a l a Wlntiow r x i
rime 11j;0
N o w move the mouse cursor back into the title bar data w i n d o w of Figure 7.1, and right-click the mouse. T h e small w i n d o w m e n u
ssapm 558pm 5:58pm 5:58pm 5:5apm 558pm 5:56pm 5:5Bpm 5:58pm 5 58pm 5:58pm 5:58pm 5:58pm 5:58pm 5:58pm 5 58pm 5 58pm
0 12700 12585 12656 12687 12740 12709 12691 12777 12862 12864 12876 12925 12980 13056 12975 12944 12974 13032
lYMUl Daily H L 12560 12661 12489 12637 12728 12635 12805 12574 12791 12! 12; 12! File 12! 12! 12! 13C 13C 13! 12860 12967 13032
0 12595 12636 12685 12725 12713
V 0
01 0 0 0 0 0
12960 13030 13090
0 0 0
Ploti 62.73 64.77 66.66 68.16 67.19
Plot2 68.11 67 88 67 70 67.06 67 01
Help
Cancel
um—Tim—rim—n 13021 13059 13156
68.29 71.28 73.55
0 0 0
BTTO—rrf
n
72 72 72 64 72 78
Plots 56.88 59.11 59.44 59.82 60.14 60,37 60.73 61.22 61.70 62.18 62.73 63.38 64,03 64.33 64.46 64 63 64.92 65.29
Figure 7.2
174
•
CALCULATING MARKET PRICE OBJECTIVES
Price Projections
by Reverse-Engineering
Indicators
•
175
nizable format for the Excel program. Step 1 of 3 is displayed in
still the one in memory.
through the three steps required to import this file into a recog-
go through all the steps that follow and discover that the old file is
Excel will display the Text Import Wizard that will walk you
not warn you that the file was not overwritten and revised. You will
Once you have decided on a file name, press Enter or click OK
Figure 7.4. Ensure that this screen has the following selected:
File Origin will be Windows (ANSI). These are my defaults but
Now we need to import the saved text file into Microsoft Excel.
Original Data Type is Delimited, Starting Import row is 1, the
as illustrated in Figure 7.2. That was the last step from TradeStation.
From START>PROGRAMS > E X C E L , open the Excel spreadsheet program so that a new screen is displayed. Select from the
not necessarily yours, so check them. You are ready to click the next button.
button, and move on to the last step.
displayed in the window. Double-click on this file name to open it.
your default as I never seem to need to change it. Click the Next
move the file over to a different computer. The text file yen.txt will be
Tab box. The Text Qualifier should be the quote symbol. It may be
file was saved. In this example the file is on the floppy disc I used to
and be sure to uncheck any others. I always have to uncheck the
left to All Files (*.*). Then select the directory or drive where the data
options to select in Figure 7.5 will be Delimiters—select Comma,
appear. You will need to change the default file type on the bottom
The second step is displayed by the Text Import Wizard. The
top menus F I L E > OPEN. The screen displayed in Figure 7.3 will
X Mic.rnsioft Vxr.pA y
Eile
Edit
View
Insert
Farmat
lools
• l a f l y l alRilS^I * l * l « 5 l < ? l
Data
Window
Help
i.,1sl#|4e|[ioo5
xfA
^ Qlwl
X Microsoft t x c e l File
Edit
I^
Floppy
ifci
View
Buukl Insert
Format
lools
• I g l H l a l i a J y - l j j ^ l ^ i a i ^ l ^l^l
Data
Vjlindow
Help
LSlgUlFSos
SIAI
m oi>^
Text Import Wi?arf
IE
|Alid Lookfl:
T h e T M I W i z a i d h a s detetmined that j w u t d a t a is D e i m i t e d
A1
If this is correct, c h o o s e Next, or c h o o s e the D a t a T y p e that best d e s c r i b e s jiour data. pDriginal D a t a T y p e C h o o s e the file type that best describes your data:
© OSiStaJij O
Fixed W i d t h
• C h a t a c l e r s s u c h a s c o m m a s or tabs s e p a r a t e e a c h field ( E n c e i
4.0 standard).
- Fields a r e aligned in columns with s p a c e s b e t w e e n e a c h field.
Start Import at
flow:
|i
File Origin:
g
|Windows (ANSI)
Q
P r e v i e w of fJe A : y e n . Ixt.
•Date'.'Time','0','H','L','C','V','01','Plotl','Plot2 08/02/96,1758,10709,10735,10670,10681,0,0,0.00,0.00,0.0 08/05/96,1758,10684,10704,10644,10657,0,0,0.00,0.00,0.0 03/06/96.1758,10657,10730,10643,10676,0,0,0.00.0.00,0.0 08/07/96,1758,10676,10805,10670,10793,0,0,0.00,0.00,0.0 0B/Q8/96 1758.1Q797 10838.10773.10828,0,0 Q,QO.0.OQ.Q.OlLl
3lile(s)found.
Cancel
| |
< Back
|||
Next)
||
finish
|
F i g u r e 7.4
F i g u r e 7.3
176
^
• CALCULATING MARKET PRICE OBJECTIVES
Ble
Edit
View
Insert
Format
Tools
Data
Window
Help
file
Edit
View
Price Projections by Reverse-Engineering
insert
Fflrmat
lools
Data
Window
Indicators • 1 7 7
Uelp
Dii^iHl alfiily-l I ext Import Wizord
JAhal
f o x l Import Wizard
JArial
Stop 3 o1
This screen lets you select each column and set the Data Formal.
This screen lets you set the ddimitefs your dala contains. You can see hosM your text Is affected in the p r e v i e w below.
rOelimite
•
•General' converts numeric values to numbers, date values to dates, and all rerrtaNng values to text
O Tteat consecutive delimiters as one
Space
other
•
Semicolon
•
B Comma
pColumn Data FotmatO General O lent
O Do NoHmpoit Column (Skip)
Textaualifier
r~
rData Review-
1758 1758 1758 1758 Il758
08/02/96 06/05/96 08/06/96 08/07/96 nfl/nfl/96
[TiMie
Date
I1067Q 0644 110643 10670
Cancel Cancel
| |
l o t l ?la
106S1 10657 10676 0670 10793 |0 n773 l i n a ? a In
10681 D 10657 0 10676 0 10793 D
Iin773 linfl?fl In
1
<>
|||
Nex>>
||
Enish
| |
.00 0.00 .00 .00 on
3.0 0.0 0.0 0.0
In
||
Neiit>
||
Fmh
1
|
Figure 7.6 Figure 7.5 Microsoft
File
I n the Data Preview group, cHck o n the c o l u m n o f data that has the dates. T h e c o l u m n should be highlighted and may already
FXCRI
Edit
L31 HDate a a U B •a U
View
Insert
Fjarmat
lools
Qata
Winidow
U FTiI B J s j s j
JArial
be h i g h l i g h t e d w h e n this screen first appears. T h e n we move to the C o l u m n Data Format options o n the top right. C l i c k o n Date (General is the default that needs to be changed i n Figure 7.6). T h e last step is to select M D Y i n the list for A m e r i c a n users or D M Y for the rest o f the world. N o w just click Finish. Voila! It has been done. Your data and indicators have just been transferred into Excel and should look similar to Figure 7.7.1 w o u l d suggest saving the file i m m e d i a t e l y as an Excel file rather t h a n as text. I w i l l leave you at that step, k n o w i n g that you have data f r o m w h i c h to learn h o w to create charts. I t is u p to you to learn
Time
12/25/96
Ua
12/18/96 12/19/96 12/20/96 12/23/96 12/24/96
m Ul IH Ul l U
12/16/96 12/17/96
Bl D
0 1758 1758 1758 1758 1758 1758 175B 1758 1768 1758 1758 1758 1758 1758 1758 1758 1758
12/6/96 12/9B6 12/10/96 12/11/96 12/12/96 12/13/96
12/26/96 12/27/96 12/30/96
m • U Uji
H 11227 11299 11344 11338 11282 11312 11397 11382 11386 11360 11407 11423 11415 11445 11467 11492 11557
L 11177 11286 11301 11262 11245 11274 11367 11342 11327 11336 11403 11387 11407 11440 11447 11472 11555
Help
« J %J.
C
11301 11354 11373 11338 11364 11402 11433 11420 11395 11420 11441 11435 11457 11473 11510 11617 11626
11290 11343 11338 11281 11313 11390 11385 11384 11362 11408 11430 11410 11457 11468 11494 1153Q 11622
Ploti Plot2 Plots 52.15 57.23 56.92 55.62 57.48 56 86 55.21 57.05 66.79 50.66 56.74 56.52 63 54.61 56.37 58.15 53.56 66.45 57.71 53.79 56.6 57.61 54.19 56.55 56.47 55.07 56.5 58.92 55.82 56.61 60.49 56.36 56.78 58.3 56.7 56 84 61.8 57.94 57.06 62.59 59 57.3 64.45 59.71 57 61 66.93 60.73 58.02 72.24 62.36 68.63
h o w to use Excel. T h e c h a r t i n g is relatively easy. Just k n o w that some o f the reverse-engineering projections require you to k n o w
Figure 7.7
178
• CALCULATING MARKET PRICE OBJECTIVES
the indicator f o r m u l a and to have Excel generate the results, rather t h a n just p l u g g i n g i n the indicator values f r o m TradeStation.
Pr/ce Projections by Reverse-Engineering
Indicators
• 179
rXUS.IYMUl naily 0 7 / 0 6 / 9 8 C = 1 4 0 2 4 <-57
Figure 7.8 displays the data and the RSI i n an Excel chart. W e knew the target for the RSI was point 2; it was the same indicator level that developed at point 1 where a G a n n and Fibonacci confluence target zone overlapped. N o w we can solve for y and determine the price that w i l l move the oscillator to point 2. T h e projected price for the Yen/$ is at 133.72. Excel can draw these new levels on the chart i f desired. T h e last bar i n this chart is the hypothetical projection. Figure 7.9 returns us to TradeStation and the actual market move that developed o n June 16, 1998. T h e price low that t h e market recorded was 133.67, based on Spot currency market. T h e difference between t h e estimate and the actual market price l o w c o u l d n o t have been m u c h closer. T h e oscillator is just slightly higher than our
•
'M
• • 'J
'A
'J
'S
'0
• • 'N
b •
'96
••V '
V • 'V
' 'M
•
• ' j ' ' 'J
Figure 7.9 Aerodynamic Investments Inc., © 1996-2011, Daily Market Report, www.aeroinvest.com Source: TradeStation. © TradeStation Technologies
original assumption and has not f o r m e d a perfect double b o t t o m . T h e reason is that the market required more t h a n one day to realize the objective. W h i l e i t took four days for the market to realize the YEN/$-Dally (solution for unknown price 'y'l
price objective thereby having an impact o n t h e oscillator, i t d i d not m o d i f y the price objective. T h i s is not a m e t h o d you w i l l want to use for short intraday market moves. I t is a m e t h o d to consider when a significant market extreme needs to be identified i n case the worst-case scenario develops. T o determine a corresponding price, you can pick any indicator level you want. I n fact, you can use any other indicator to reverse-engineer a price objective. One o f the most valued indicators I use to reverse-engineer an indicator to project a market move is displayed i n Figure 7.10. I n the chapter that addressed trend lines, it was recommended that a horizontal line be plotted to mark the historic highs and lows for an
Figure 78 Aerodynamic Investments Inc., © 1996-2011, Daily Market Report, www.aerolnvest.com
indicator. T h i s helps to keep i n perspective the idea that an oscillator extreme displayed i n the chart window is i n fact a lifetime extreme or
180
• CALCULATING MARKET PRICE OBJECTIVES
f XUS.IYMUl Ddlly
VK.N
Price Projections by Reverse-Engineering
Indicators
• 181
fancy by p l o t t i n g the same differential as a histogram and a line chart. T h e zero line is where the closing price crosses this simple m o v i n g average. N o t so high-tech after all, b u t it works like a c h a r m i n extreme market situations. T h e reason why the oscillator extreme was used to project a price extreme i n June 1998 was that the weekly Yen/$ chart displayed i n Figure 711 was recording a m a x i m u m displacement equal to the one seen i n 1995. W h e n a market records a displacement equal to the extreme unmatched over a three-year period, that is of great interest. T h e indicator i n Figure 7.11 is the exact one used for Figure 7.10: the differential between market closing prices and a simple short-period
Figure 7.10
m o v i n g average. T h e difference is that the displacement was just plotted as a line this t i m e and the histogram and zero axis were not
you are i n reality just detrending the m o v i n g average, or i n this case
of time. I w i l l view a year of daily data and pick the m a x i m u m dis-
added. W h e n you plot a m o v i n g average of prices as an oscillator,
a lesser displacement that remains unchallenged over a long period
placement extremes to establish a worst-case market scenario. I n Figure 7.10 the extreme oscillator low at p o i n t X was clearly the m a x i m u m displacement i n this daily chart of the Yen/$ prior to June 16, 1998. T h e weekly chart warned that it was t i m e to make a projection by solving for the u n k n o w n price variable y. T h e chart shows the actual displacement i n this indicator w h e n the market intervention on June 16, 1998, occurred. T h e oscillator at p o i n t Z was displaced an equal distance to that w h i c h occurred at p o i n t X. T h e price projection f r o m this indicator w h i c h used an entirely different oscillator was 133.40. T h e chart itself shows the actual market move at p o i n t y that f o r m e d a market b o t t o m at 133.67 low. W h a t is this magical indicator displayed i n Figure 7.10? T h e oscillator i n Figure 7.10 is just the difference between the closing prices and a simple short-period m o v i n g average o n price. I made it look really
Figure 7.11 Aerodynamic Investments Inc., © 1996-2011, www.aeroinvest.com Source: TradeStation. © TradeStation Technologies
182
• CALCULATING MARKET PRICE OBJECTIVES
the differential between closing prices and the average. A detrended indicator is frequently easier to read when extremes occur.
Price Projections by Reverse-Engineering
Indicators * 1 8 3
T h i s m e t h o d may be tested by a market only once or twice a year. However, I tend to calculate m y p r i m a r y trading markets for
well w o r t h the effort.
intraday move equal to y -I- 1 day, you had better not be greedy and
market's unexpected move could be heading. T h e preparation is
market move i n the opposite direction. I f the market produces an
are madly scrambling around you t r y i n g to figure out where the
w i l l be the intraday pivot p o i n t that w i l l likely precede a forceful
f o r t i n g to sit c a l m l y w i t h a price objective i n h a n d w h e n others
emotional environment. T h e level you calculate for price y -I- 1 day
others become engulfed i n total panic and chaos. It is really c o m -
day as the market attempts to f o l l o w t h r o u g h on the prior day's
able target. It is a price objective that offers a secure milestone when
the indicator is capable of m a k i n g a double b o t t o m the very next
and I then k n o w the worst-case price projection is the more prob-
an intraday projection for the next extreme. I n other words, assume
table I have defined for a single days session is quickly exceeded,
happens next? First, use the extreme that just occurred to define
the market w i t h unexpected news. Usually the support or resistance
historical displacement, as occurred i n Figures 7.10 and 7.11, what
are possible or present. It is used on the days that a shock wave hits
engineering an indicator, and the indicator challenges the former
worst-case projections on a regular basis w h e n extreme conditions
I f the market realizes a price objective derived f r o m reverse-
objective realized.
rate the new extreme displacement into the indicator f o r m u l a and
that means step to the sidelines whenever this technique has a price
m o v i n g average and needs about three new price bars to incorpo-
priate action to u n w i n d immediately. T h a t is a significant signal
another extreme. T h e reason is that the indicator includes a simple
Cash has realized an extreme w i t h o u t Futures and taken the appro-
this m e t h o d , a m i n i m u m of three days to pass before calculating
ally it is the Futures market, but I have been grateful to k n o w that
h i g h . M y experience has been that you then need to allow, using
advance, only one index w i l l realize the extreme projection. Usu-
an intraday basis. T h e indicator w i l l now be o f f the extreme l o w or
Cash S & P and Future price extremes. I n a m e l t d o w n or extreme
show you a double b o t t o m i n the indicator. It w i l l develop only on
i n extreme environments, S & P traders should calculate b o t h the
the close for the market is m u c h higher, the oscillator w i l l never
As a final note for this particular application of price projection
expect more f r o m the market. As a major reversal often follows and
i n effect to reset the price projection capability of the indicator. Recall f r o m the m o v i n g average discussion i n Chapter 2 that some indicator formulas w i l l shift to a permanent position. K n o w i n g an indicator shift c o u l d develop w h e n new data enters the screen w i l l explain why we need to refrain f r o m using any indicator u t i l i z i n g m o v i n g averages for price projection purposes u n t i l after it has shifted into its permanent chart position.
chapter 8 PRICE OBJECTIVES DERIVED F R O M POSITIVE A N D NEGATIVE R E V E R S A L S IN T H E RSI
O
f all the methods in the first edition, this method of recognizing positive and negative reversal signals in RSI was the most often
incorrectly applied technique by early readers. Part of the problem is the change in market volatility. But the primary reason for the misunderstanding is not recognizing when a method is an analytic method versus a trading technique. I had written that this is one of the methods I use along with Fibonacci, Elliott and Gann analysis, but I never gave you a detailed example to show when a method changes its weighting and becomes far less favored when it becomes part of the decision tree for trading. Therefore, I now have the opportunity to clarify the use of reversal signals in general and to develop a new discussion about the differences between tools that help us analyze markets versus those better suited to trade them.
185
186
• CALCULATING M A R K E T PRICE OBJECTIVES
The original discussion defining the signals themselves remains
Price Objectives Derived from Positive and Negative Reversals in the RSI • 1 8 7
the indicator does not. 'Qiilinust be^^^
now within your
ling reversal signals relative to price.
bearish divergences present in this weekly D M K / $ chart between
divergence relationships between oscillators and price before tack-
Figure 8.1 uses bold lines to mark the conventions of bullish and
on opposite sides of a momentum channel. Therefore, focus first on
EURUSD will uncouple and cease to exist in its present form.
time. Reversal signals and divergence signals develop in oscillators
the Deutsche Mark is valid because there is a very real risk the
to recognize divergences, consider skipping this chapter until a later
long term to very short intraday intervals. Even the example using
technical skills before moving on. If you are still training your eye
valid. The signals form in all markets and all time horizons from
reversal, the RSI will develop a new oscillator low relative to a for-
will occur when the market makes a new price low on close and
reversals. There are positive and negative reversals. In a positive
the closing price high is recorded. Conversely, bullish divergence
offer rough estimate price projecrions. The new signals arc called
the indicator. In this chart the RSI is clearly at a lower peak when
in the same chart and displays examples of the new RSI signals that
market forms a new^price high on close that is riot confirmedj^y
Figure 8.1 conveniently has both bullish and bearish divergences
the price data and indicator. Bearish divergence occurs when the
'
mer oscillator low when the price data is at a higher pivot low. Look PXUSDMMUL-Weekly
at the positive reversal in Figure 8.1. It makes perfect sense that this indicator pattern would be bullish as the oscillator is becoming oversold when the market is trading at a higher level. This tells us that the market is building underlying strength when this indicator formation is present. Conversely, a negative reversal develops when the RSI is able to develop a new momentum peak relative to a prior oscillator peak, but the closing price associated with the new oscillator peak is at a lower price level. As the oscillator is becoming overbought when prices are at a lower level, it warns that the market is beginning to erode and become weaker. In Figure 8.1 a positive reversal pattern is labeled that forms from fairly blatant oscillator lows. Reversal signals do not have to
'92
'93
'94
'95'
he at oscillator peaks and troughs that form extremes. The negative reversal in Figure 8.1 shows that the first RSI peak is not at an
Figure 8.1 Aerodynamic Investments Inc., © 1996-2011, www.aerolnvest.com Source: TradeStation. © TradeStation Technologies
obvious extreme pivot. However, the second RSI pivot high on the right side of this negative reversal signal is clearly associated with a
be used for trading purposes and should never be used to calculate
over on the oscillator.)
the oscillator pivot. Any method that only sees closing prices cannot
subde and closer to the zero line or under two averages crossing
the high or low, when we want to know the associated price for
about most oscillators. The strongest signals are often those more
Keep i n mind that the closing price for a price bar is used, not
at blatantly obvious pivot extremes. (This observation can be said
highest level for the positive.reversal price projection calculation.
they can be much stronger market signals than reversals that form
with price. We will always look for the momentum peak that is the
the larger trend for the oscillator, it is called a hidden signal, and
will not be the closing price high as oscillators form divergences
method is discussed. When an RSI reversal signal occurs within
located between points W and X. Frequendy the momentum high
lower closing price. This will be clearer when the price projection
Price Objectives Derived from Positive and Negative Reversals in the RSI • 1 8 9
188
•
CALCULATING M A R K E T PRICE OBJECTIVES
The positive reversal in Figure 8.2 is the same signal introduced
or assess stop placement. That is why this is only an analysis method.
X. As describedfor a positive reversal, the momentum extreme at
oscillator high marked Y that is the highest momentum extreme
Y, where Y is the extreme momentum low between points W and
oscillator lows I have arbitrarily called pivots W and X, then an
The negative reversal has three oscillator pivots labeled W, X, and
to you in Figure 8.1. It now has three oscillator pivots labeled: two
point Y in the RSI may not be the actual price low because diverFXUSDMMUL-Weekly
DMIvS
gence may be present. ^The oscillator will always determine the CQirect
closingjjrice Jo use^
To make a price projection from a positive reversal signal, use the following formula: (X - W ) + Y = the new price target In Figure 8.3 the closing price associated with the RSI low at point W is 1.5835. The closing price associated with the RSI low at point X is 1.5957. In a positive reversal the closing price associated with pointX must always be greater than or equal to the closing price associated with the RSI pivot at point W. It is this positive differential '92
'93
'94
'95'
that defines a positive reversal signal in the indicator. Therefore, the price projection from near 1.5957 is (1.5957 - 1.5835) = 0.0122 + Y
Figure 8.2 Aerodynamic Investments Inc., ® 1996-2011, www.aeroinvest.com Source: TradeStation. © TradeStation Technologies
The differential is then added to the closing price associated with the momentum extreme at point Y. The closing price that
190
• CALCULATING MARKET PRICE OBJECTIVES
FXUSDMMUL-Weekly
Price Objectives Derived from Positive and Negative Reversals in the RSI • 1 9 1
FXUSDMMUL-Weekly
DMIv$
17000 -16000 -15000 -14000
highest price bar on close = 1.7540
-70.00 -60.00 -50.00
Positive Reversal
Price target from 1.5957 at X is 1.7537
-40.00 -30.00 '92
'93
'94
Figure 8.4 Aerodynamic Investments Inc., © 1996-2011, www.aeroinvest.com Source: TradeStation. © TradeStation Technologies
Figure 8.3 Aerodynamic Investments Inc., © 1996-2011, www.aeroinvest.com Source: TradeStation. © TradeStation Technologies corresponds to the m o m e n t u m h i g h at Y is 1.7415. Therefore, the final target is the differential 0.0122 + 1.7415 = 1.7537. T h e actual price at close o n February 11, 1994 was 1.7540, and it marked the end o f the dollar rally at that time. T h e formula to use for calculating an estimated market decline f r o m a negative reversal is as follows:
and the result must be subtracted f r o m t h e closing price associated w i t h t h e R S I oscillator extreme at p o i n t Y . T h e signal i n Figure 8.4 is actually a negative reversal that follows a series o f price p r o j e c t i o n signals w i t h i n this decline. W e w i l l look at each of these negative reversal signals and the price targets that develop i n Figure 8.5. There are four negative reversals labeled i n Figure 8.5. (There are actually eight signals present. C a n you locate the other four?) T h e
Y - ( W - X ) = the new price target
three pivots that we use to make the price projection w i l l name the
h X Y E X Y is i n fact the same signal marked i n Figure 8.4 as W X Y .
W . Therefore, the points that create the differential are switched,
are signals abc, def, g h i , and bXY. There are also signals at e X Y and
always be less t h a n or equal to the associated closing price at p o i n t
signal being described. T h e negative reversals labeled i n this chart
I n a negative reversal t h e second p i v o t h i g h at p o i n t X w i l l
192 • CALCULATING MARKET PRICE OB)KCTIVES
FXLISDMMUL-Weekly Target from b = c- (a-b) = 1.6658 (1.7185-1.7137) = 1.6610 Target from e = f-(d-e) = 1.6433-(1.6715-1.6706) = 1.6424
^1
• 02/11/94
'V
Price Objectives Derived from Positive and Negative Reversals in the RSI • 193
define a trend using a 14-period RSI discussed in the first chapter, a 14 period should always be used for RSI in all time horizons. Sometimes Forex traders ask, 'Why are my spot prices different from yours?' Some traders who have been in the business for years are unfamiliar with how their Spot market data is created. So I had better digress for a moment and explain why differences can occur.
Target from h= i-(g-h) = 1.5396-(1.5980-1.5741) = 1.5157
04/21/95 •
The major quote vendors—Reuters, Bloomberg, TradeStation, and CQG, to name just a few—must create Forex spot market quotes from contributing banks and major financial institutions
Target from X= Y- (b-X) = 1.5000-(1.7137-1.5813) = 1.3676
which report their trades to the vendor. This information is gathered in several ways. Some contributors have trades captured and
Sometimes it is the bid and ask that is reported by a contributor
92
agreements. Other data vendors collect trades in secondary ways.
Actual close for the lowest price bar (week of April 21, 1995) was 1.3678
electronically transmitted directly to the vendor with exclusive
Figure 8.5 Aerodynamic Investments Inc., ® 1996-2011, www.aeroinvest.com Source: TradeStation. © TradeStation Technologies
and not an actual market level that was traded. Regardless, the more contributors a vendor has reporting trades, the more accurate will be their market indication. Reuters has the largest number of contributors reporting Forex market levels at this time. Bloomberg
data feeds and that solves the problem. But this is just a heads-up
ended the decline was 1.3678. Because of the range rules that
do not meet your needs. Some of these products allow third-party
tioned in the beginning, the actual close for the price bar that
the same quote system may find significant differences at times that
1.5000 - (1.7137 - 1.5813) offers an objective at 1.3676. As men-
is primarily a retail stock specialist, the data for a Forex trader using
tive from X is calculated using the formula Y - (b - X ) . Therefore,
ranges from a Forex trader using Reuters. If you have a vendor that
DMK/$. Finally, the last signal is bXY, where the price objec-
TradeStation data, as an example, you may have different price
are 1.5396 - (1.5980 - 1.5741), giving a 1.5157 objective for
value indications. What this means is that if you are working with
is from pivot h. The formula to use is i - (g - h). The results
teric cross rates and secondary markets can be given fair market
1.6453 - (1.6715 - 1.6706), which equals 1.6424. The third target
tors as possible for their exclusive databases so that the more eso-
1.6610. The target from the RSI pivot e will equal f - (d - e), or
Software vendors are constantly trying to collect as many contribu-
the new price objective. Therefore, 1.6658 - (1.7185 - 1.7137) =
was second when last I checked before the 2008 bank implosion.
The target from the RSI pivot b is calculated c - (a - b) to obtain
194
• CALCULATING MARKET PRICE OBJECTIVES
Price Objectives Derived from Positive and Negative Reversals in the RSI • 1 9 5
back and f o r t h w i t h w i l d swings the volatility is h i g h . Vola-
i n both i f you trade futures only.
day i n a straight line have low volatility. W h e n markets chop
to the spot market. Targets and technical analysis should be done
and h i g h volatility? Keep i n m i n d markets m o v i n g day after
ous software platforms w i l l exist. Futures traders must have access
3. Does the signal or m e t h o d work equally well i n times of low
that problems w i t h l i m i t e d and different data sources into the vari-
calculation.
following:
cannot see price ranges because RSI uses closing prices i n its
raise about any m e t h o d or f o r m u l a that was new to me are the
so you k n o w where you are w r o n g i n a t i m e l y manner? RSI
nals can be used i n a real-time e n v i r o n m e n t . Questions I w o u l d
or see-saw back and f o r t h , how accurate is the price objective
B u t there should be a lot of questions i n your m i n d before the sig-
nals do not p e r f o r m consistently w e l l . W h e n markets chop
ple that was chosen to i n t r o d u c e y o u to reversal signals i n RSI.
t i l i t y does not i m p l y trend. Here is where RSI reversal sig-
T h e weekly D M K / $ chart i n Figure 8.5 offered a perfect exam-
T h e calculation developed by J. Welles W i l d e r for RSI is 1. Does the indicator work equally well for any market i n any
stated as follows:
time horizon? T h e m e t h o d and techniques you favor should work i n any market. However, you may f i n d one m e t h o d could be favored over another w h e n markets accelerate versus stall i n consolidation ranges. For example, a simple detrended movi n g average plotted as an oscillator is useful i n strong trending environments. 2. Is the indicator as useful i n a rising market as it is i n a d e c l i n i n g market? H o w w o u l d you check this? Use the cross-hairs o n your pointer and t h e n scan f r o m right to left across the entire history of data available. L o o k at the amplitude of the
RSI = 100 - [100/1 + RS] where RS = Average of n period closes up / Average of n period closes d o w n . T h e f o r m u l a begins w i t h calculations for average gain and average loss over the past 14 averaged periods. T h e average gain and loss is determined by using closing prices only. • First average gain = sum of gains over the past 14 periods/14 • First average loss = sum of losses over the past 14 periods/14
point swing before the market close.
a bigger trend change develops?
moved lower. It never knows i f the DJIA experienced a
needed to fit the t i m i n g where a fifth wave move ends before
mean all 14 days w i t h i n the f o r m u l a had closing prices that
w i t h the ends of t h i r d waves i n price and t h e n divergence is
average gain equals zero. I n a 14-period RSI a zero value would
ally turns? A t m o m e n t u m extremes do you find these a l i g n
trading applications. I n Wilder's f o r m u l a w h e n RSI is 0 the
and t h e n require a W-shaped b o t t o m before the market actu-
fore^ better suited to analysis work t h a n risk management i n
the signal. Does the oscillator f o r m a simple inverted A top
Reversal signals only 'see' the closing prices and are there-
signals as a displacement f r o m zero. L o o k at the pattern of
1,000
196
• C A L C U L A T I N G M A R K E T PRICE OBJECTIVES
Cjt^ow
m u c h t i m e after any signal does the market take to react?
Price Objectives Derived from Positive and Negative Reversals in the RSI • 1 9 7
indicator fail? W h a t is the probability it w i l l fail? A signal that consistently wrong.
evaluate a technique. Sitting on a position going nowhere is
fails w i t h h i g h probability can also be valuable because it is
Consider opportunity costs as a risk exposure variable when y o u
costly i f other markets are m o v i n g i n fast trends that you would have traded had your capital not been tied to a ship anchor. 5. Is this a signal that has merit as a trading signal? T h i s is the question I d i d n o t list i n the first e d i t i o n . A signal can keep u on the right side of a market trend, b u t it may not improve r market t i m i n g to enter and exit positions. T r a d i n g signals l i m i t your capital d r a w d o w n exposure and allow y o u to react q u i c k l y and see f o l l o w - t h r o u g h i n the market shortly after. T h e N Y S E Advance-Decline l i n e is not a trading tool
Use this indicator pattern w i t h i n the RSI to explore some of these questions. W e k n o w the weekly D M K / $ chart offered a perfect scenario f r o m w h i c h to learn h o w to identify f u t u r e market objectives f r o m the RSI reversal signals. But do these signals work equally well i n other markets? Figure 8.6 is a weekly chart of the S & P 500 futures market. T h i s example uses a weekly t i m e horizon to keep the t i m e interval the same, only changing the market for comparison. T h e A p r i l 1997 low allowed the RSI to f o r m a low that was below
as i t can stay at extremes for very l o n g periods of t i m e w i t h no reaction i n price. I t has no ability to t i m e the market. For
iP-Weekly
K S I P O S I T I V E K E ' E K . S A L Price ProiocHon
outright buy/sell position traders the V I X is n o t a trading tool as the extremes give you no help o n correct market entry or exit t i m i n g . O p t i o n traders w i l l speak o u t l o u d l y against this view, b u t they trade i m p l i e d volatility and cannot enter stops based o n market levels. T h e V I X is of little value to a futures trader w h o simply wants to buy and sell o u t r i g h t positions. positive and Negative RSI reversal patterns ca'nnot
'see'
the full, price range of a bar. Therefore using the price targets / derived f r o m these signals as a place to hide stops w o u l d be
patterns themselves give guidance w i t h m e r i t i n trends and
^
an inappropriate use of the technique. However, the reversal
1 /
transitions. Therefore they are of value i n strategy development and better left o f f the battlefield. 6. W h y does_an indicator w o r k a a d h o w do y o u increase the probability of its accuracy? Under what conditions w i l l the
Figure 8.6 Aerodynamic Investments Inc., ® 1996-2011, www.aeroinvest.com Source: TradeStation. © TradeStation Technologies
198 • CALCULATING MARKET PRICE OBJECTIVES
the oscillator extreme recorded in 1996. As the oscillator pivot at point X is produced from a higher price level, there is another positive reversal signal present. The target the signal offered at 911.35 was realized and then was significantly exceeded. It confirmed the trend, but if you exited the trend because of the target from this signal, how would you reenter the position? The market keeps going with no significant retracement. Strong trends agtually need indicators that are not normalized in a 4:1 time comparison. Moving on, we need to consider how this indicator performs when a monthly chart is evaluated or an intraday time interval is the focus. The monthly chart of the S&P 500 Index is displayed in Figure 8.7. The positive re<'ersal signal that formed in 1997 in the weekly chart was just one signal within an entire series of positive
Price Objectives Derived from Positive and Negative Reversals in the RSI • 199
reversals that had developed throughout the monthly chart. Signals abc, bde, fgh, and ijk have all had their price objectives realized by the market. The method is true to the trend indicated, but gives no indication as to how rough a ride it could be toward the target. The signals do not give a sense of how strong the trend will be that follows. That is true of any oscillator, but the positive reversal has a pivot low within the target calculation that fails to mark the exact market entry. Pivots b, d, g, and /' are not the final price lows before the larger trend resumes. Therefore this again shows that this method is better suited to analysis work that develops strategy rather than market entry/exit trading applications, because it has limited value for execution tactics. There are numerous smaller signals not marked in this chart, but as their price projections are inconsequential now, they are not labeled. The crash of 1987 produced an enormous positive
sap
500 Sluck InJux Monllily
10600/98 C-1100,63
-1000 00 -800.00 11'
^i«ri,i'l
40 level, which we know is the support range for the RSI within
are at a higher level on the RSI y axis than the pivots that form at points fand g. This is an interesting observation that would make
C -80.00
83
84
35
reversal signal in the RSI with point b forming a bottom above the the context of a bull market. Then we see that the three signals that
-400.00
•600.00
follow develop at higher levels. The RSI pivots that form at i and /'
if
,.'.'»„„„r»l'
RSI w « Zones 79.17
[J V 82
86
1o
87
91
92
93
94
95
one wonder at what point could these signals warn of an end to can be answered by comparing signals that form along the same
-60.00
the expansion cycle in the market and the uptrend. This question
•70.00
•50.00 -40.00
horizontal level. Does the market always respect the signals that develop in the same range where pivots i and ; are located? Can
98
97
96
you determine a probability from such a comparison? The answer is yes. Study how many times a signal develops on a specific sup-
Figure 8.7 Aerodynamic Investments Inc., ® 1996-2011, www.aeroinvest.com
port or resistance level within the indicator.
Source: TradeStation. © TradeStation Technologies
200
• CALCULATING MARKET PRICE OBJECTIVES
Be aware that the amplitude, or distance between the momen-
Price Objectives Derived from Positive and Negative Reversals in the RSI • 201
5P~Weekly
tum highs from the RSI pivot lows, are becoming narrower. The rise from point
KSI P O S I T I V E KEV£KSAL Price Projection 1) The number of days or periods between W and X ?
to c is much wider than the rise from point d to e.
Each has a narrower rise than the previous signal Now we are
2) I s i t the differential between E S I level 1 and E S I level 2 ?
beginning to actually think about what makes an indicator work and to question what has to occur when it has its best performance.
3) E S I range the signal developed? Is a signal at an E S I pivot 40 more significant than a higher level?
Another observation is that the number of periods th^t separate points i and / are less than the spread between points f and g, which in turn is less than the spread between b and d or d^and b.
4) Is the amplitude of E S I significant?
Therefore the element of time has entered our evaluation of the signals. When signals develop quickly they have a higher plpb-
5) Maybe the spread between moving averages on E S I ?
ability of being correct. Our curiosity should prompt us to ask i f these same characteristics can be observed in different markets with uptrends? What happens in downtrends? Figure 8.8 is a segment of the weekly S&P market that developed a positive reversal signal. Segments of the signal are labeled
Figure 8.8 Aerodynamic Investments inc., © 1996-2011, www.aeroinvest.com Source: TradeStation. © TradeStation Technologies
14-period RSI with two averages; and the bottom oscillator is the
of my career.
oscillator in Chapter 12); the middle oscillator is the Wilder
how detailed my examination of RSI signals was in the early years
tum imbedded within a very fast RSI (more will follow on this
Chapter 14 to illustrate the concluding Zen story. They show you
oscillators under the data: my Composite Index which has momen-
segments of five hand-drawn RSI charts on the last five pages of
but starting to get a little over extended. There are three different
probability that the signal's targets w i l l be realized. You w i l l find
of wave iii up with the implication that the uptrend is incomplete
the less time the pivots need to form the signal, the higher the
shll displayed. The rally is believed to be working toward the end
W make a difference? The answer is yes because the tighter, or
2011. It is a real-time working chart with wave structure and targets
on the signal's accuracy? Does the spread or delta of X minus
Figure 8.9 is the monthly Gold chart captured on August 2,
so a statistician could study them. Does the amplitude impact
Do signals of this nature develop in other formulas? Yes they do. You can take the spread of two simple moving averages and find reversal signals develop in any detrended oscillator.
simplest one of all as it is a detrended average. A detrended average uses two moving averages on price. I used a one-period average that simply connects the closing prices, then
202
• CALCULATING MARKET PRICE OBJECTIVES
TradeStation Chart Analysis - @G.. [SjilJ Gold - Monthly L:1643.20 21.50 1.33% 6:1643.10 A-.U43. ..I 2,600.00
Price Objectives Derived from Positive and Negative Reversals in the RSI • 203
a fairly fast moving average like a B-period simple moving average. When the slower moving average is at the point where the two averages are the same value (the slower average is about to cross through the faster average), the spread between the two averages is 2,200.00
1966.39
1147.88
iii
1^1,(00.00
zero. When the two averages separate, the spread is positive when the faster average is above the slower. It is vice versa for downtrends, as the faster average will track below the slower moving average, yielding a negative spread between the two. Plot the spread of the
7^
~ll
OfiO
OQI
averages as a displacement from zero and you have created the simplest oscillator possible. In flash crash scenarios this is the best
<00.00
oscillator to use. However, all oscillator indicators are similar. They
Composite
all use the difference between two. When plotted as a spread dif-
100.00
ferential above or below zero it is considered to be a detrended technique. However, not all will be normalized by forcing them to
20.00
track between zero and 100.
RSI with Avgs
In Figure 8.9 there is a strong uptrend in Gold in 2010. A line has been drawn under the three different oscillators labeled A, B, and C. In this strong uptrend the
only
oscillator that did not
develop positive reversal signals was the 14-period RSI along the
DeTrended OscI
line B under the RSI indicator. The detrended oscillator in C produces the reversal signals that are the most pronounced and easiest to read. I read the signals forming in the Composite Index and detrended oscillator as confirmation of the uptrend. However, I have no interest in calculating their respective price objectives
Figure 8.9 Aerodynamic Investments Inc. Dally Market Report. Source: TradeStation. © TradeStation Technologies, inc., 2001-2011. All rights reserved. No investment or trading advice, recommendation or opinion is being given or intended.
because I know I have other methods that are far more accurate that we have discussed in earlier chapters. We also have the new Gann chapter approaching in Chapter 9 that will reveal far more to you than was released in the first edition. Between Gann and
204 • CALCULATING MARKET PRICE OBJECTIVES
Price Objectives Derived from Positive and Negative Reversals in the RSI • 205
Fibonacci confluence targets this method does not come close for developing trading targets. However, the pattern formation remains of value and forces us to look deeper into our other methods when these formations develop. This chart in Figure 8.9 happens to contain a method using boxes to make geometric targets along with Fibonacci projections that are more accurate. The boxes are projections from internal wave structures which use full price ranges within earlier parts of the move. It is like a swing trader projecting equality relationships to earlier price swings, but this method utilizes the mid-point of the developing trend to project the future swings and subset corrections that will develop based on geometric relationships. This method is fully detailed in my book Fibonacci Analysis. RSI was not the only formula for developing positive reversal signals to reinforce and confirm the trend, but the RSI is the only %mula that tracks the ranges discussed in Chapter 1. The RSI lows along line B also hold the moving averages on the RSI indicator. This offers a place to add or reestablish a position that may have
Figure 8.10 Aerodynamic Investments Inc., © 1996-2011, Dally Market Report, www.aerolnvest.com Source: Charts by Market Analyst 6, © 1996-2011
been scaled back.
moved the oscillator to such an extreme displacement below zero
compare historic crashes.
lowed the rally into the weekly high of April 15, 1930, never again
moving averages. Oscillators in the style of line C can be used to
devastating slide into the 1932 lows. The declining swings that fol-
far away from zero as required in a strong trend to separate two
retracement in the Dow Jones Industrial Average that lead to the
formulas that are not normalized. This means they can travel as
from this momentum low was the historic 50 percent counter-trend
Oscillator lows along lines A and C are positive reversal signals in
to record the extreme placement in Figure 8.10. What followed
along line C just outperformed the RSI signals in this Gold chart.
in place near -75 from the 1929 Crash, a horizontal line is drawn
Recognize that the simple detrended oscillator with signals
This is an extremely important study. The week of November 10,
as occurred in the first swing down into the November 1929 low. market does not challenge the 1929 extreme displacement again
tive bench mark in future market extreme events. With the extreme
In Figure 8.11 the history of the weekly oscillator in the DJIA
1929 established an oscillator extreme that was used as a compara-
206 • C A L C U L A T I N G M A R K E T P R I C E O b j E C T i v E S
Price Objectives Derived from Positive and Negative Reversals in the RSI • 207
JETS-
Figure 8.12 Aerodynamic Investments Inc., © 1996-2011, Daily Market Report, www.aeroinvest.com Source: Charts by Market Analyst 6, © 1996-2011
MMmmMimmimmmmmm'imMmmmmTiL. Figure 8.11 Aerodynamic Investments Inc., © 1996-2011, Dally Market Report, www.aerolnvest.com Source: Charts by Market Analyst 6, © 1996-2011
follows. It is important to see the pattern that old extremes are tested after new historic lows are made. Figure 8.12 shows history challenges the o l d 1929 m o m e n t u m extreme i n the 1980's, b u t it is not u n t i l the October crash o f 1987 that
i n g the 1929 crash becomes the perfect support level for market
this m o m e n t u m l o w begins t h e multi-decade b u l l market that
After t h e 1987 crash, t h e o l d d i s p l a c e m e n t established dur-
displacement occurred the week o f September 10, 1974 and f r o m
prior to the 1987 crash and it defined support again after the crash.
ket reversal can be sustained. T h e challenge to the 1929 extreme
i n 1929. T h e 1929 extreme was i m p o r t a n t support for t h e D J I A
created i n 1929 must be tested again as support before a major mar-
m u c h o f the history o f this oscillator is the o l d extreme established
low is established. Take notice that the old extreme displacement
a new extreme m o m e n t u m low is established. T h e heavy line under
u n t i l the week o f November 28, 1973. A new historic m o m e n t u m
208
• CALCULATING MARKET PRICE OBJECTIVES
Price Objectives Derived from Positive and Negative Reversals in the RSI • 209
sal trends^ However, in Figure 8.13 the DJIA chart established a
historic extreme near -1511. But once again the rally did not
These olc[extremes become very important levels to begin rever-
again in August of 1998. September 11, 2001 established a new
form a recovery unhl a test was made near -478, the prior historic momentum extreme low.
extreme in October 1987 was established near -478. It was tested
entry throughout the balance of the 1980's. The new oscillator
new momentum extreme low in October 2008 at a level of-2296. As of November 2011 the old historic extreme at -1511 has not been retested. If history repeats; and we know it does, the bear market of 2008 is not over and the rally unfolding cannot build a long horizon bull market. Only time will tell if we are about to see a negative reversal develop that carries an historic message of what lies ahead.
Figure 8.13 Aerodynamic Investments Inc., © 1996-2011, Daily Market Report, www.aeroinvest.com Source: Charts by Market Analyst 6, © 1996-2011
chapter GANN A N A L Y S I S : C A L C U L A T I N G P R I C E AND TIME OBJECTIVES
Introduction
I
n Chapter 2 the following statement was written in the discussion about cycles. 'We easily miss the fact that a cycle is a rhythmic
fluctuation that repeats over time with reasonable regularity. It is not until the rhythm persists over a long span of time that we recognize this regularity cannot be the result of chance.' A lot of people have been waiting for me to write a Gann book. I promise it will be written and will fully disclose a careers lifetime of research and application, but a solid introduction can be offered in a chapter that can begin your journey and offer enough information for you to stay clear of
the many guesses that are in print at the moment. This chapter follows 15 years of study and I likely have at least five more years ahead just to complete all the original work of Gann's that I was given by Elinor Smith Sullivan, a former pilot who did contractual flights for •
211
212
• C A L C U L A T I N G M A R K E T P R I C E OBJECTIVES
Gann and many other business men of that era. (see Breakthroughs in Technical Analysis, edited by David Keller (Bloomberg Press)). Therefore, if you have the first edition of this book, rip the Gann chapter out; the place where I find myself now is light years away
Gann Analysis: Calculating
Price and Time Objectives
•
213
is the natural law of celestial motion, first recognized by Copernicus, where the Earth's revolution around the sun has a significant effect on the timing of when life springs back to life, blooms, matures, withers, and decays in a rest phase. These two cycles of celestial motion equinox as a timing mechanism within the markets.
The books in general about Gann are often misleading. What is a
we accept easily, but the third is more difficult, the precision of the
Many use the methods of Gann's son and they have been led astray.
beginner to do? Working directly from Gann's courses is the best way. The aim of this chapter is to offer an introduction to a few methods that will get you started and give you enough information to know when you are being duped. Gann was a master of cycle analysis. He used several different approaches with the aim of finding confluence between price, time, and diagonal analysis. One of Gann's approaches to identifying when a trader has permission to exit or enter a market is to use what are known as the 'laws of vibration' or 'natural laws.' These are
Expansion and Contraction Properties between Cycles Before we go any further, consider the following Wheat charts i n Figures 9.1 through 9.5. The agricultural markets will be extremely important i n the years ahead and even though you may only trade an equity index, treasury, or Forex markets, it is important to know that coming years will place greater emphasis on grains. History has
two very antiquated terms which refer to celestial motion. Gann's prestige grew in 1909 when Richard D . Wyckoff interviewed him for The Ticker and Investment Digest. Wyckoff further witnessed a
Annual Average Wheat in England (1259-1938). Compiled by Bureau of Crop Estimates
trading period with an impressive win:loss ratio. Keep in mind that the year of the interview was 1909. If you want to know what makes a trader, study the materials you think brought them into their era of success. I made the mistake of beginning with modern resources, when books written in that era would have been easier to use as the language was the same as that used by Gann. As an example, books of the late 1800s use the term natural law freely to describe the 'stellar vibrations' of our solar system. It is a natural law, which we now call 'diurnal motion,' where the Earth's rotation on its axis has an influence on us to move from a cycle of conscious alertness to a state of sleep and back again every 24 hours. It
Figure 9.1
214
• C A L C U L A T I N G M A R K E T PRICE
OBJECTIVES
Dally WHEAT 1900 to 2011
Gann Analysis: Calculating
Price and Time Objectives
•215
taught us that the cost and availabiUty of food is a more poHtically destabiUzing force t h a n a f i n a n c i a l crash. Figure 2.10 i n Chapter 2 showed the European W h e a t prices i n
1000-
the fixed cycle work o f E d w a r d Dewey. As part o f that discussion I mentioned the data was collected by L o r d Beveridge o f L o n d o n . For years the only chart I had seen was the Dewey chart that spans data f r o m 1500 to about 1850, yet L o r d Beveridge reportedly spent a l i f e t i m e collecting European market prices. H o w far back was he able to go i n Wheat? H e went back to 1259. T h e interval Edward Dewey charted for us makes it hard to see a truly remarkable c o m parison to more current W h e a t markets.
Figure 9.2 Daily Wheat 1900-2011 Source: Market Trader Gold, Alphee Lavoie and Sergey Tarasov
Figure 9.1 includes all the data L o r d Beveridge collected, w i t h a continuation into 1938 by the Bureau o f C r o p Estimates i n L o n d o n . I f you are a W h e a t trader and have never seen this historical data before, you m i g h t t h i n k it looks oddly familiar. Figure 9.2 is the daily W h e a t prices f r o m 1900 to 2011. I f you do not see the relationships and similarities easily between Figures 9.1 and 9.2, t u r n to Figure 9.3 where the historical data o f 1259 to 1938 i n E n g l a n d is an overlay on the historical market data o f W h e a t i n N o r t h A m e r i c a f r o m 1900 to 2011. I f I placed the data f r o m 1259 to 1938 any closer over the data f r o m 1900 to 2011 by adjusting the scale o f the y axis, you w o u l d not be able to see the upward price spikes i n the underlying set o f data at all. Cycles do indeed repeat, but they do not all mirror a fixed interval of time. Most charts that examine the m i r r o r i n g properties of cycles suggest t h e x axis should r e m a i n a constant to support fixed cycle theory. However, Figure 9.3 is a comparison that shows that cycles do repeat but the cycle may experience an expansion or contraction attribute. T h e relative ratios that define the significant
2 1 6 • CALCULATING MARKET PRICE OBJECTIVES
Gann Analysis: Calculating Price and Time Objectives • 2 1 7
versus the other suggests cycles have fractal properties. As you
relationships to one another.
charts have similar proportional spacing. The length of one cycle
logarithmic and have harmonic ratios that explain their internal
overlay of 1920 to current. The principle market spikes in the two
sion and contraction changes have a mathematical basis. They are
prices from 1259 to 1938 have cycle beats that align to the data
beats within the larger cycle remain similarly placed. These expan-
ics and harmonic theory is a precise science. There is a great Web
ars as Pythagoras and Plato.
monic proportion is about. Harmonics fit within the field of phys-
defined the curriculum of the Quadrivium, studied by such schol-
buzzword; their contents show the authors have no idea what har-
be unrelated. These four areas of study are four of the seven that
books have hit the market lately using the word 'harmonics' as a
of one another though they seem from the outside looking in to
spiral pattern from rectangles are harmonic, but not all. Several
These four fields of study will mathematically flow into and out
may have a Fibonacci ratio. Some ratios that create the Fibonacci
etry, harmonics (the theory of music), numerology, and astronomy.
properties. As an example, Elliott is not harmonic just because it
sis you must have a comprehensive working knowledge of geom-
to take away the understanding that some ratios have very special
the education path he followed. To fully comprehend Gann analy-
In the next discussion regarding harmonic ratios I only want you
alone. Gann was a thirty-second degree Freemason and that was
selves have a mathematical basis of unique harmonic cohesiveness.
cess. The answers are not easily extracted from his original charts
study hundreds of charts it becomes clear the proportions them-
To understand Gann, you need to study what brought Gann suc-
site you can explore at Georgia State University—they have an app
Harmonic Ratios
as well for iPods and iPads: hyperphysics.phy-astr.gsu.edu/HBASE/ hframe.html. (If the link changes, try a search using 'HyperPhysics
Gann student from the main menu.
see that cycles have expansion and contraction phases themselves
However, this site offers several useful categories of interest for a
this knowledge can be applied within our industry. The start is to
discussion on Pythagorean Temperament in the 'Sound' category.
branch, and branch yet again. But my task is to stay focused on how
of Physics and Astronomy. Be sure to work your way toward the
wisdom. There are endless rabbit holes to explore that branch and
to the top of a Google list.) This site is hosted by their Department
by Gann can spend decades in pursuit of his understanding and
C.R. Nave Georgia State University'; my test moves the site right
Readers new to Gann will discover that those of us sincerely intrigued
per second that is mathematically related to the other one.
ratios. Consider Figure 9.3 where the compressed years of Wheat
each string length when plucked produces a vibrating frequency
cycles expand and contract is contained within their proportional
lengths. Together they produce a pleasing sound to the ear because
contraction attributes. The mathematical glue that explains how
monic ratios originate from the vibrations of two proportional string
is to gain a mathematical understanding of these expansion and
While we do not concern ourselves directly with pitch, har-
that proportionally change cycles relative to one another. Our job
218
• C A L C U L A T I N G M A R K E T PRICE OBJECTIVES
• Lengths of 180 to 360, a ratio of 2:1, is called an octave. The tone from the shorter string, exactly half the length of the longer, will vibrate exactly two times faster than the longer string. • Lengths of 216 to 360, a ratio of 5:3, is called a major sixth. • Lengths of 240 to 360, a ratio of 3:2, is called a perfect fifth. The perfect fifth had great significance to the Pythagoreans. It has great significance for us as well. As an example, a Gann Wheel can calculate a perfect fifth price projection from a pivot point. Our Solar System contains perfect fifths. Have you ever heard of the phrase 'harmony of the spheres'? Past scholars in history meant it literally. • Lengths of 270 to 360, a ratio of 4:3, is called a perfect fourth. • Lengths of 288 to 360, a ratio of 5:4, is called a major third. • And a ratio of 1:1 is called the first or keynote. I prefer to think of it as the price pivot note from which all others are derived.
Gann Analysis: Calculating
Price and Time Objectives
• 219
corrections, but it cannot define when it starts and ends as Gann's work can offer. When Gann analysis creates a congested mess of cycle targets in a wide interval of time on a chart, this is not confluence. It can be thought of as disharmonic noise in time. The result is a market that is unable to decide the direction it w i l l move. Noise can be thought of as two sine waves fighting one another. The congestion of cycle beats must be worked through over time. When the market breaks past from the congested time period the trends that follow are often very free flowing and strong. Even i f for no other reason, this is why you should give Gann analysis a chance. The concept of vertical confluence targets i n time is very similar to when a market is trapped within a cluster of Fibonacci ratios on the horizontal axis. Then, when prices break through the congestion levels of resistance or support, the market trends freely toward the next confluence zone on the horizontal axis. The same concept applies to Gann time analysis on the vertical axis.
Vibrating strings can be drawn as oscillating sine waves along a zero axis. Harmonic waves share some common intersecting points on the zero line. W h y we end up learning about harmonic ratios is that they explain why certain price objectives are respected by the market more often than targets derived by other methods. Targets derived from harmonic ratios also reveal which target is of greater importance i n a series of objectives. One of the primary reasons I entered this field of study was to see if Gann analysis could identify when not to trade a market. Technical analysis has no other method that can more clearly show when to stay out of a market and identify for how long. Elliott wave has its guideline of alternation between
Time Cycles: Examples of Gann's Analysis on the Vertical Axis Time objectives are significant when time cycles congest in very narrow bands that confirm one another. I f confluence is important on the vertical axis, there must be a way to test what cycle should be used. Figure 9.4 is a subset of daily Wheat data from the 1900 to 2011 series. The software is called Market Trader Gold by Alphee Lavoie and Sergey Tarassov. The program will swear at you in Russian English and is far from intuitive to use, but it is brilliant because a Russian space program wizard and a professional
220
•
CALCULATING M A R K E T PRICE OBJECTIVES
Gann Analysis: Calculating
Price and lime Objectives
• 221
astrologer have developed the perfect search engine to identify cycles that apply to Gann market analysis. I use a program called Market Analyst with flashier graphics to chart and present some of the astronomical research results, but some of the cycle conditions derived from Alphee's program cannot be duplicated in the Market Analyst. It is very difficult to find software that is easy to use, stable, consistent, and all inclusive i n one package. Some readers will know that Peter Pich developed the first Gann charting program, called Gann Trader 3. It was accurate, but incomplete and very thin to nonexistent on the astronomy side needed for time analysis. W i t h his passing, the product remains in Microsoft DOS format and is slowly becoming extinct. There is no easy solution right now . Market Analyst is rewriting their entire software product in need of a new platform to resolve several issues. I have not seen the new platform at this time. I have invested years of detailing feature specifications for this product. But the previous versions have been unstable for advanced Gann users. It is hard to stay true to Gann methods in a very exclusive and small market. Soon as numerous additions are added to any product it becomes more challenging to learn. However, this one has the greatest potential of getting it right. Figure 9.4 has an imbedded cycle that slowly expands the timing between cycle beats and then contracts. This is common for cycles based on orbital motion because planets follow elliptic patterns that give the appearance a planet can change speed along the path. To see the expansion and contraction character of a cycle, use Figure 9.4 to look at the x axis near 1994. The cycle beat is nearly aligned to the axis hash mark denoting the start of the year. As you scan left the hash mark for a year on the x axis no longer aligns with
222
• CALCULATING M A R K E T PRICE OBJECTIVES
Gann Analysis: Calculating
Price and Time Objectives
• 223
in ancient Egypt. Gann travelled to Egypt to further his studies.
of the data, but the alignment with the start of a year returns. As
more interesting when you know that Mercury had great significance
the cycle beat. The contraction phase is not visible in this segment
System is called geocentric. The Earth s tilt on its axis creates several
physics departments. Focus on discussions of motion.
June and July. The perceptions form Earth as the center of our Solar
mation posted by universities for astronomy courses offered by their
the Sun moves faster in December and January, and more slowly in
astronomy from the hiternet. Begin by exploring the online infor-
not uniform? The seasons are of different lengths of time because
from earth it is called geocentric. Beginners can learn the basics of
Speed? D i d you know the mohon of the Sun along the ecliptic is
tions relative to one another. I f we consider their positions as seen
direction, or mathematical orbit measurement has been evaluated.
planets. As planets move around the sun they change their posi-
multiple combination, individual speed, acceleration, position,
looking at a planetary cycle based on the aspect angle between two
ets in combination? Well in fact every planet, every angle, every
of expansion and contraction rhythms within the history. You are
Why look at the Sun and Neptune rather than some other plan-
the chart scrolls toward 1900 the cycle beats clearly display phases
to the barycenter of the Earth-Moon system.
and Neptune appear to be very close to one another as seen from
perturbations of our orbit by other planets and our motion relative
L I is a formula that will map every occurrence when the Sun
small differences in the length of seasons from year to year due to
cycle beats are again caused by the elliptic orbit of the planets.
mathematical issues that must be considered. In fact there are also
In Figure 9.4 the expansion and contraction character of the
record.
produced a different cycle beat. The angles between Gann's Fan
from the cycle beat over the entire 700 years of Wheat data on
be in opposition to one another. The opposition aspect would have
tested further to see what kind of market action might be expected
been 180 degrees from Neptune (plus or minus four) they would
mine the significance of this event. Only then are the signals
example that the Sun is conjunct with Neptune. Had the Sun
markets. The cycle must then be statistically evaluated to deter-
plus or minus four degrees i n orb is close enough to say i n this
with water define one of the cycles of great influence on the grain
ets are in orb. A zero degree alignment is rare. Therefore we accept
Interestingly the Sun that gives light and Neptune associated
Earth. We refer to the four degrees criteria as the period the plan-
lines have been found to be of significance in aspect planetary cycle studies.
Astronomy is a science and the universal language of science is mathematics. The new cycle in Figure 9.5, L2, is called an Ingress
only one degree of angle along the celestial horizon has been
Mercury as the center. This mathematical relationship becomes
study there would be 360 degrees between each cycle beat because
a mean distance of 1.618 when measured in astronomical units from
sun passes that precise angle it forms a cycle beat. In this particular
Analysis that Ceres is needed to show how the planets all add up to
the Sun enters a specific degree on the celestial equator. As the
idly. The asteroid Ceres is important. But you will read in Fibonacci
study. L2 is directly under the price data and marks every occasion
Alignments with more than two planets will magnify the task rap-
224
• C A L C U L A T I N G M A R K E T PRICE OBJECTIVES
Gann Analysis: Calculating
Price and Time Objectives
• 225
r
300200TRANS lO)-*!*?)
Efficiency l e i t for Asi^qilog^fli E v e n
Figure 9.5 Aerodynamic Investments Inc., www.aeroinvest.com Source: Market Trader Gold, Alphee Lavoie and Sergey Tarasov
requested (Sun entering Aries). Figure 9.5 includes an efficiency test for this astrological event. The cycle beat is on the zero axis and the -20 to +20 scale shows 20 days before and 20 days afterward. Gann gives time cycles in his books, but he does not give you charts to make it easy to see his references visually. It is not until you create the chart that it is clear Gann did not reference the start or end of significant price swings. This is very confusing at first, but Gann was teaching us what cycles he used for specific markets. Patrick Mikula makes a valuable study in his books, but they are incomplete and Patrick did not filter out John Gann's work that must be ignored. Gann's work is amazing considering it all had to be done by hand. Be aware he made errors and omissions because he lacked the ability to use a computer. He also never traded a market overseas. Do not assume what works in North America translates exactly in China, as an example.
'Zl'Jioin^Z'!^^^^^^
investments Inc., Daily Market
Source: Charts by Market Analyst 6, Copyright 1996-2011 We are still discussing vertical cycle analysis used by Gann to define time objectives. Figure 9.6 is a two-week chart of Corn prices. This is a different Ingress study where the planet Venus is tracked when it enters the ninetieth degree or cusp of Cancer. With the cycle beats marked on the chart you can easily see the
226
• C A L C U L A T I N G M A R K E T PRICE OBJECTIVES
Gann Analysis: Calculating
Price and Time Objectives
*227
ets move in a direct (eastward) motion around the Sun, but the planet
vertical lines on the bottom of the price window. The expansion
caused by its being lapped by another planet, or vice versa. Both plan-
expansion and contraction character of the downbeats on the and contraction phases are also rhythmical. We have only touched on two different kinds of astrological cycles Gann used. Other cycles you will want to learn about, other than ingress movement and aspect relationships, arc solar and lunar eclipses, moon phases, and retrograde motion to name a few more. In the book Breakthroughs in Technical Analysis (Bloomberg Press), I had a copy editor who corrected my correchons describing retrograde motion several times. I finally submitted a title to the graphics department so the image would be correct, although the copy editor explains the retrograde movement backwards. We are all so visual no one has brought this inconsistancy to the publisher of my attention. However, it is wrong in that book. So let me use a NASA image to do a better job on this particular time cycle explanation, (see Figure 9.7). Most people use the term 'retrograde' to mean retrograde motion. Retrograde motion is the apparent backward motion of a planet
with the inside (smaller) orbit moves faster than the planet on the outside (larger) orbit, and when it passes the slower moving planet, each sees the other one as apparently moving backward relative to its usual motion around the sky. In this 'retrograde' motion, neither planet is actually moving backward; it only appears that way, during the time that one laps the other. There are several periods we Gann analysts study to find the best correlation with a market: the period when the planet enters its retrograde motion, the interval of time just before retrograde when the planet appears to decelerate its orbital speed, the time when the planet seems to stall and have no motion, the period it returns to direct motion, and the period when it accelerates away and returns to the original speed along its normal orbital path. There are two other types of retiograde movement in astionomy, but we do not concern ourselves with the backward path of some moons as they orbit around their planets or if a body spins in reverse on its axis to most other bodies in the solar system. (On Earth everyone is shifting to the east. 'Clean cups please.' A.K.A. Alice in Wonderland.) Easi
I wish I had seen the NASA illustration earlier because it makes retrograde motion drawn within market data so much easier to understand. Retrograde Motion is found in a Gann market chart in two different ways. The first is to mark the cycle itself for the planet as was illustrated in Figure 9.6 when an ingress cycle was identified. How-
Ofbit •
'
ever, retrograde cycles vary because inner planets create a downbeat
1
Vl/es,
that is very narrow in time and the outer planets create wide bands
fNASASP-AJial
over longer intervals of time because they move so slowly. This Figure 9.7 Source: www.history.NASA.GOV/SP-4212/p3a.html
changes the width of the cycle duration.
228
• CALCULATING MARKET PRICE OBJECTIVES
Examples of Gann's Analysis on the Diagonal Axis: Planetary Lines
Gann Analysis: Calculating
Price and Time Objectives
•
229
let's say to view Venus you need to turn 47 degrees. Now add 360 to 47 until the value enters the range of prices on the chart. That works for the DJIA for the moment, but Figure 9.8 is trading in a
range of Corn. Therefore, factors are involved.
chart is when planetary lines are plotted as channels. Figure 9.8
range where 360 cannot be added to 47 without exceeding the price
The other way you will see retrograde motion represented in a Gann
the chart. During periods of retrograde the channel line begins a
Earth s view that the lines change very slowly.
base and 360 * 0.1 is then added to the new channel position on
motion as well, but the planets move so slowly from day to day from
where it can be seen. Let's say the new position is 49 degrees as the
closest to the Sun. The outer planets create orbits with retrograde
equals 0.1. With each new day the planet has moved to a new angle
diagonal channels derived from inner planets, or the five planets
The multiplier for one harmonic in Figure 9.8 is one degree
is a daily chart of Corn futures. The squiggly lines are typical of
geometric tools to step through a process that becomes very precise.
tion at a specified degree within a circle. Within the 360 degrees.
not be addressed here because we need Gann Squares and other
tion of all the planets at that moment, each planet will have a posi-
change. But the process of correctly defining the factor to use can-
Exchange on market close, and take a snapshot to freeze the loca-
specified market. When you know the correct factor it does not
to accept. If you are standing in the middle of the New York Stock
chart 360 units higher, or by means of the factor required for the
a channel line derived from a planet is a tougher challenge at first
channels are created from the first line and replicated through the
easy to do; however, believing the market has any relationship to
wave action as seen in the NASA illustration for Figure 9.7. The
How do you plot a planet's orbital movement on a chart? It is very
When the channel forms a straight line it is mapping an outer planet that is exceptionally slow. Figure 9.8 has the channels for three different planetary lines. These particular lines cannot be reproduced in Market Analyst. The factor used for Corn used the term 'one harmonic.' The word harmonic in this context is widely misunderstood in numerous Gann books. If you want to take a single channel and subdivide it in two, it will have a parallel line in the middle of the channel. We would consider this a 50 percent division line in a conventional channel, /
/
/
Figure 9.8 Aerodynamic Investments Inc., www.aeroinvestxom Source: Market Trader Gold, Alphee Lavoie and Sergey Tarasov
but it is not linear geometry being applied. Spherical mathematics will state two harmonics have been created. If one harmonic creates a channel line every 360 units from the y axis, then two harmonics
230
• C A L C U L A T I N G M A R K E T PRICE OBJECTIVES
Gann Analysis: Calculating
Price and Time Objectives
• 231
w i l l be drawn every 180 units. You need the original channel range of 360 drawn i n one color to f i n d it again. T h e subdivision lines w i l l be different colored lines. Three harmonics w i l l mean a range spread of 360 is divided into equal thirds. So the math is easy. But there is a lot of misinformation on the Internet and i n print. Planetary lines show you one example of Gann's diagonal calculations. D o I personally use them? N o . But I do not trade f r o m movi n g averages o n price, or use standard deviation volatility bands, or favor conventional channels applied to any chart for any t i m e horizon. Have I ever needed channels? Yes. It was a tool that solved the problem when a former boss asked me to give a price target for the Italian Lira and T h a i l a n d Bhat currency cross. I a m showing my age as this cross n o longer exists. But the data looked like Swiss cheese w i t h gaps after every bar as this was the d e f i n i t i o n o f t h i n market. To solve the problem I created a line chart and used channels. T h e trader that favors channels i n general w i l l be intrigued by this example of Gann's work o n a diagonal axis. T h e last example demonstrated just one m e t h o d along the diagonal that u t i l i z e d planetary m o v e m e n t . N o w we w i l l step away f r o m methods derived f r o m astronomy and focus on a few geomet-
Figure 9.9 Aerodynamic Investments Inc., www.aeroinvest.com Source: Charts by Market Analyst 6, Copyright 1996-2011
ric applications. u n i t o f t i m e is equal to one u n i t o f price. T h e d i f f i c u l t y follows
Examples of Gann's Analysis on the Diagonal Axis: Fan Lines and Gann Squares
for people because they do not k n o w or have forgotten the simple m a t h associated w i t h a box. I n Figure 9.9 there is a m o n t h l y chart o f the N i k k e i 225 Index. I n t h e same chart are four boxes. Box
360 degrees. For the box to have sides o f equal length and still be
differences here. )ust keep i t simple to develop the concept. O n e
the center, what shape w o u l d I have drawn? A circle. A circle has
I w i l l n o t be addressing the issue o f calendar versus t r a d i n g day
corners and had each point along the box an equal distance f r o m
often misunderstood. T h i s is simply a box whose sides are equal.
n u m b e r one was created as the square of 90. I f I smoothed out the
Squaring price (the horizontal axis) and t i m e (the vertical axis) is
232
• C A L C U L A T I N G M A R K E T PRICE OBJECTIVES
Oann Analysis: Calculating
Price and Time Objectives
» 233
an exact 45-degree angle.
equal length. How many sides are there i n a triangle? Three. How
multiple of two. Therefore the square of 45,90,180 etc. will contain
one. Turn the circle with 360 degrees into a triangle with sides of
ognize this about the square of 90, or any square of 90 that is a
right-angled. Four times 90 equals 360 degrees. A l l right, try this
of other outer boxes. It becomes extremely important that you rec-
connected as a solid, the four corners must be equal to 90 to be
very big wall to create the chart—not practical. So how do you fit
people a problem.
21,000. I would need 21,000 days to 21,000 price units and a very,
equals 120. I do not see the difficulty here, but this causes many
time relationships. This chart has an old price high still visible near
has three sides of equal length. Therefore 360 divided by three
Nikkei does not accommodate the use of one unit for the price and
at the corner? The answer is 120 degrees. A n equilateral triangle
Here is a new question for you to consider. The range of the
many degrees will be between two sides of the triangle that connect
a Gann software vendor believes. As soon as you start messing with
and y axes will affect how the box looks to us.
mine the correct factor to use. In addition, it is not experimental as
of a computer screen. Computers and how we elect to scale our x
with a sound-bite quick answer. There is a very precise way to deter-
Yes. It will often look less than perfect because of the pixel array
correct? That is one of the big questions that cannot be answered
fect? Are the lengths of the sides for height and length really equal?
such a price scale into a chart and keep the geometric proportions
Now study box 1 in Figure 9.9. Why does it look less than per-
the horizontal, vertical, and diagonal.
four boxes. Multiply four boxes by 2 and the screen would display 16.
tion, they do not understand how the dimensions interlock between
height and the box length will be twice as long also. Therefore I get
your sources advise you Gann analysis is only about experimenta-
a box by two, the one side (height) doubles for the dimension of
has turned into an experiment that is no longer Gann analysis. I f
two? Most people do. You have forgotten this is a box. If I multiply
thinking it might be better if a Fibonacci ratio were used, the analysis
will you create across the Nikkei chart? Did you think the answer was
the proportional relationships between width and length, such as
Next, if you have one box and multiply it by two, how many boxes
one unit to the right, equally dividing the right-angle corner of
corner is divided equally by two.
ner with a growth ratio of 1:1, the line will move one unit up and
the line passes from corner to corner. Therefore the right-angled
triangle into very specific angles. If you create a line from the cor-
lies between lines ah and hcl It has to be an angle of 45 degrees as
corner on the bottom left. A Gann Fan subdivides a right-angled
box. What is angle ahc that defines the top left corner of box 1 that
is the addition of Gann Fan lines that subdivide the right-angle
the corners with lines they will cross dead center i n the middle of a
with four Gann Squares as described in Figure 9.9. The difference
truly a box with equal lengths in height and width, when I connect
Consider Figure 9.10 where the monthly Nikkei data is displayed
Box 1 has a cross in it that connects the corners. I f my box is
Notice a line connecting the corners of box 1 will pass through the corners of box 4. Every line of 45 degrees will connect corners
90 degrees. It becomes a 1 x 1 line that is a 45-degree line through the chart.
234
• CALCULATING M A R K E T PRICE OBJECTIVES
Gann Analysis: Calculating Price and Time Objectives • 2 3 5
many units along the x axis you move before moving up the y axis scale. So 1 X 16 is over one unit and up 16 units. We mention the use of 1 x 3 and 3 x 1 separately because these are thirds and they do not multiply by factors of two and do not retain an important relationship about to be demonstrated. I f you have software with a tool that looks like Gann Fan lines, they must be these specific angles. If not, they are simply speed lines used to observe acceleration and the slope of data swings. Figure 9.10 contains circles so you can see more easily the Gann Fan lines that intersect box corners and the diagonal crosses. These relationships between Gann Fan lines and box proportions are important. These relationships also relate to the Gann Wheel that defines the horizontal axis. When a software vendor has an error in any one of these three tools, the Gann Square, Gann Fan, or Gann Wheel, it will be recognized by the advanced Gann practitioner because the factors used will no longer be correct that bind these proportional attributes. If I multiply my Gann Fans drawn in Figure 9.10 by two, where would the 45-degree line that is in the middle of the chart be drawn next? It would shift and be drawn exactly on top of the line above the 1 X 1 in Figure 9.10 and become the new 1 X 1 line. Why? It is a line and not a box. So the 1 x 2 moves to 1 x 4 and the 1 x 4
1 X 3 and 3 x 1 will intersect the corners and intersecting crosses
3 X 1 and 1 x 3 will be on degrees of 18.75 and 71.25.
many boxes form inside one box that is divided by three? Nine. The
26.25, 45, 63.75, and then 82.5 and 86.26. The thirds created from
separately. But Gann used to subdivide each box into thirds. How
will dissect the right angle at the following degrees: 3.75, 7.5, 15,
moves to 1 X 8, etc. That is why the 1 x 3 and 3 x 1 are menhoned
A Gann Fan is a series of lines subdividing 90 degrees. Each line
The ratios needed to create these lines precisely from price and
within the nine boxes of a single larger box exactly.
Fan lines that do not connect to any corner or middle intersection
then 1 X 2, 1 X 4, 1 X 8, and 1 x 16. In each, the first value is how
I have not made a comment about the highest and lowest Gann
time are 16 X 1, 8 X 1, 4 X 1, 2 X 1, 1 X 1 (the 45-degree angle line),
236
• C A L C U L A T I N G M A R K E T PRICE OBJECTIVES
Gann Analysis: Calculating
Price and Time Objectives
• 237
has to be more to it, and there is indeed.
be confusing because he calls them moving averages and not trend
analysis. But diagonal analysis is never used alone. Therefore there
Ganns courses and he describes how he draws Gann Fan lines, it can
I would always favor Astrological cycles over this method for time
and find the boxes continue above and to the right. When you read
Gann Square will have inconsistent results and poor performance.
of the top left box is heading toward a corner when you zoom out
2011.) This method of just looking for intersecting points from a
and height. The line circled in Figure 9.10 that bisects the middle
tive from astrological cycles for the SP500 will be October 22,
additional boxes of equal size are added to a larger chart in width
a diagonal target in the year 2011. (The Gann vertical time objec-
in Figure 9.10. They will bisect corners or intersecting crosses when
measurement from thirds. Point A was not i n Figure 9.10 either,
ing about Gann analysis.
B was not even identified in Figure 9.10 because it falls upon a
touch on the tip of the iceberg to give you a conceptual understand-
a Gann Wheel and they are confirmed by diagonal angles. Point
That is a good question which I have to leave with you. I can only
they occur when price targets have been realized as derived from
traders. How do you find the 45-degree angle in a square of 144?
only signals in this chart of Gann significance. Why? Because
a diamond-shaped appearance and will be of interest to soybean
the bottom left. The circled intersections at A, B, and C are the
90. We use other squares. For example, the square of 144 will have
Gann Fan area under the 45-degree-angled line originating from
I have just demonstrated some of the attributes of the square of
and 9.10. Two gray areas will help your eye stay within the white
working with a harmonic cycle that is better named a moving average.
Figure 9.11 is the same monthly Nikkei chart as Figures 9.9
lines. When you delve more deeply into Gann you will realize you are
involved on the most elementary level. The Nikkei price high in
9.10 the intersection of box 4 (bottom right) has a circle showing
can identify for you and offers an introduction to the geometry
importantly, Gann did not use the boxes in this manner. In Figure
nal converge at points A, B, and C. This is what Gann analysis
cycles are not the best cycles to use in trading applications. More
(not shown). Therefore all three cycles in price, time, and diago-
all they have done is to create a fixed cycle again. We know fixed
which developed a confluence target zone from astrological cycles
Books and books write about these patterns giving you time, yet
bottom left box as resistance. It is also a vertical target in time
when the intersection is extended vertically down to the x axis.
right box. It is also using the Gann Fan line radiating from the
cross in the middle are always half the length or width of the box
zontal axis. It bisects a descending 45-degree line in the bottom
length of each box is equal from one to another. The lines that
240-dcgree price target from a Gann Wheel falls on the hori-
to be the time objectives for trend changes. Think about this. The
diagonal, and vertical axes. For example, point B is where the
Some describe the intersecting corners and centers of the boxes
are of interest because confluence develops on the horizontal,
very precise. There is no room for guessing. But many books do guess.
adding objectives on the horizontal axis only points A, B, and C
ing that the relationship between Gann boxes and Gann Fan lines is
but could have been as it fits the criteria for that discussion. By
The key thing to take away from this discussion is the understand-
238
• CALCULATING MARKET PRICE OBJECTIVES
G a n n Analysis: Calculating
Price and Time Objectives
• 239
G a n n used other wheels such as the hexagon wheel, w h i c h is a honeycomb pattern. T h e honeycomb pattern is better for calculati n g i n units o f thirds such as 30 and 60 degrees. T h e wheel called the Square o f N i n e starts at the center w i t h a value o f 1 and the first box ends at the n u m b e r 9. I n Figure 9.12 you can follow the numbers to 9 and then r u n into the first tricky part where beginners can make an error. T h e b o t t o m left corner requires you to move outward by one square and then the numbers w i l l continue to increase by one sequential u n i t again u n t i l the next outer square is f i l l e d again. T h e numbers 81 and 82
GANN W H E E L A N A L Y S I S 0
45»<
June
92 93 94 95
May 6
97 98 9 9 1 0 0 102
Figure 9.11 Aerodynamic Investments Inc., Daily Market Report, www.aeroinvest.com Source: Charts by Market Analyst 6, Copyright 1996-2011
103
.32 11 1 6 W 270'
2 1
11
4
12 9
2007 was n o t discussed. M u c h has to be overlooked i n this i n t r o -
19
S 23 2 2 !
79 7 « 77 76 75 7 4 > 3
ductory discussion.
Gann Price Analysis on t h e H o r i z o n t a l A x i s : The Gann Wheel H o w do you extract price objectives f r o m a G a n n Wheel? Figure 9.1 is a G a n n W h e e l called the Square of N i n e or a harmonic wheel.
Nov?
F«b4
Dct2l 180 '<
A e r o d y n a m i c I n v e s t m e n t s Inc
http:/A«ww.aeroinuest.com
Figure 9.12
240
• C A L C U L A T I N G M A R K E T PRICE OBfECTivES
Gann Analysis: Calculating
Price and Time Objectives
• 241
move develops. Currencies love 90 degrees. Each market has its
another and how this pattern becomes a calculator.
Bonds seem to be attracted to 120 degrees w h e n a counter-trend
rotation t h r o u g h the wheel. Here is how the numbers relate to one
and 360. A r e all these angles important? N o — n o t i n all markets.
row. T h e numbers w i l l be incremental as you move i n a clockwise
interest to us i n financial markets are 45,90, 120, 180, 240, 270, 315,
pay attention you can easily f i n d yourself o n the w r o n g c o l u m n or
left out the targets f r o m three other degrees. T h e angles o f greatest
on the b o t t o m left clearly illustrate this movement. I f you do not
T h e horizontal and vertical cross-sections display a double line so that I can later rotate the cardinal cross and angles o f the circle
own personality that is consistent. How
do we calculate the m a j o r support levels f r o m a stock
also have been used.
opposite $15. (We w o u l d still say, '$23 is 180 degrees u p f r o m $15,'
G a n n used a wheel w i t h increments o f .50 cents. A factor c o u l d
then 180 degrees up f r o m $15 is $23. T h e target o f $23 is directly
need targets t h a t d i d n o t have such a large spread. T h e r e f o r e
45 degrees up f r o m $15.' N o w skip over to 90 degrees and f i n d $19,
used a value o f $15. B u t a stock t h a t trades i n this range w o u l d
o f numbers i n rows and columns. W e w o u l d use the phrase '$17 is
objectives for the next example. T h e first example we covered
n u m b e r 17 is the value where the 45-degree line bisects the path
i d e n t i f y these price targets, you w i l l be able to o b t a i n the price
the center o f the circle travels toward the top right corner. T h e
at $4, the n u m b e r just below $15. I f y o u can f o l l o w the wheel to
45-degree price objective f r o m 15 is 17. T h e 45-degree l i n e f r o m
numbers 10 a n d 9.) N o w skip to the 360 degree target, w h i c h is
adjustments; just read the G a n n targets straight o f f the wheel. T h e
180 degrees d o w n is at $8. ( D o n o t forget to j u m p the corner at
the n u m b e r 15 is aligned w i t h zero, we need make no f u r t h e r
left diagonal. T h e $11 level w o u l d define 90 degrees d o w n . A f u l l
360-degree angle marked o n the top o f the wheel. Because
and
left. I t is located at $13 where the 45-degree l i n e crosses the top
G a n n W h e e l . T h e number 15 is perfectly aligned under the zero
terclockwise. T h e first target w i l l be o n a 45-degree l i n e to the
using a Square o f N i n e G a n n Wheel? F i n d the n u m b e r 15 o n the
f i n d support because the n u m b e r s decrease as you move c o u n -
were trading at a price of $15, what w o u l d the price objectives be,
d i d to f i n d resistance levels, we w i l l move counterclockwise to
the wheel becomes a calculator to create a price objective. I f a stock
pivot o f $15 using the wheel? Instead of m o v i n g clockwise as we
as a fixed u n i t w h e n zero' is set over a starting price. Here is h o w
fractions? T h e r e is indeed more t h a n just these basics.
how a G a n n W h e e l is used. To make your first example easy I have
for the Nikkei? H o w do you use a wheel for a market that trades i n
f i l l i n the blanks and complete the wheel i n this diagram.) T h a t is
off the wheel once the start is set correctly. H o w do you use a wheel
up f r o m 15 is 34. (You w i l l have to add t w o numbers f r o m 32 to
T h e first target f r o m 87 w i l l be 92. It is just a matter o f reading right
level.) A t 270 degrees, $28 is the price objective. A f u l l 360 degrees
that the zero line at the top now crosses t h r o u g h the n u m b e r 87.
the wheel, the results o f the angles are always ' u p ' f r o m the starting
the M y l a r overiay o n the top o f the numbered boxes to the left so
began. As l o n g as you are o b t a i n i n g higher price objectives f r o m
W h a t i f the price low is $87? T h e start has to be changed. M o v e
even t h o u g h 180 degrees is directly opposite the place where we
242
• CALCULATING MARKET PRICE OBJECTIVES
The best Gann Wheel on the market is made by the Lambert-
Gann Analysis: Calculating
Price and Time Objectives
• 243
45 degrees (up) = ((SQRT($A$27)) + 0.25) * (SQRT($A$27) + 0.25)
360 degrees = ((SQRT($A$27)) + 2) * (SQRT($A$27) + 2)
Why go through all these time-consuming steps?
315 degrees = ((SQRT($A$27)) + 1.75) * (SQRT($A$27) + 1.75)
Let me plug in the starting value and run a list of angles to targets.
270 degrees = ((SQRT($A$27)) + 1.5) * (SQRT($A$27) + 1.5)
wheel. But if you have a computer, why recreate a physical wheel?
240 degrees = ({SQRT($A$27)) + 1.37) * (SQRT($A$27) + 1.37)
made a great effort to create a graphic reproduction of a physical
180 degrees = ((SQRT($A$27)) + 1) * (SQRT($A$27) + 1)
You will find Gann Wheels within Gann software and they have
120 degrees = ((SQRT($A$27)) + 0.67) * (SQRT($A$27) + 0.67)
great job of writing a booklet that describes the logarithmic series.
90 degrees = ((SQRT($A$27)) + 0.5) * (SQRT($A$27) + 0.5)
Gann Publishing Company (www.wdgann.com). They have done a
The Square of Nine is a very cool calculator. A four year project
spreadsheet and you must still reference where you entered the
this calculator.
a habit of cut and pasting formulas in other locations within the
offer. But we can look at one of the mathematical properties behind
I have entered 34 i n column A, and row 27. Use '$' i f you have
Egypt. It has much more to it than just this brief introduction can
$A$27 is a fixed cell location where the pivot price low is entered.
to decode it took me through a Journey back in history and to Cairo,
The Gann Wheel is really a square root calculator. The square root of 15 from the first example is 3.873. Now add 2 and it equals
original price low. The spreadsheet results for a price low of $34 are:
5.873. The square of 5.873 is 34.49. Rounded to 34 it is 360 degrees 34.00
this method creates issues once you try to extend past 360 degrees.
start
up from 15. This is a full 360-degree rotation of the circle. But using
45
90
36.98
40.08
120 42.26
180
_j
46.66
240 51.85
270
315
53.74
360
57.47
61.32
The simple formula above is not the whole story. The logarithmic properties of the wheel will cause errors if you do not understand what you are working with when doing math in harmonic ratios. But stay within the first 360 degrees and you will be able to create your own wheel.
The Gann Wheel results for a price low of $34 are:
37
34.00
45
start
53
51
46
42
40
270
240
180
120
90
61
57
360
315
These results in the spreadsheet track closely to the values
Create Your Own Gann Wheel in Excel To identify levels of resistance from a price low, use the following formulas in Excel for each of the degrees. The formulas are:
extracted from the Gann Wheel. However, the number 34 is very close to the center of the physical Gann Wheel. How well will these formulas track i f you move further out or away from the center of the Wheel?
244 • C A L C U L A T I N G M A R K E T P R I C E O B J E C T I V E S
The spreadsheet results for a price low of 721 are: Start 721
45 734.49
90 748,10
120 757.43
180
240
7 7 5 . 7 0 796.45
270 803.80
315
360
8 1 8 . 0 4 832.41
The Gann Wheel results for a price low of 721 are:
Gann Analysis: Calculating
Price and Time Objectives
• 245
45 degrees (down) = ((SQRT($A$27))-0.25) * (SQRT ($A$27) - 0.25) 90 degrees = ((SQRT($A$27)) - 0.5) * (SQRT($A$27) - 0.5) 120 degrees = ((SQRT($A$27)) - 0.67) * (SQRT($A$27) - 0.67) 180 degrees = ((SQRT($A$27)) - 1) * (SQRT($A$27) - 1) 240 degrees = ((SQRT($A$27))-1.37) * (SQRT($A$27) - 1.37) 270 degrees = ((SQRT($A$27)) - 1.5) * (SQRT($A$27) - 1.5)
start 721
45
90
120
180
240
270
315
737.00 748.00 760.00 776.00 796.00 804.00 321.00
360 832.00
?15 degrees = ((SQRT($A$27)) - 1.75) * (SQRT($A$27) - 1.75) 360 degrees = ((SQRT($A$27)) - 2) * (SQRT($A$27) - 2)
The resistance targets from the spreadsheet for 45 degrees, 120
The method of convention when using a physical Wheel is to
mulas track if you move even further out or away from the center
/ do not recommend using the spreadsheet beyond 360 degrees.
degrees, and 315 degrees are slightly off. How well will these forof the Wheel?
continue past 360 degrees as you might with Fibonacci expansion
The spreadsheet results for a price low of 1624 are: Start 1624
45
90
120
180
270
240
315
calculations to identify a target at the 2.618 level as an example. 360
1644Jil 1664.55 167&45 1705.60 173630 1747.15 1768.11 1789.20
The Gann Wheel results for a stock price low of 1624 are: start 1624
45
90
120
180
315
270
240
360
1647.00 1664.00 1677.00 1706.00 1738.00 1747.00 1770.00 1789.00
But this is not the most accurate way to produce a longer horizon price objective. The set of formulas I have released above will work well until you need a conversion. How do you put the Treasury T-Bond price scale with 32nds into a Gann Wheel? How do you handle the Nikkei Index? How do you handle currencies that trade less than one? How should an intraday scalper adjust the spread of values and keep
important steps remain as you advance.
315, I recommend that you do. The results from the Gann Wheel
addressed here, but they show you where you will be heading and
series of numbers represents. As a result most people do not use
inner-wheel dimensions or visa versa? These questions cannot be
sheet will always be a tad off due to the type of calculation the
of a Gann Wheel, why would you want to convert them into the
the formulas above are correct. The 315 angle i n the spread-
them meaningful? I f you have values obtained from the outside
You are not going to track the Gann Wheel much closer so
will give you levels to practice reading from a Gann Wheel itself. To find targets from a price high to identify levels of support, the formulas will be the following;
For years I used the following i n the S&P. It is different, but extremely successful for how that market moves intraday. It will also start you off with a clue on how to handle the questions raised above.
246
• C A L C U L A T I N G M A R K E T PRICE OBJECTIVES
The formula for the S&P 500 is: 45 degrees (down) = ( ( ( S Q R T ( $ C 2 9 ) ) - 0 . 2 1 ) * ( ( S Q R T ( $ C 2 9 ) ) - 0.21)/10) + 3 0 0 9 0 degrees = ( ( { S Q R T ( $ C 2 9 ) ) - 0.5) * ( ( S Q R T ( $ C 2 9 ) ) - 0.5)/10) + 3 0 0
Chapter 10
120 degrees = ( ( ( S Q R T ( $ C 2 9 ) ) - 0 . 6 7 ) * ( ( S Q R T ( $ C 2 9 ) ) - 0 . 6 7 ) / 1 0 ) + 3 0 0 180 degrees = ( ( ( S Q R T ( $ C 2 9 ) ) - 1 ) * ( ( S Q R T ( $ C 2 9 ) ) - 1)/10) + 3 0 0 2 7 0 degrees = ( ( ( S Q R T ( $ C 2 9 ) ) - 1.51) * ( ( S Q R T ( $ C 2 9 ) ) - 1.51)/10) + 3 0 0 3 6 0 degrees = ( ( ( S Q R T ( $ C 2 9 ) ) - 2 ) * ( ( S Q R T ( $ C 2 9 ) ) - 2)/10) + 3 0 0
Cell $C29 adds 3,000 to the handle of the traded price level. Why is there a conversion in cell ' C to add 3,000 to the price?
USING O S C I L L A T O R S WITH THE E L L I O T T WAVE P R I N C I P L E
Why do I divide by 10 and add 300 in separate steps? It makes life with Gann much easier down the road when you have to work with global markets of all kinds. Why do I use -0.21 and not -0.25 for 45 degrees down? I ' l l leave you with that carrot for thought to think about. There is a lot of information in this chapter. Why would I give this away? This is not the complete story to be sure. But nothing has been held back in this discussion about Gann, because the more you know about Gann, the more you will want to know about Gann. By giving you an honest foundation and solid footing to begin, you might be able to avoid some of the nonsense online that will send you off in the wrong direction. The only way I know to end this chapter is with a promise for the future.. 'to be continued'.
Objections and Misunderstandings
1
am not going to remove or change the original text because it had tremendous positive feedback. But first, may I please ask you
to consider the last chart that was published in the 1998 edition in this chapter. It is the next chart in Figure 10.39 (August 1998). M y suggested forecast for the DJIA using just the principles of R.N. Elliott is likely skipped over by readers today because they believe it to be the current chart for the DJIA. Compare chart Figure 10.39 (August 1998) to the recent chart in Figure 10.40 (August 2011) with a projection for the 2012 year ahead forecast. Figure 10.40 (August 2011) has a vertical line marking where August 1998 occurs in the price data. The indicated coiling pattern • 247 •
248
• CALCULATING M A R K E T P R I C E OBJECTIVES
Using Oscillators with the Elliott Wave Principle • 249
D J I A - Q U A R T E R L Y S e p t e m b e r 1 9 6 2 to A u g u s t 1998 A E R O D Y N A M I C INVESTMENTS INC http:/Ai>n»w.aerolnv*>t.coin
Elliott W*ve Intetpretttion Connie Brown, CMT Aug 1998
Triangle not intended to be scaled to the x-axIs for time forecast.
Figure 10.39
A e r o d y n a m i c I n v e s t m e n t s Inc., © 1 9 9 6 - 2 0 1 1 ,
Daily Ma rket R e p o r t , w w w . a e r o i n v e s t . c o m Source: TradeStation. © TradeStation Technologies
that followed had not even begun the first swing in 1998. IN fact, the last swing up from the end of wave 4 into wave 5 of (3) had not
Figure 10.40 A e r o d y n a m i c I n v e s t m e n t s Inc., Daily Mar k e t R e p o r t , www.aeroinvest.com Source: Charts by Market Analyst 6, Copyright 1996-2011
developed at the time of this suggested path. Yet that upswing prior
decline. Now I will leave you with the current forecast in Figure
my early studies of Gann and not using Elliott.) Then the coiling
wave E, the final swing in the pattern, to finish with a spectacular
chart that is an error of one bar. (The time target reference was from
tive patterns in this chapter. The outcome for the DJIA still requires
start of the first swing down began in January 2000. For a quarterly
in the major rally that was incomplete. You will read about correc-
chart states that May of 2000 is when the start might begin. The
actual data. Not an unacceptable forecast when the market was still
in the scaling of the x axis now or then. However, the text under the
suggested contracting triangle is likely an expanding triangle in the
pattern was a multi-year contracting triangle. Time is not implied
pattern in the DJIA began to track the forecast with one twist; the
to the start of the coiling pattern did develop. The suggested future
250
• CALCULATING MARKET PRICE OBJECTIVES
Using Oscillators with the Elliott Wave Principle • 2 5 1
'Analysts have been caught on the wrong side of the market
4,550 is a potential capitulation bottom if 5,400 cannot hold. Keep
'It is too subjective.'
10.40 (2011) for the DJIA. The minimum target is 5,400 but likely
Industrial Average.'
was first suggested in 1998. Now only the last leg is missing after
or with diametrically opposing views about the Dow Jones
in mind none of the coiling price swings were in place when this 13 years. How was this done? I know the guideline of alternation between corrective waves and I know that wave structure is fractal in nature. If you learn about the Elliott Wave Principle you will know
'Too many alternative market scenarios can be suggested by the same analyst.'
'If you scan different time horizons, for the same market, you
2 books on Elliott Wave. The book was very large so we split it. The
'Different analysts can interpret the exact same pattern in different ways.'
these things too. You may be interested to know that I have written first should be released April 2012 called Mastering Elliott Wave
can find any pattern you want to prove your market opinion.'
Principle: Elementary Concepts, Wave Patterns, and Practice Exer-
I can take these issues further by addressing each separately.
Global Cash Flow Analysis. My first editor moved to Bloomberg
common points often raised as objections toward this methodology
Wave Analysis: Complex Patterns, Intermarket Relationships, and
Did I leave any out? Probably, but I have certainly listed the most
cises. The second will follow shortly after called Advanced Elliott
tend that there are several causes contributing to the subjectivity
goes into a book is mind boggling.
the 'nay' voters? Don't jump to that conclusion so quickly. I con-
But it is no reflection on the publishers themselves. The work that
Did an Elliott wave practitioner just cross the line to stand beside
Wiley So my books cross all publishers in our industry That is rare.
Let s start with the first objection: 'It is too subjective.' Yes, it is.
from McGraw-Hill and Bloomberg then sold their book division to
lished services that violate the basic tenets of the Wave Principle.
useful by readers. I have a new way to help the 'wave deaf' that
interpretations demonstrated in countless newsletters and pub-
without edits. The real-time scenario that develops is said to be very
issue. The first problem is that there are a vast number of wave
The balance of this chapter remains as written in the first edition
has been favored by many traders who struggled in my new books. Help is on the way if this describes you. But this chapter will give you a real-time example that walks you through all the patterns in an environment that makes the first steps easier for most.
This problem can be resolved by specifically highlighting some very apparent and common misunderstandings within our industry However, these misunderstandings are not the only cause contributing to the subjectivity argument. I will show you in a few
proven wrong by the market. The chart in question will offer an
Principle are the following:
real-time context, I knew the interpretation would most likely be
common arguments used to build a case against the Elliott Wave
for this methodology. However, when this chart was labeled i n a
in our industry today is R.N. Elliott's Wave Principle. Some of the
pages a wave interpretation that meets every rule and guideline
Perhaps the most hotly debated and controversial analytic method
252
• C A L C U L A T I N G M A R K E T PRICE OBJECTIVES
Using Oscillators with the Elliott Wave Principle • 2 5 3
That is the point where the object will be balanced and the left or
can hit all the right notes, but it still comes out wrong.
the object at its fulcrum point by using no more than three fingers.
as others might be tone deaf or musically challenged. Some people
Now before you pick the object up, study it closely. You must pick up
in the movement of the markets, and some people are 'wave deaf'
all you gain by moving the object is information about its weight.
others would pay to hear played. There is a rhythm and harmony
do. Put the object on the floor by only grabbing the extreme top so
sufficiently to learn how to turn these same notes into music that
tool that is used to poke logs or shovel ashes. Even a broom would
duce a sonata written by Beethoven, few will develop their skills
a pencil is too short. Live in a big city? Find an object like a fireplace
people can learn how to press the correct keys on a piano to repro-
pen or pencil and think you will improvise. That will be too easy as
analogy can be offered to illustrate the difference. While many
stick from a tree that is at least two feet in length. Don't look at your
for the market's future path will require a higher level of skill. A n
not. I do not mean that you have an ability to paint or draw. Find a
data versus developing a wave interpretation that will provide a map
right-brain issue. You already know if you have an artistic ability or
The difference is that the skill level required to label the current
Might you be wave deaf? It really comes down to a left-brain/
interpretation that most ElHott wave practitioners would have used.
difference. He could not see the adjustments, and one day we had
there is no balance or proportion in their interpretation.
the picture would look straight to anyone else who could tell the
to state in the same breath that they will likely be wrong because
quietly 'fix' his final calculations that my eye also required so that
Wave Principle is to tell people they are technically right, but then
ture was perfect. However, my mother, an interior designer, would
overriding another technical method. The hardest aspect of the
perfection from the floor or ceiling. Scientifically he knew the pic-
confidence to listen to my gut feeling when price structure suggests
father would have been wave deaf as he measured every picture to
up wave frequencies and structure very quickly. That gives me the
picture to offset the multiple parallax errors within the room. My
that market s price chart. In equity indices, I know that I can pick
will not look straight if you use a level as you have to balance the
time because they are unable to see proportional balances within
rarely perpendicular or parallel. A picture in such an environment
tion. Unfortunately, some people cannot see wave structure at any
an old Victorian home where the corners, floor, and ceiling are
my current intraday problem in bonds is only a temporary situa-
you have to run for a level to rely on scientific fact? Good luck in
and actually adds weight to the longer-horizon outlook. Therefore,
proclaim a picture is straight when it is hung on the wall? Or do
time as it fits the wave character of the larger picture for this market
Another easy test. Are you the one in your family who gets to
I know why intraday bond charts are complex and choppy at this
you likely have a sense of proportion and balance in your mind's eye.
then consider wave counts only if they are very distinctive. However,
more than one finger's width to find the perfect fulcrum point, then
an example, I currently rely on Gann analysis for U.S. T-Bonds and
you pick it up. If you need to make only a minor adjustment of no
character or frequency of a market s action for a variety of reasons. As
right side will not suddenly take a dive back toward the ground when
If not outright wave deaf, one can be 'wave handicapped' to the
254
• CALCULATING MARKET PRICE OBJECTIVES
Using Oscillators with the Elliott Wave Principle • 2 5 5
on several occasions.
parallax distortion, and the shift made the picture appear straight
from their quote screens know I had to eat humble pie for dinner
base boards. The change was needed because the ceiling caused a
The institutional traders who used to monitor my live wave counts
shifted nearly seven centimeters between its top corners from the
sions when the market has proven my opinion to be dead wrong.
horrified to learn he couldn't see that a large painting had been
that particular outlook was correct, there have been other occa-
him measure a picture to prove we had made a change. He was
within its environment.
The S&P 500 is my specialty, and you will fully understand how
of applying the Elliott Wave Principle.
As an example, start your wave count in the middle of the market's
from using price structure alone. This is a very unorthodox method
a few methods that can help you find higher-probability scenarios.
that went into that wave interpretation will be entirely different
the basic building blocks of the Elliott Wave Principle. There are
a labeled chart will be defined, but the underlying considerations
ion this is the most important area of study once someone knows
technical indicators. The final results will look the same to you, as
pretations as a real-time trading tool might be weaker. In my opin-
summary, you will see that I construct wave interpretations from
understand other people's wave interpretations, but your own inter-
derive different opinions by the end of this chapter. In a capsule
cally handicapped or not. You can still learn to label charts and
two people can utilize the Elliott Wave Principle differently and
Anyway, if you think about it, you already know if you are graphi-
just used indicators.
monitored the live updates throughout the day on their Bloomberg,
Principle alone and you will be out of sync more often than if you
tional traders to consider with their own analysis. These traders
step off track using indicators alone. However, use the Elliott Wave
intraday real-time wave counts with market objectives for institu-
chronized dance will assuredly step off track. O f course, we can
500 Index. In addition to an evening report, I transmitted frequent
tion from either the Wave Principle or an indicator and the syn-
Atlanta. I was accountable for the day-to-day analysis of the S&P
will develop. However, overlook an important piece of informa-
Prechter a few years prior to the Centennial Olympic Games in
fect synchronization with most of the twists and turns the market
Dow Jones Industrial Average.' That is true. I worked for Robert
strated. Find the right balance, and it is possible to move in per-
side of the market or with diametrically opposing views about the
between indicator direction and price structure w i l l be demon-
point people make is that 'analysts have been caught on the wrong
to the subjectivity of our wave interpretations. The balancing act
We are not quite done with the subjectivity argument. A strong
with the Wave Principle and how they are used will contribute
few different methods used in this chapter to develop a wave count.
Therefore, the technical indicators one uses i n conjunction
move instead of from its origin or extreme high or low. You will see a
Reuters, Telerate/Bridge, and D T N quote systems.
I can see how that last sentence may be used out of context. But Wave Principle is of little value, let me point out the flaw in that
ish on the U.S. stock market. I will show you why later. While
before you jump to the conclusion that I have implied the Elliott
During the time I worked for Bob Prechter, I was extremely bull-
256
• C A L C U L A T I N G M A R K E T PRICE OBJECTIVES
Using Oscillators with the Elliott Wave Principle
• 257
lead only to an even stronger leg down in these markets? So the Elliott Wave Principle becomes the much-needed tool we can use to obtain information about the size of the reaction we can expect from any given indicator signal. Elliott is hard to live with, much harder to live without. Just don't live with it alone! Another common objection is that there are 'too many alternative market scenarios that can be suggested by the same analyst.' We can also fold into this same argument: 'Different analysts can interpret the exact same pattern in different ways.' An intellectually honest Elliott wave practitioner will admit that there are times when there are no discernible wave structures firom which to build a highly probFigure 10.1 Aerodynamic Investments Inc., ® 1996-2011, www.aerolnvest.com
able interpretation. It is better to openly admit that sometimes the Elliott Wave Principle has little more to offer than the development of a slew of scenarios. Every analytic method will have periods when better off keeping the ideas being considered to ourselves until the
that cannot be corrected by just adding another indicator or devel-
it is not decisive. Sometimes we just have to wait, and we would be
hasty assumption. Indicators have a very serious inherent weakness oping a new formula. Figure 10.1 displays similar sell signals when the Composite Index and RSI diverge in four different markets. How do you know just how large a correction will unfold from these identical signals? Oh sure, we have calculated targets, lots of targets, but which one will be the likely bottom? The monthly Hang Seng signal preceded a 56 percent decline. A daily S&P signal has so far produced a 6
market becomes more defined. Elliott takes numerous hard knocks in this regard. In an effort to pick up a lost trial, we may openly discuss all that we might think or hypothesize fi-om little more than muddled, congested price bars that form wandering data tracks on our screens. Guilty as accused. But recognize that one who 'sees' waves in market data will lose his or her own center of inner balance when he or she has lost the pattern of the primary markets.
incentive to dig deeper into the markets to see what is being missed,
that a countermove up can be expected, but will the next sell signal
So the inner struggle that goes on when you 'lose i t ' is the needed
will this market go?' You can see that the markets are oversold and
small world when the market's footprints have again been identified.
sell signals leave us with the same question: 'Just how far down
position within a larger framework of the Universe. All is right in my
cent, and it had not bottomed. These identical bearish diverging
its quiet knowing that 'my' S&Ps are tracking and are at a certain
found its first pivot. The weekly DMark futures chart fell 20 per-
My whole day goes out of whack just because my world has lost
percent decline. Gold fell 29 percent before the Composite Index
258
• C A L C U L A T I N G M A R K E T PRICE OBJECTIVES
Using Oscillators with the Elliott Wave Principle
• 259
as you do to calculating where you believe the market might go i f
peace. Sounds odd, but is very true.
as much effort to defining the levels where you might be wrong
if for no other reason than to restore a sence or balance and inner
the better view has been proven wrong.
market scenario. It can be strongly stated that there is no doubt in
battle, it is hard to be objective about a new market direction once
those who do not understand why there should be an alternative
simply shows we are trying to think ahead, knowing that when in
alternative at all. If you fall into this group, you might be among
not a hedge to a favored opinion. Preparing an alternative scenario
before we move on: Some people will not tolerate seeing the word
you are right. Within this framework the alternative game plan is
Just one last addition to the too-many-alternatives argument
different problem. Far too often analytic services read ' I f not this,
had first considered to be a much lower probability.
But i f there is no favored outlook to begin with, that is an entirely
prepared in advance to explain what might be developing that you
Just step back. Reassess your position. Then move forward again.
that you have stepped off track, then there should be a scenario
you or themselves as their higher-probability scenario was wrong.
market exceed or penetrate the marker you have defined as proof
would do xyz,' then they are not being intellectually honest with
level that identifies where you believe you are wrong. Should the
'You see, I was right all along because we said the market might
target level. That outlook should always be stated with a market
If analysts ever use their alternative market scenario to state
your mind that the market will move in a certain way to a certain
applied.
will be wrong. The analyst who does not have an obligation to state
It warns me that something is lacking in the methodology being
do-or-die level to show where the analyst believes his favored outlook
then that,' and they use this sentence construction all the time.
The problem with the alternative view is when you are not given a
whole. However, the shorter the time horizon, the faster you can
be more or less, but risk exposure to capital must be very precise.
are just the smallest building blocks of the whole and must fit the
made an error on the risk portion of the equation. The reward might
will go through how larger patterns subdivide. Short time horizons
prefer to be wrong on the reward aspect of the ratio than to have
wave count is usually expressed by traders who are wave deaf We
should be and further defines our risk-to-reward ratio. I would much
This objection about picking the best time horizon to support your
Knowing where we are wrong will dictate where our stop placement
can find any pattern you want to prove your market opinion.' Sigh.
to her longevity than will be the accuracy of her market forecasts.
was, ' I f you scan different time horizons for the same market, you
trader must have about where she may be wrong is more important
The last common objection mentioned at the start of the chapter
where he is wrong is not trading that market. The accuracy that a
focus only on the larger pattern. The latter group often gets caught
slippage exposure will be horrendous. Therefore, you must devote
within a larger pattern will be able to react faster than traders who
stop below the third swing in a triangle that we call 'wave c.' The
larger game plan. Traders who focus on the smallest building blocks
pretation is poor money management. As an example, never place a
see when a market is beginning to develop a move not within your
Placing stops near market levels that negate a favored wave inter-
260
• CALCULATING MARKET PRICE OBJECTIVES
Using Oscillators with the Elliott Wave Principle • 2 6 1
wave patterns w h e n the chapter concludes.
we can address this topic fairly, we need to discuss the basic patterns
w i t h m u c h more i n f o r m a t i o n than a cataloged listing of 13 E l l i o t t
focused on the long-term pattern w i l l be caught more often. Before
the new count up for a trade. So, right or w r o n g , we should end up
h o r i z o n to fit their longer h o r i z o n o p i n i o n . Traders and analysts
to reassess your o r i g i n a l game p l a n and understand w h e n to pick
i n an u p d r a f t or d o w n d r a f t because they tend to force the short
of the E l l i o t t Wave Principle.
Chapter 1, Figure 1.4 showed H o n g Kong's H a n g Seng Index on June 3, 1998, and the c o m m e n t was made at that t i m e w h e n
Impulsive Waves
the chapter was w r i t t e n that the H a n g Seng market d i d not have a b o t t o m i n the daily chart. L o o k i n g back n o w o n July 24, the mar-
G a n n confluence zone.
a straight-line illustration into a H i g h - L o w - C l o s e ( H L C ) bar chart
declined to 7,351, w h i c h is a major Fibonacci support cluster and
or line-on-close charts because many people find it hard to translate
ket low i n that chart was 8,351 o n June 2. T h e market has since
Let's try something different. W e w i l l not be discussing stick charts
over to start this t r a d i n g example, it gave the analysis and t i m e tar-
a trader! So no line charts this t i m e .
edition, I w i l l have to transfer one paragraph f r o m the first edition
that it is the market action above and below the close that can k i l l
As the G a n n chapter has been entirely replaced f r o m the first
i n a real market environment. It is also more i m p o r t a n t to realize
I n a d d i t i o n , it is really b o r i n g for the reader to have to wade
gets for the market interval that follows.
t h r o u g h pattern after pattern w h e n there is no real context i n w h i c h
that warn the current correction may not be over.'
have used every single m e t h o d o u t l i n e d i n this book i n order to try
and price confluence point, and we will soon turn to other methods
transmitted live around the world? N o t really. But be assured that I
agree that this is still a fairly respectable reaction to a Gann time
have global distribution. Is that any different f r o m a market report
DJIA from the high is only a 5.5 percent decline. But you might
I have a chance of being dead w r o n g i n a book that is expected to
current low as this is written to 8,855.7 A 512.1 point decline in the
not want to spend the t i m e w r i t i n g one. So let's do this in real-time.
objective, and there has been a decline from a high of 9,367.8 to a
to read a descriptive list about E l l i o t t patterns, and I certainly do
'A market top occurred July 20, 1998, near a time confluence
to see h o w all these patterns may flow together. You do not want
and find the starting p o i n t for a real-time scenario that w i l l give me the best chance possible to start o f f on the right side of the market.
So that is the market we w i l l track.
E l l i o t t Wave Principle w o u l d become completely lost i f they were
I f I step o f f track, it w i l l be extremely informative to show you how
the DIJA because I a m concerned that traders u n f a m i l i a r w i t h the
real market environment that w i l l become a time-sequenced event.
track a real-time scenario rather t h a n our futures market. I chose
needed. T h i s w i l l help you understand how they can be used i n a
T h e S & P traders m i g h t be frustrated that I w i l l use the D J I A to
W e w i l l apply all the techniques previously described as they are
262
• CALCULATING MARKET PRICE OBJECTIVES
Using Oscillators with the Elliott Wave Principle • 263
the real-time journal.
time on the S&P itself at some point within this chapter. But the
Elliott Wave Principle: a five-wave pattern. Then we can begin
always be used for the 24hr period.) I promise to spend extensive
First, we need to discuss the most basic building block within the
Edition note: This is no longer a concern or factor. The data must
S&P Globex problem that will distort even the simplest pattern.
more complex than it should be for someone less experienced. (2nd
to use as a start. However, I'll stay with the DJIA and avoid the
now in the S&P 500 futures market. It would make a difficult topic
is much closer with a high at 1,199.40 and shows a better pattern
to see the Globex night session data that we have to contend with
five-wave pattern. Figure 10.2 is a one-minute bar chart of the Cash
wave structure against the SP500.)
waves for ever more (at least I warned you!), begins with a simple
the Toronto TSE is a good alternative Index to track for cleaner
The world of R.N. Elliott, and your lifetime sentence to count
DJIA always creates a cleaner wave pattern. (2nd edition note: Now
patterns, then you should not apply them to technical indicators
excerpt is from that live interview:
tern purposes. In my opinion if you cannot use tick charts for wave
KWHY-TV being interviewed by Richard Saxton. The following
bar chart. As mentioned earlier, don't use tick charts for wave pat-
Friday, April 17, 1998, I was in the Los Angeles television studio
S&P 500 Index so it cannot be subdivided any further into a smaller
I had created a problem for myself in a public broadcast. On
either. Tick charts are bars where a fixed number of trades define a Richard: Are you looking toward Dow 10,000, as most analysts are,
bar. As an example, a 25-tick bar chart will have no time axis. While
now that the 9,000 level is behind us? Connie: No, 9,000 is of no particular significance in the methods I use. What will be far more important is how the DJIA reacts to 9,339. Both Gann and Fibonacci work suggest this is a critical
Aerodynamic Investments Inc. IMPULSIVE- Simple 5 Wave Pattern
area. If we can pass through 9,339 without a hitch, we will reach
Find
1. Wave 2 cannot retrace the start or origin of wave 1
10,000 fairly easily. But I do not think the market will be able to break through 9,339 without some sort of [a] reaction.
2. Wave 3 cannot be the shortest impulse wave (Does not mean wave 3 has to be the longest.)
So, unwittingly on my part, DJIA 9,339 became public record
3. Wave 4 cannot overlap the termination of wave 1
and was given much more attention by some than was ever intended. I am now in a bind as the DJIA closed four days ago at 9,337.9. I did not even know that was the close until e-mail messages flooded my inbox. In truth, I think the target was missed as the high in place is 9,367.8. The September S&P target
Aero Home Page
Services
Learn More About Gann/Elliott
E-Mail
Figure 10.2
264 • CALCULATING MARKET PRICE OBJECTIVES
Using Oscillators with the Elliott Wave Principle • 265
action forms the second correction that becomes the fourth wave:
recommended for use as diagnostic tools. We will have to come back
Then you have traders who want to take some profits, and this
using tick charts is not recommended, Point-and-Figure charts are to this later so that we do not stray too far off our present discussion. The expansion cycle of a market unfolds in a pattern of five
Rule 3: Wave 4 cannot overlap the end or termination of wave 1.
next move. So, sure enough, after some profit taking, they step into
this causes the first correction we call 'wave 2':
sidelines of the market waiting for a chance to participate in the
assumes the prior trend is still the dominant market direction, and
that developed in the third wave position have been sitting on the
larger move. The first leg is called 'wave 1.' The rest of the crowd
is a guideline. Meanwhile the traders that missed the strong rally
ers who move the market in what will later be seen as the start of a
The Wave Principle only has three rules. Easy. Everything else
waves. A new trend begins with a few well-informed or lucky trad-
the market and buy in unison, forming a fifth wave up in the direc-
itself once again.
the normal character for a market in a very short horizon chart will
three-wave movement down. The five-wave cycle can then repeat
market after years of sitting before a computer screen. Whatever is
little more. Now the whole pattern is ready to be corrected with a
the first wave. That comes from knowing the personality of your
is happening in this market and their orders press the trend just a
kets, such as currency markets, characteristically retrace most of
left to buy? Right. The retail traders who just discovered something
In Figure 10.2, wave 2 down retraces part of wave 1. Some mar-
tion of the larger trend now in force. In a very short horizon, who is
Rule 1: Wave 2 may not break below the origin of wave 1.
someday be seen in that market s longer horizon charts. From the end of wave 2 a strong rally then develops because the
Waves 1, 3, and 5 are called impulsive waves as they are clearly dominant and define the direction of the larger trend. Waves 2 and
struction of the different patterns. Some quote systems allow you
third wave is most often the strongest:
feel the rhythm unfolding in a market before diving into the con-
on the same side of the market for one reason or another so the
makes their head spin. It is far more important to understand and
they realize a new trend is unfolding. Everyone is entering orders
they are built upon blocks offivesand threes and threes andfivesit
same time as the majority are trying to establish a position when
fanatical about the internal construction of these waves and how
now squeezed out, and they have to unwind their positions at the
4 are the corrective phases of the five-wave cycle. People get so
sleepy traders who assumed the former trend was still in force are
to compress the x axis. By compressing the scale, you force yourself
ing every little squiggle wiggle in the chart.
compared to waves 1 and 5.
do within any time horizon as it is too easy to get caught up count-
does not have to be the longest; it just cannot be the shortest when
you cannot see the finer details. This is something I will frequently
This rule actually leads to a common misunderstanding. Wave 3
to study the rhythm and proportions unfolding in the market because
Rule 2: Wave 3 cannot be the shortest.
266
• CALCULATING MARKET PRICE OBJECTIVES
Figure 10.3 Aerodynamic Investments Inc., © 1996-2011, www.aeroinvest.com Source: TradeStation. © TradeStation Technologies
Using Oscillators with the Elliott Wave Principle • 267
Figure 10.4 Aerodynamic Investments Inc., © 1996-2011, www.aeroinvest.com Source: TradeStation. © TradeStation Technologies
the right side of the market.
a 40-minute DMK/$ chart on the far right. Your first task is
balance and proportions, but you didn't need them as you were on
the weekly Hang Seng Index, the weekly NASDAQ Index, and
another major sell signal. We have guidelines to help us determine
Figure 10.3 displays three markets with a compressed x axis:
to please mark each chart so that balanced five-wave patterns emerge. Look at the Hang Seng chart first on the left. Would you mark
Next, review the weekly NASDAQ chart in the middle of Figure 10.4. I deliberately had to scroll this chart back so that you could not see the present data. The real chart is a five-wave advance.
high is much more forceful than the first.
warned you that the rally into the 1,200 level in late March was
wave structures in this chart. The secondfive-wavepattern into the
then followed. You had the right rhythm. Your indicators would have
proportions within their relationships. There are two pairs of five-
you would be on the right side of the market because a large rally
as they have similar character. As a group, they also have balanced
you thought a five-wave structure ended at the low in early 1998,
be able to see that some of these waves should be grouped together
Guess what—it does not matter which one you prefer because if
NASDAQ chart in Figure 10.3, you have a problem. You need to
Or would you feel five waves down ended at the low in early 1998?
However, if you found afive-wavepattern from bottom to top in the
five waves down from the top of the chart data to the very bottom?
268 • CALCULATING MARKET PRICE OBJECTIVES
Using Oscillators with the Elliott Wave Principle • 269
interesting.
doing so. The data has been compressed precisely for this reason so
in a weekly DJIA chart, our real-time scenario could get pretty
immediately start to subdivide it as we cannot stop ourselves from
projections are correct and we now have afive-wavepattern visible
the fifth wave in the first five-wave pattern in this chart. You'll
coincides with the end of a five-wave pattern. If my Gann time
Textbook Elliott wave practitioners may become troubled with
next corner.
is that you did not need that level of expertise to make money from the pattern.
both directions correctly begin with an immediate left turn at the
catalog and connect perfect patterns endlessly together. The point
McDonald's restaurant.' The scale is entirely different, although
into trouble toiling over every little squiggle with an objective to
'Turn left at the next corner and go straight until you reach the next
is why some of the strongest textbook Elliotticians get themselves
and go straight until you reach California' as opposed to stating,
moment as I know I jumped ahead of you. Stay with me.) But this
were giving us directions by stating, 'Turn left at the next corner
new, note that I'll explain the 'previous fourth' comment in a
larger road map that we can turn to as a guide. It is as if someone
a major rally. So stick with a view of the total picture. (If you are
from the Elliott Wave Principle that we gain a sense that there is a
right to the previous fourth wave target zone and then produced
cators. It is only after we have added new information obtained
When the first five-wave pattern concluded, the market declined
those market corrections from just observing divergence in the indi-
subdivide into 5s or 3s. Who cares about the internals in this chart?
In Figure 10.1 we were unable to determine the magnitude of
that you are unable to see clearly if any structure has internals that
The last market in Figure 10.4 is a 40-minute DMK/$ chart on
ends. The novice will soon understand why.)
the range of the previous fourth wave. That is extremely helpful
be scrambling to again reverse or step aside where this chart data
a five-wave pattern will be to this zone, which is anywhere within
price low in this chart. (The more experienced wave counter would
ted line marking the start and end of wave 4. The target following
both would have been looking for a rally in the dollar from the
10.2. Where wave 3 topped and wave 4 bottomed, there is a dot-
Elliott wave practitioner will be on the right side of the market as
zone the vicinity of the previous fourth wave. Take a look at Figure
the end of this chapter. However, both the novice and the advanced
market corrects a completed five-wave cycle. We call this target
is part of another pattern that the novice will easily recognize by
know to anticipate where the next target will be located when the
advanced Elliott wave analyst will recognize that the last leg down
staring at these redundant patterns so often in all time frames, we
OK for the information you have been given so far. The more
in the opposite direction to correct the entirefive-wavecycle. After
data in this chart as afive-wavepattern from top to bottom. That's
we have a five-wave pattern in place, there is going to be a move
the far right. If you are just learning, you will likely mark the entire
I had to jump ahead, so let's back up and fill in the gap. Once
We said markets move in five-wave patterns in the direction of a
of the correction that will follow from any indicator signal that
Let's go back to the NASDAQ chart in the middle of Figure 10.4.
to know in long-horizon charts as we now have a sense of the size
270
• CALCULATING MARKET PRICE OBJECTIVES
larger trend. This chart shows two five-wave advances in a clearly
Using Oscillators with the Elliott Wave Principle
• 271
Extending 5 Wave Pattern
defined bullish trend. The total picture is not a complete five-wave pattern, and the second pattern of five waves is clearly stronger than the first. What is happening? This market is extending, which means that once another five-wave pattern forms over the first two in this chart, we will then have a larger five-wave pattern. The reason is that the impulsive waves in positions 1, 3, and 5 are themselves constructed from smaller five-wave units. So three sets of five-wave patterns that still abide by the three rules stated for the simplest fivewave pattern in Figure 10.2 will become a larger five-wave pattern. Let's take a closer look at an extending five-wave pattern in Figure 10.5, and then we can start our day-to-day tracking of the DJIA. Take a look at the last five-wave cycle in Figure 10.5. Each wave is labeled I through 5 into the price bottom. The start of the last five-wave cycle is marked wave (4) at the top and the bottom is
Figure 10.5 Aerodynamic Investments Inc., © 1996-2011, www.aeroinvest.com
marked wave (5). As you look up the data and then study the whole
and below it.
price decline is so strong, however, that we already know that it is
bar or swing and counter-balance corrections and rallies above
ture is present as more price bars become available for study. The
edition has been removed. Start i n the middle of the strongest
shorter time horizon so that we can confirm that a five-wave struc-
tion. So the quick reference people found confusing in the first
the completion of wave (3). That is why we drop down to the next
trends in the new Elliott Wave books that just went to produc-
be subdivided into five distinct waves from the top of wave (2) to
I go great length to describe how to start in the middle of larger
ful that it is nearly a straight line, and it is hard to see that it can
in the middle of the chart. I know, this one sounds screwy at first.
within a larger five-wave cycle. Wave (3) in the middle is so force-
A market that is extending is much easier to count if you start
structure, you can see that this last move down is the fifth wave
pattern, and we may have to use this technique at some point. It is
smaller five-wave patterns that are complete.
midpoint. Our DJIA scenario could create a lengthy extending
this third wave down, it must have internal building blocks that form
have to make an adjustment and use the stronger move as the new
through ignorance. Regardless of the time horizon we use to view
stronger than the first that defined our midpoint? We would then
ent time horizons to pick a pattern that best fits your view is wrong
What i f the next five-wave pattern that the market forms is even
a third-wave decline. The assertion that you should look at differ-
272 • CALCULATING MARKET PRICE OBJECTIVES
Using Oscillators with the Elliott Wave Principle • 273
always balance, rhythm, and proportion that should be given the greatest thought for any chart. There are three asterisks marked at 50 percent Fibonacci retracement levels in Figure 10.5. The first asterisk is the fourth wave up within thefirstfive-wavepattern down that completes wave (1). The second asterisk marks a 50 percent retracement for all of wave (3) and becomes the area of resistance for wave (4) up. Once the entire extendingfive-wavepattern is complete, the first bounce up is to a 50 percent retracement relative to the lastfive-wavecycle in the sequence or the fifth wave. You would now want to know the Fibonacci retracement ratios relative to the entire decline as well. These repeating ratios show the market is extending, but not expanding, its scale. Proportions remain about the same. If the decline were accelerating, the corrective rallies would begin to fall short of the
Figure 10.6
Source: TradeStation. © TradeStation Technologies
50 percent retracement levels marked in this chart. We covered considered so a new chart had to be constructed.
not venture down that same road twice. But the internal Fibonacci
it is not the best interpretation when balance and proportion are
this topic in great detail within the Fibonacci chapter so we will
five-wave patterns that follow one after another. It takes three sets
balance.
subdivided into three individual sets of complete, though smaller,
tant. Figure 10.5 gives us an overall picture that has symmetry and
In Figure 10.5 we saw how a larger five-wave pattern can be
ratios and resulting math grids on the screen are extremely impor-
is true even when our supporting indicators offer major signals to
writing style in an effort to make my task a little easier.
knowledge that the market has reached an important pivot. This
time market patterns that develop. Permit me to switch to a journal
that meet minimum requirements and sit back, content with the
use within the Wave Principle so that we can keep up with the real-
Figure 10.2. However, it is not enough to simply find five waves
building our knowledge about the basic building blocks that we
and they cannot break the first three rules that were defined for
the remainder of this chapter. The challenge will be to continue
of completefive-wavecycles to construct a largerfive-wavepattern,
Let's begin to study the DJIA real-time and track it through
It is now July 24, 1998. In Figure 10.6 we have a seven-minute
complement our view that thefive-wavecycle is now complete. rules and guidelines of the Elliott Wave Principle. First locate the
so that it could be used in my evening market report. However,
In Figure 10.6 the five-wave pattern labeled will meet all the
chart of the Dow Jones Industrial Average. The chart was labeled
274
• C A L C U L A T I N G M A R K E T PRICE OBJECTIVES
Using Oscillators with the Elliott Wave Principle
• 275
i to v for Minute, and finally i to 5 for Minuette.
nity to do so now. The high in this DJIA chart at 9,367.8 was made
Degree uses (1) to (5), Minor Degree uses the numbers I to 5, then
discussed how to label subdivided waves, we can take this opportu-
which follow are the more common ones you will see; Intermediate
pivot levels labeled 1, 2, 3, 4, and 5 in Figure 10.6. As we have not
follow your interpretations.
V. Beneath wave v we put the number 1 to show where we think
Just know you have to be consistent so that others might be able to
that we refer to as one lesser degree should be marked i , i i , i i i , iv, and
faxed, but the method I have been using is not a whole lot better.
confusion with the larger structure unfolding. So the smaller subset
.1 to .5. The dots are very hard to read, particularly if the chart is
be labeled with something other than numbers 1 through 5 to avoid
electronic word processing. But others have used .1 to .5 and then
sions within wave 1 are marked with only an asterisk. They need to
ters. I will use r to 5', then 1' to 5< as the primes easily work within
five-wave decline that forms wave 1 down. The individual subdivi-
short time horizon, it is very easy to run out of conventional charac-
of July 24. Start from the price high in this chart and follow the first
When you subdivide wave structure all the way down to a very
on July 20, 1998, and the chart shows the correction into the close
on their screen or chart. This interpretation would be the best fit.
sions in the chart next time.
as most people try to find the best fit for only the data that is present
number when we run out of ways to differentiate additional subdivi-
plete? That is how most analysts will end their wave interpretation
the strongest part of any market move. We will use an underlined
an extending five-wave pattern that is now complete. Or is it com-
wave because it is wave i i i within 3. This wave position is always
the price low. That's it. We have just labeled every twist and turn in
wave iii) and the end of wave i i i is what we call the third-of-third
then we have an extending pattern that creates wave 5 down into
with asterisks between the end of wave i i (better said, the start of
if you need a quick recap. The correction becomes wave 4 up, and
small five-wave pattern. The small five-wave cycle that is marked
covered this phrase eariier, and you may want to turn to Figure 10.2
run out of labels, as wave iii itself subdivides further into another
then see a correction unfold back to the previous fourth wave. We
wave 2 up that leads to wave (3) down. Within wave (3) we again
We left off where wave 3 down was complete in Figure 10.6. We
the first five-wave cycle ended. Then there is a correction marked
the Composite Index is diverging with RSI.
R.N. Elliott's, although we keep running out of ways to subdivide
wave cycle even though there is a buy signal at the price low where
as one lesser degree. Bob Prechter's labels make more sense than
However, I do not think that this chart represents a completed five-
As the patterns subdivide into smaller portions, we refer to each
bined. D i d I make an error in my identification of wave 3 itself?
1 to 5 within circles are used for Primary Degree. The degrees
distance of the price decline that unfolded i n waves 1 and 3 com-
cycle Degree, I to V denotes Cycle Degree, and then numbers
is that the distance traveled in wave 5 is greater than the total
Elliott Wave International. Waves (I) to (V) are reserved for Super-
others will view their analytic task as complete. The first problem
in the book Elliott Wave Principle: Key To Market Behavior from
We must now take the chart in Figure 10.6 past the point where
intraday data. The following labels are described in more detail
276
• CALCULATING MARKET TRICE OBJECTIVES
Using Oscillators with the Elliott Wave Principle
• 277
invaluable technical i n f o r m a t i o n for constructing wave counts.)
plete structure w i t h really b e a u t i f u l proportions and ratios w i t h i n
to go back to pick this i n f o r m a t i o n up first as the RSI reversals offer
N o , I do not believe so. T h e reason is that wave 3 is clearly a c o m -
the true t i m i n g signal.
these market turns.
deliberately set up as an early w a r n i n g . T h e crossover i n the RSI is
is divergence between the RSI and Composite Index to w a r n us of
the crossover i n the Composite Index that occurs earlier has been
nificant pivots that mark the end of waves 1, 2, 3, 4, and 5. There
or below the longer m o v i n g average on the oscillator. I n this chart
sell signals w i t h i n m y indicators to see i f they align w i t h the sig-
near the intersection where the short m o v i n g average crosses above
market's data. I t h e n need to study the c o m p l e m e n t a r y buy and
tor pattern relative to the m o v i n g averages is a signal that coincides
wave 3 d o w n . There must be something else going on w i t h i n this
4 up. W e previously demonstrated that the most significant oscilla-
can see the tremendous symmetry w i t h i n the segment defined as
Finally, the buy signal of greatest significance produces wave
this u n i t . I have added some visual guides o n the chart so that you
extended, b u t it d i d not rescale its symmetrical proportions. T h e
without
Fibonacci price projections. I n Figure 10.5 the five-wave pattern
tant relationship between oscillator versus price movement. You are
W h e n this indicator pattern develops, I k n o w to adjust all m y future
very weak. W e had charts i n prior chapters to h i g h l i g h t this impor-
eruption to a larger mathematical grid for future internal structure.
only m o v i n g a relatively small distance upward. T h e bounces are
wave position or a market c o i l i n g i n preparation for an expansion
indicator. However, the Composite Index advances while prices are
Composite Index is a characteristic pattern f o u n d w i t h i n a t h i r d
bounce each t i m e the Composite Index tests the trend line for that
W h a t d i d we just find? T h e channel that you see developing i n the
the July 22 date correctly warns that the market w i l l attempt to
evidence that we have been t r y i n g to find. ' I t is?' you may ask.
w i t h each new oscillator peak that develops. T h e divergence near
horizontal channel o f resistance. T h i s is the significant piece of
the RSI are developing negative reversal price projection signals
to a series of four S labels denoting sell signals that align w i t h i n a
early. It is not. L o o k more closely. B o t h the Composite Index and
At the top of wave 4 i n Figure 10.6 we have a sell signal that leads
T h e buy signal that forms into the b o t t o m of wave 3 appears
on the wrong side of the market if the oscillators prices
can move
retracements a l l challenged a 50 percent ratio of a p o r t i o n o f its
following.
the buy signal leads to an indicator pop into the last bar o n the right
the RSI signals that I a m referring to very difficult. You m i g h t want
quake occurs. N o w follow the indicators past this channel where
RSI. ( I f you skipped the RSI price projection chapter, you w i l l find
as a seismograph recording of a m i n o r tremor before a larger earth-
the RSI that meets the amplitude requirements we defined for the
that f o r m a horizontal channel i n the Composite Index is the same
up w i t h i n wave 3 d o w n . There is also a negative reversal signal i n
wave structure. T h i s indicator pattern of consecutive sell signals
pared to prior oscillator peaks. T h i s signal marks the end of wave iv
have a five-wave pattern that w i l l likely develop into a larger five-
where the oscillator makes a new h i g h at a lower price level c o m -
larger movement, and the decline was orderly. I n Figure 10.6 we
W e t h e n see the Composite Index forms a negative reversal
278 • CALCULATING MARKET PRICE OBJECTIVES
Using Oscillators with the Elliott Wave Principle • 279
So we now know the market is going down further after it attempts
Lane. The indicator screams SELL.
travel any displacement above or below zero that it may want to go.
of this chart. The term 'an indicator pop' was coined by George The oscillators are overbought, and the two indicators have diverging peaks within a close proximity to one another. This is exceedingly bearish. The chapter on the RSI revealed that peaks
a back-and-fill pattern from nearby levels. The next step is that we have to think of a wave count that fits this additional evidence. We also have to factor in the negative reversal signal present in
of technical evidence within our current puzzle.
with the most recent data. Both the RSI and Composite Index are
finding a revised wave count that might compliment all the pieces
cal information available from the final indicator positions aligned
to a new proportion with the next move. Here is how I go about
tors display this strong signal. However, this is not the only techni-
series of horizontal sell signals that warn us the market will rescale
should be considered very strong signals. In this chart both indica-
the oscillators and work with the information we gained from the
that form divergences within a spread less than nine periods apart
overbought at oscillator highs that form at lower price levels com-
The chart in Figure 10.7 is the exact same chart we saw in decline from the high, rotated that segment of data 180 degrees,
reversal signal. Reversal signals develop only in bear markets.
Figure 10.6, except that I have made a copy of the first part of the
pared to other oscillator peaks which we now know is a negative
extend into a larger structure. We knew that from the oscillator
It is making a peak at a level (compared to the indicator's y axis)
tern that displayed symmetry because it in fact was still trying to
Composite Index peaking in the most current data of Figure 10.6.
to see that the market had not formed a completed five-wave pat-
few bars of Figure 10.6. How do we know this? Look closely at the
and pasted them to the last bar in Figure 10.6, it was very easy
will not make a major decline from the market levels in the last
When I copied the beginning cycles within our five-wave pattern
mation warning us not to 'bet the farm' immediately. The market
ing data. Why? We know markets create symmetrical patterns.
market will decline further. But it is also giving us more subtle infor-
and then pasted the copied segment onto the back end of our exist-
All right then, we know the chart is clearly telling us that the
that is higher than any other peak that this indicator created within this chart. Recall a prior discussion that stated, when an oscillator successfully creates a new high or low by breaking through a former horizontal range of resistance support for that indicator, the market will have sufficient punch to come back and test the former indica-
2 1
1
li
1
Rotated 180 degfrees
j
Copied
tor range after the new indicator extreme has been alleviated. In my experience the oscillator nearly always returns to test the former 3
range of resistance support for that indicator. That is why I use the Composite Index, because it has not been normalized and can
5
Figure 10.7
280 • CALCULATING MARKET PRICE OBJECTIVES
Using Oscillators with the Elliott Wave Principle • 2 8 1
warning us that an expansion may form.
but not the exact puzzle piece that we need.
tion now because of the series of horizontal sell signals in the chart
tern is what we need. OK, so the pattern in Figure 10.7 is helpful
choppy pattern into its conclusion. We cannot make that assump-
not be an immediate breakdown. A down up and then down pat-
start might be how the market's move ends by repeating a slow
that the oscillator position into a new high is telling us there will
will look like as it nears a resolution. In other words, a slow choppy
in Figure 10.7 helps us see an extending pattern, but do not forget
you to identify price targets and also what the character of the move
iii within wave 3 down. The graphically manipulated interpretation
between the first part of a market's move and the latter will enable
this. What had been labeled wave v of 5 down is now marked wave
the Wave Principle. Knowing that proportional balances may occur
rotated. Use the thumbnail method if you have difficulties seeing
will be able to do in your mind very quickly after you start to utilize
are still present, though upside down, because the data has been
onstrated in Figure 10.7 before we move on. This is something you
Figure 10.7, you will see that the original labels from Figure 10.6
One last note about the graphic cut-and-paste technique dem-
signals. If you look at the pasted segment added to the right side of
which we will cover later in this chapter.
nicely.
are price projections within corrective patterns and from triangles,
versa. So the guideline of alternation fits our present scenario really
will produce the correct objectives. What we have yet to discuss
you have a simple wave 2, you can expect a tricky wave 4 and vice
you need to learn as the prior techniques in the Fibonacci chapter
the amount of time each takes to complete their corrections. If
in the Fibonacci chapter. So there are very few new relationships
at Figure 10.2 again, you will see that waves 2 and 4 alternate in
tionship by applying the techniques we discussed at great length
of patterns that form within wave positions 2 and 4. If you look
because you already know how to calculate every Fibonacci rela-
the Wave Principle has a guideline for alternating the character
will not spend a lot of time on price projections for Elliott waves
into a fifth wave. I like that idea, especially when we consider that
Fibonacci relationship such as 1.618 relative to the first wave. I
down up pattern the oscillator wants, and then we can break down
5 is frequently equal in size to the first wave or will develop a
is incomplete? A more complex correction would incorporate the
It is important to know for price projection purposes that wave
Take a closer look at wave iv up in Figure 10.7. What if wave iv
longer time horizons would do for us now.
projections, and Gann. Gann will have the highest weighting.
for a pattern that best fits our market view. Ha! Let's see what the
tive reversal signal we have in place in Figure 10.6, Fibonacci swing
might see us change our time horizon and think we are looking
To determine the target for the price low, I will combine the nega-
of graphing a future path as shown in Figure 10.7. Someone else
a new low, and then we can see what else the market will give us.
ing up to a longer time horizon, rather than going to the trouble
rent fourth wave is incomplete, and then we can expect a decline to
remove some of our subdivisions in the seven-minute chart by mov-
and-fill scenario, so let's go with the following game plan. The cur-
Someone might think it would have been much easier to just
There are no cycle lows on the horizon to challenge the back-
282
•
CALCULATING MARKET PRICE OBJECTIVES
DJIY-15 min
Using Oscillators with the Elliott Wave Principle • 2 8 3
DJIY-60 mm 07/24/98 0=89373 +44
07/24/9S C=89399 +52 93500
1-93000 92500 92000 -91500 -91000 -90500 -90000 -89500 -89000 -88500 -88000 10:45
'7/17
' '7/20 '
'7/21
7/22
7/23
eh'
' '6/12' ' ' ' 6 / 1 8 ' ' '
'6^4'
' '6/30'
'
W
''7/13''
'7/17'
'
'7/23
Figure 10.9 Source: TradeStation. © TradeStation Technologies
it is the most direct and displays fewer back-and-fill retracements.
7/24
However, you might notice the pattern of sharp breaks throughout Figure 10.8 Source: TradeStation. © TradeStation Technologies
wave 5 in this interpretation and the increasing volatility. These visual clues in Figure 10.8 are harder to read than the indicator
Figure 10.8 is the exact data that was displayed in the sevenminute chart now plotted in a 15-minute bar chart. The labels
signals that we discussed in Figure 10.6, but the wave count is clearly intact and defines a five-wave structure in this time horizon.
might have been made from the price high to the current low in
down in this interpretation still appears to be the strongest because
offered in Figure 10.9. While we can still see that afive-wavepattern
to plot the X axis. The wave structure that unfolds within wave 3
ute bar chart in Figure 10.6 remain visible in the 60-minute chart
not touched this chart other than to change the time period used
waves in a 60-minute chart? The labels applied to the seven-min-
labels in place that were added to the seven-minute chart. I have
pressed into a 15-minute chart, will it continue to look like five
time horizons in TradeStation, that system will try to keep the
If an extending five-wave pattern can hold together when com-
in this chart are those we used in Figure 10.6. When you change
284
• C A L C U L A T I N G M A R K E T PRICE OBJECTIVES
Using Oscillators with the Elliott Wave Principle • 2 8 5
someday by an unexpected surprise.
the completion of a five-wave pattern and the decline unfolding now
beyond the boundaries of your o w n market, you may be clobbered
was advancing. T h i s prior work shows that we were working toward
and do not review the global picture to see what m i g h t be brewing
still has the o l d labels displayed f r o m the analysis w h e n the market
also happen. I f you get too complacent i n the shortest t i m e horizons
what has just come into view. T h e rally that developed into the h i g h
horizon does not w a r n you that a change is brewing? Yes, this can
this longer time frame, we must consider w i t h m u c h greater interest
completed five-wave rally displayed i n the 60-minute chart. O n the
Figure 10.9 fits into an even bigger picture for this market.
chart w i t h i n Figure 10.10 shows the current decline relative to the
it is becoming clear we must k n o w how this current five-wave rally i n
into a rally that has been f o r m i n g an extending pattern. T h e top right
we now have to step back and look at an even longer time horizon as
Figure 10.9, we want to see how this latest five-wave structure fits
wave w o u l d have been the most likely target for a m i n o r pullback. So
N o w that we have a completed five-wave rally present i n
has already exceeded the previous fourth wave. T h e previous fourth
corresponds to the five-wave rally highlighted i n a box i n the daily
it. W h i l e I work f r o m the smallest b u i l d i n g blocks possible and work
T h e second arrow to the right i n this weekly chart marks the low that
Elliott Wave Principle differently f r o m the way Bob Prechter applies
bottom right, using a compressed x axis, is the weekly D J I A chart.
W h a t you are seeing demonstrated is one of the ways I apply the
upward into a longer time h o r i z o n , he starts w i t h an extremely l o n g h o r i z o n interpretation and then looks for substructure w i t h i n progressively shorter t i m e frames to c o m p l e m e n t his longer h o r i z o n view. I take the approach o f b u i l d i n g f r o m the small and w o r k i n g upward because m y risk management dictates accuracy i n the very short t i m e horizons. I can be less accurate i n the really big picture because I a m always revising the underlying components w i t h i n a market as they develop. I have little interest i n the really big picture just as long as I ' m trading f r o m the right side over the next few hours or days. Prechter, o n the other hand, cannot change his view every few minutes or days because his greater interest is i n the really big picture that may define major pivots w i t h i n terms of decades. So this is one way that we apply the Wave Principle differently. I spend extensive t i m e reviewing t h e bigger picture w h e n m y short-horizon b u i l d i n g blocks begin to step o f f track or they are out of sync and challenge the longer-horizon view. But what i f the short
Figure 10.10 Source: TradeStation. © TradeStation Technologies
286 • CALCULATING MARKET PRICE OBJECTIVES
Using Oscillators with the Elliott Wave Principle • 287
the market ultimately develops. Good night.
ject of a great debate. The debate concerned a triangle pattern, and
developed several pages ago in Figure 10.6. Let's see what pattern
that were used in the past are still present. This pattern was the sub-
has been slipping nicely within our down up scenario that was
10.10. You will see that all the former signals and wave interpretations
but not tonight. It is now 4 a.m. on July 25, and the Globex session
at the window displaying the daily S&P 500 futures data in Figure
I will walk through the approach I use for longer horizon work,
ing us that an expansion to its present scale could develop. Now look
these indices and the underlying S&P 500 sectors themselves. So
from the present high. No wonder our seven-minute chart is warn-
be the next step to study our weekly and monthly bar charts for
the market showed respect to this same mathematical grid derived
need to show you how I analyze long-horizon charts as that would
the remaining arrows bring to your attention prior pivot levels where
I've seen enough to know this could get ugly. To go further, I will
test that we discussed in the Fibonacci chapter. It does pass this test as
is on my screen. I do not need all the answers to trade tomorrow.
that formed in October 1997 would meet the internal relationship
time target. I do not care what this current Elliott wave pattern
and what is being tested is to see if the current high and major low
wave pattern is present that developed into a major Gann price and
data. The weekly chart, bottom right, has a Fibonacci grid added,
we will come back to this chart if the DJIA makes its own triangle. We will discuss corrective patterns later. The price high in this chart may in fact be the resolution of a largefive-wavepattern that began in October 1997. The guideline
Corrective Waves
to bindfive-wavecycles together.
going to lead to a fairly healthy correction with all the evidence
Elliott wave vocabulary to include the corrective patterns that serve
Composite Index when it diverged dramatically with the RSI is
more useful than letter shapes and direction, we need to build our
fourth wave is admittedly very large. The sell signal present in the
futures market. But to discuss these down up patterns in terms
have arrived already within this target zone. But the range of the
in Figure 10.6 is what has developed in the DJIA and S&P 500
will target 'the vicinity of the previous fourth wave.' Hello. We
The down up market action our indicators warned us could occur
that the Wave Principle provides us with is that afive-wavepattern
building against the DJIA. The indicators in the daily S&P chart
While most people will develop an ability to see a five-wave
more than one individual building block is generally required to
may not be correct. All 1 really care about is that a possible Elliott
terns should look like in a real market environment. In addition,
the high? That interpretation from the October 1997 low may or
these patterns should look like, but it is not as clear what the pat-
tional five-wave pattern marked from the October 1997 low into
cycle. It is clear what a textbook line drawing representation of
averages in this chart will be violated. Also what about that addi-
it difficult to see the patterns that correct a completed five-wave
we therefore know the long-term trend lines marked by moving
structure within price data fairly rapidly, some people may find
show that a bottom is not in place in this time horizon either, and
288 • CALCULATING MARKET PRICE OBJECTIVES
complete a market correction, and there is some confusion about
Using Oscillators with the Elliott Wave Principle • 289
•the following shapo is folded into a cube shape, which picture best represents the
the connector between these individual patterns, the dreaded X wave. I can understand why the confusion exists because I too had to struggle with corrections until Dave Allman taught me how to feel the pattern for an individual structure. Feel a pattern? Yes. Allman walks you through hundreds of sequential charts to help you gain a feel of a particular pattern in his Advanced Workshops. Although I cannot do that here, I can provide some analogies that
• 2
Which of the designs best completes the following sequence?
may help you. Before we look at our first corrective pattern, you need to know
Q
the level of aptitude you presently have for working with graphic patterns that change their orientation. I am dyslexic in that tables of numbers and letters are difficult for me to read. Tabular characters tend to transpose their positions in my mind's eye. I am not
• 3
II ^ [in— nil 1 1/ Ml
L-iiiii
II
n—
III! J
= liiiiTI » 4
0
Which of the figures below the line of drawings best completes the senes?
r,
an options trader for that reason because the strike price tables and calculations for delta, gamma, and theta in tabular form are a visual nightmare for me. I work with them only on rare occasions
0
D
!/•
Which of the figures below (he drawings best completes the series 9
and with great anguish. But the flip side of this limitation is that I have a photographic memory and a graphic aptitude that more
9
than compensates for the tabular weakness. The point is that you
«
ilk
may not know that graphic puzzles are difficult for you. Corrective patterns are just puzzles once you know their underlying building blocks. Let's try a quick visual test to see where your starting point might be located.
M E N S A International
http://www.m9nsa.org
Figure 10.11
There are four graphic puzzles in Figure 10.11. You should give
are like graphic puzzles, and markets frequently ask us to 'complete
International. Their Web site can be found at www.mensa.org.
similar to a market problem. Elliott wave corrective wave patterns
an introductory sample quiz, considered elementary, by MENSA
why these questions have the solutions they do and why each is very
original guidelines for these questions. These four questions are from
and these questions are from their Mensa Workout page. I'll explain
yourself no more than four minutes to select your answer based on the
290 • C A L C U L A T I N G M A R K E T P R I C E O B J E C T I V E S
Using Oscillators with the Elliott Wave Principle
• 291
a decline, rotated it 180 degrees, then pasted it to the back of the
accounts. That is the fun part!
Figure 10.7 showed one example where I copied the first portion of
when we 'complete the series' correctly, we put money in our trading
relationship that is developing within a market's price structure.
the series.' The difference is that our quizzes are more fun because
rotated 180 degrees with each new frame as the stick is alternating
the markets will create for us. The correct answer is B.
The stick simply shows that the orientation of the entire graphic is
long, and the first question in the quiz is far easier than the puzzles
side opposite the stick decreases by one. The correct answer is C.
tion is the type of puzzle an Elliott wave trader is solving all day
number of lines on the side with the stick will increase by one, the
The reason such a bold analogy can be offered is that the first ques-
Elliott wave application when adding indicator signals. While the
environment about as comfortably as I work with options tables.
While this problem is harder than the first, it is very much like an
will be applying R.N. Elliott's wave patterns in a real-time market
actual data in that chart. Question 3 is the same type of puzzle.
If you had any trouble with the first question in this quiz, you
The second question in the quiz shows how you might be
sides. The correct answer requires a final rotation.
it is the only solution offered that fits in the series.
versus a stock index. Nonetheless, it is the same pattern. In ques-
row has lines radiating into the center. The correct answer is E as
may not be a distinct N when you look at it in the data for Euros
while the second only had five? The first row is solid. The second
terns. For example, the up-down-up or N pattern referred to eadier
the series.' D i d you notice that the first column had six petals,
interference or noise. Market character can impact corrective pat-
Question 4 is another graphic puzzle asking you to 'complete
thrown off balance by different market character that serves as
issue again, and it is no reflection on one's intellect. Please accept
seminar or conference.
bonds are a breeze (ugh). It is a left-brain/right-brain development
am not giving you the answer to this revised question; maybe in a
day long without effort. Even duration and convexity calculations in
relationships have already been defined to extend the series. No, I
your firm who can manipulate delta, gamma, and theta tables all
you would be asked in an actual Mensa test. The mathematical
using this particular instrument. I bet you are the options wizard in
ing for the missing fourth shape?' That would be closer to what
Elliott piano, but most likely you will never learn to play the music
be the next pair of shapes that follow within this series after solv-
you will learn the mechanics of how to press the correct keys on the
easy question. A harder question would have been, 'What would
struggled with all of these questions, please accept that with work
circles. The black and white colors simply alternate. This is an
and you will likely have little trouble. At the other extreme, if you
hexagons would be followed by two exterior hexagons with inner
ready to charge ahead into any corrective pattern at full speed
vice versa. The correct answer is A. Two outer circles with internal
How did you do? I f these questions were easy for you, we are
tion 2 when the outer shape is a hexagon, the center is a circle and
indicators from prior chapters will be applied in a real-time scenario.
the orientation of a pattern to correctly identify the mathematical
that this next section may be harder for you, but stick with it as the
Question 3 is one I really like as you frequently have to change
292 • CALCULATING MARKET PRICE OBJECTIVES
Our indicators told us to look for a price movement that would look like the letter N. This pattern fits into the family of corrective
Using Oscillators with the Elliott Wave Principle * 293
higher than would have been possible from a solid surface. All divers must then abide by the force of gravity.
subdivided into afive-wavecycle.
below the board's horizontal level. The resulting displacement will
which can be in the direction of the larger trend, will not be easily
diving board, but then his weight presses the diving board down
will frequently be obvious is that the b-wave portion of the move,
diver takes the same approach by taking a step to the edge of the
our task much easier. The market action will get choppy, and what
hand, the Expanded Flat pattern is the diver on a springboard. The
will always be hard to count. The process of elimination makes
under your feet before your final jump from the edge. On the other
is that the final move to the edge of the platform or springboard
much higher than the start of your first leap because there is no give
for a moment at the top of Figure 10.12. What is important to know
edge of the platform, then jump as high as possible. You do not get
it. Here is the key to this entire pattern: Forget the 3-3-5 references
is a solid structure. You jump up, land with both your feet on the
that the market action is counter to thefive-wavepattern preceding
left of Figure 10.12 is the diver taking off from the platform, which
movements of the 'dive,' or correction, are labeled a-b-c to identify
into the air. The Flat pattern is the exact same feel. The Flat on the
for a declining market as demonstrated in Figure 10.12. The three
themselves, you saw divers launch themselves from a springboard
pattern can develop in an upward trending market as can develop
ment on television and never watched any of the televised events
board or platform in a weightless environment. Hence, the same
the Summer Olympic Games. Even if you saw only an advertise-
pattern is that the 'diver' can bounce on either side of the diving
you ever watched a springboard diving competition? Think back to
The only difference between the diving analogy and the market
patterns called 'Flats.' They are illustrated in Figure 10.12. Have
ter what kind of three-wave pattern it is. The middle leg within Flat
The spring displacement of the diving board then projects the diver
If it is not afive-wavestructure, it has to be a three. It doesn't mat-
be lower than it was with the approaching step toward the edge.
corrections will be choppy because there is a lot going on. A diver Flat (3-3-6)
Expanded Flat (3-3-5)
will lift only one knee before planting both feet on the edge of the board. Why? Because he is about to change his direction of trend and he needs to stabilize his prior movement before he can be projected in a new direction. That is exactly what the market is doing. Don't get hung up on categorizing all the internal squiggles. If you can sniff out a b-wave, you have it made. The point I am trying to make is that it doesn't matter if the first leg looks like a five because the choppy pattern to the edge of a diving board is far more important. Feel them first in your gut. The problem with representative line
Figure 10.12
294 • CALCULATING MARKET PRICE OBJECTIVES
Using Oscillators with the Elliott Wave Principle • 295
over, look at this page from the next page while holding the page in
wave patterns used without the aid of other indicators are deadly.
rective pattern under the springboard. Can't see it? Flip your book
pattern's action within a market. In my opinion that is why Elliott
illustrated in Figure 10.12 except that Bonds are forming the cor-
charts for these patterns is that you never get a chance to feel the
Bias problem solved.)
type of tension or inner expectation that we might have when we
put it in MS Paint. The image can be flipped and inverted easily.
Flat patterns in Figure 10.13. The indicators contribute the same
front of a light source. (When you trade take a screen capture and
Look what happens when we add an oscillator to the corrective
know how a springboard should ultimately move when it is overextended from a horizontal position.
In the September Bond chart in Figure 10.13 notice the positive reversal marked in the Composite Index that forms in the oscillator
the perfect textbook pattern in the real world.
former trend. The character in this 20-minute chart is exactly what
it had to move well past the start of wave a? No. So don't look for
Flat pattern is completed, the Bond market is slow to resume the
does not move much past the origin of wave a. Did anyone ever say
is meant by 'different character for a market.' Once the Expanded
posite Index forms a double top. Notice that wave b in this example
in Figure 10.13. A very short hme horizon was selected to show what
lows on either side of the asterisk. At the top of wave b the Com-
A 20-minute September Bond futures chart is displayed on the left
end of afive-wavepattern in Figure 10.4. The top of wave c of (4)
Flat pattern itself for the correction is no different from the market
Now you know why the first decline in early 1998 was not the
Bonds frequently take time to make a trend change. The Expanded
The middle chart in Figure 10.13 is the Weekly Hang Seng.
Bond traders are accustomed to viewing in longer time frames as
USDS :'0 min
HHAYWet;
BSIMISISmin
in Figure 10.13 was accompanied by a negative reversal signal in the Composite Index. (The acceptable amplitude of the signal in the Composite Index is much larger than what was defined for the RSI.) Notice that wave b down (the decline from wave a to the next low that leads to wave c up) coincides with a buy signal in both the RSI and Composite Index. An asterisk adjacent to the Composite Index when it is testing support marks a very strong signal. Turn back to the DMK/$ chart in Figure 10.4. Earlier I mentioned that you would soon know why the new low in the dollar within this chart was not part of the originalfive-wavepattern. The chart displays a five-wave decline (into a double bottom) followed
Figure 10.13
by a perfect Expanded Flat. We are able to make a price projection
Source: TradeStation. © TradeStation Technologies
296 • CALCULATING MARKET PRICE OBJECTIVES
Using Oscillators with the Elliott Wave Principle • 297
'Flat.' I know, the wave 'a looks like afive-wavepattern, and the
be the correct interpretation.
three-wave decline, and that carries the message it is a corrective
will be significantly reduced that your Expanded Flat pattern will
into the low is not a five-wave structure. It must therefore be a
ship. Do not exceed this measurement; if you do, the probability
Remember wave four cannot overlap a first wave, so this decline
wave a. The rally that follows is in fact exactly to the 1.618 relation-
five-wave structure with all the internal overlaps that develop.
of wave b from the price low in this chart for the dollar relative to
the September S&P chart is choppy, and we cannot see a distinct
pass over so that you project a 1.618 relationship from the bottom
that the key in these patterns are the b waves. Wave b down in
level near 1.7770 in Figure 10.4). Now flip your precision ratio com-
pattern and chart allow me to emphasize the point stated eadier
of wave a (from the end of the marked five-wave decline up to the
we referenced before we knew what the pattern was called. This
for the distance wave c may travel by first measuring the distance
to categorize it. Trade it.
overbought into wave C. The final clincher is that wave C up has
in the wave b position within a Flat corrective pattern, stop trying
proximity between the Composite and RSI, and the market is
you identify the chop that forms the internals of the middle leg
tive reversal in the Composite Index, divergence peaks in close
a. The best advice I can offer you for these patterns is that once
wrong if it is exceeded. In the Hang Seng chart there is a nega-
on is if wave b within this pattern will break the origin of wave
target and should be known as it also marks where you may be
eral points in the oscillators. The information you want to focus
wave c had to form a 1.618 relationship, just that it is a frequent
three waves. Nonetheless, wave b is producing buy signals at sev-
below the 1.618 relationship with wave a. I did not suggest that
textbook line-on-close descriptions all state that wave a must be
Return to Figure 10.13. Wave c up in the Hang Seng is well
a completed five-wave pattern. Wczve Cs will always be five-wave
Beware that b waves can get very emotional in bigger pictures.
sentiment that neither side of the market can win.
perfect 1.618 relationship develops with wave a to form a textbook
that they will always be a three-wave pattern caused by a battle in
longer to see if the present five-wave advance will extend so that a
Straight-line capitulation drops can be b waves as well. The key is
patterns except within triangles. You are not going to wait any
a Flat pattern, the DJIA produced an Expanded Flat. Notice that
losing momentum.
updated now in Figure 10.14. While the September S&P developed
analogy, and we can also see at the top of wave c that the diver is
Our real-time tracking of the DJIA in a 15-minute chart is
scenario. The Hang Seng chart clearly fits the springboard diver
the DJIA because it must form five waves up from the end of wave b
July 24. The S&P correction became a Flat forming the letter N
the real wodd. However, we know an Expanded Flat is forming in
pattern that our indicators warned us about in Figure 10.6 on
origin of wave a. Yes, in a perfect Elliott world, that is true. Not in
Flat. Also look at the date of this chart. This is the down up
start of wave a. I didn't say that Expanded Flats must penetrate the
view of the September S&P 500 futures market; nice looking
wave b down in both the S&P and DJIA retrace back only to the
The last example on the right in Figure 10.13 is a 15-minute
298
• CALCULATING M A R K E T P R I C E OBJECTIVES
DJIY-15 min
Using Oscillators with the Elliott Wave Principle • 299
Fibonacci ratios described in earlier chapters. With expanded daily market ranges the need for greater accuracy requires greater care than using estimates. In Figure 10.14 we have divergence between the Composite Index and the RSI. The RSI is abiding by the range rules defined for a bear market by topping near the 60 to 65 level. Do we have a market top now? No, I do not think so for one reason. If we study thefive-waveadvance unfolding from the end of wave b, it is missing a couple of bars to form the final fifth wave up. Figure 10.14 also shows the market under a 0.382 Fibonacci retracement that many people will be able to identify. Expect an early advance near the open tomorrow, and we will use the advance from the end of wave b down to the small black arrow in this chart as a tentative measure to define tomorrow's target. Why? Balance. I am a fanatic about balance because the market is obsessed with balance and proportion. Also waves one and five are frequently equal. If we should surpass a small 'pop' to form equality and balance with the first part of the move, we will have to look toward the 1.618 relation-
'7/20
'7/21
' '7/22'
'7/23
'7/24
'7/27
ship with wave a. That will also mark the level where we view our Expanded Flat pattern to be wrong. Let's see what tomorrow brings.
Figure 10.14 Source: TradeStation. © TradeStation Technologies
It is July 28, 1998, after the market's close, and we see in Figure 10.15 that the DJIA produced a minor advance on open
our larger five-wave decline from the high. Figure 10.15 shows
I use now is the confluence objectives derived from overlapping
tion. Then the second leg down became the fifth wave to complete
the market path for tomorrow. 2nd Edition note: The price target
because one leg down is wave b within the Expanded Flat correc-
high near toward the close for July 27. Now we need to determine
we created Figure 10.7. The DJIA doesn't look like Figure 10.7
relationship with wave a, which is higher than the current market
right to be on the lookout for an extendingfive-wavepattern when
of wave a will be exceeded. Where will the target be? The 1.618
into today's low was afive-wavepattern, which means that we were
wave a is only three waves, so to finish five waves, we know the end
today and headed south the remainder of the day. The decline
in either pattern: a Flat or Expanded Flat. The pattern to the end of
300
• C A L C U L A T I N G M A R K E T PRICE OBJECTIVES
DJIY-15 min 07/28/9^ C=89365
Using Oscillators with the Elliott Wave Principle
* 301
T h i s is exactly what we discussed i n Figure 6.10, w h e n we were considering Fibonacci confluence points i n the Fibonacci chapter. W e are n o w finding that eariier discussion b e i n g applied by the D J I A market. There was a strong rebound into the close, and the divergence marked o n b o t h the Composite Index and RSI, compared to the price low, warns that the near-term advance w i l l develop f u r t h e r gains. T h e fastest way for a market to develop a countertrend move is to connect t w o five-wave patterns together and t h e n resume the larger t r e n d . T h i s is not a Flat pattern as the market does not retrace all of the first leg we called wave a. Chances are f a i d y good that we w i l l see o n l y a m i n o r pullback followed by further gains tomorrow, so we need to look at another corrective b u i l d i n g block that w i l l fit this scenario. W e have all the earmarks i n place for what we call a Zigzag
pattern to develop, so
let's move on to discuss this pattern. I f you c u r r e n t l y u t i l i z e the Wave P r i n c i p l e , may I have your attention just a m o m e n t longer regarding Expanded Flats before we move o n to Zigzags? I really need to try and resolve a p o i n t about '7/21
' '7/22'
'7/23
'7/24
'7J27
'7/28
w h i c h the industry is clearly confused and d o w n r i g h t b e f u d d l e d . There is no such pattern as an 'Irregular B.' There has never been
Figure 10.15 Source: TradeStation. © TradeStation Technologies
an 'Irregular B.' Prechter used to call the Expanded Flat pattern an 'Irregular Flat.' T h e name referred to the entire pattern, and
Irregular Bs. OK?
at 1.382.
sense to call it what it really is: Expanded Flat. So please, no more
marks the 1.618 relationship) is that there was a Fibonacci cluster
expanded variation of the Flat is so c o m m o n that it makes a lot of
for f a l l i n g short of the 1.618 calculation ( h o r i z o n t a l black arrow
letter N i n bear markets and an inverted N i n b u l l markets. T h e
pattern) advanced to a 1.382 relationship w i t h wave a. T h e reason
these patterns far more often than the simple Flat, w h i c h is just the
c of 4 (the h i g h this m o r n i n g that completes the Expanded Flat
changed because there is n o t h i n g irregular about it. You w i l l see
you wanted to mark a l l the internal subdivisions. Notice that wave
wave b i n this pattern was never irregular. T h e name was wisely
how the extending five-wave dechne w o u l d be correctly labeled i f
302 • CALCULATING MARKET PRICE OBJECTIVES
The next pattern we need to discuss is a Zigzag. I cannot think
Using Oscillators with the Elliott Wave Principle • 303
Zigzag (5-3-5)
Double Zigzag
of an analogy for this pattern because if you miss this one, it is more like a hit-and-run accident and all you want to do is get the license plate number of the truck that just ran you over. You can sense these aggressive corrections on the horizon before they appear on your screen. That is why I am very suspicious tonight of the chart pattern in Figure 10.15. We know a larger correction is due in the DJIA as the entire five-wave pattern could be complete from the high. We also have a five-wave advance into the close that forms from a
Figure 10.16
way immediately or reverse. You know a Flat is not developing.
consider the Zigzag further. Watch for cycle lows, indicator signals
move that cannot retrace the first leg we called wave a, get out of the
waves clearly in place, and then try a minor pullback, we should
However, if a wave b choppy price pattern appears after the five-wave
just turn into a three-wave pattern. But should the market keep five
although five waves will make me think of a Zigzag immediately.
I do not know if the current five-wave advance into the close will
structure at all. Remember, I do not care what wave 'a looks like,
five-wave decline, so we know the market rebound is incomplete.
a victim of the Zigzag.
so that it can return the larger trend that is usually accelerating.
are wrong about the prior trend returning. You are not wrong—just
get the correction over with and out of the way as fast as possible
reverse just as quickly about the time you begin to question if you
least amount of time for the market to develop. The market tries to
to the first. They are sharp. Take few prisoners along the way. Then
common patterns in a strong trending market because they take the
to see a 1.618 relationship in the second five-wave pattern relative
will warn that a Zigzag could be near as these corrections are very
equal to thefirstfive-wavepattern that formed wave a. It is common
the Cash S&P and OEX all positioned for a larger rebound. This
The secondfive-wavepattern usually travels a minimum distance
that seem stronger than prior signals, and other similar indices like
Our DJIA chart has been accelerating. You can see that supporting information fits.
Zigzags sometimes travel in pairs. Figure 10.16 shows a Double Zigzag where we encounter our first x' wave. It is also one of the
This is a common misunderstanding. If your Zigzag internals are
Flat, and our first five-wave unit would not have been a five-wave
internals of their first Zigzag as 'wave A within a larger correction.'
the first five-wave structure. Why? Because then we would have a
that the larger correction is incomplete, they hastily label the a-b-c
to expect this sharp correction as wave b down cannot retrace all of
analysts see a Zigzag pattern develop and they have an opinion
tern. If a minor pullback develops in the DJIA tomorrow, we know
should be to connect a second corrective pattern. As soon as many
five-wave cycles back to back. That is all there is to a Zigzag pat-
most abused x waves in the business because it is not used when it
A Zigzag as displayed in Figure 10.16 is made up of two sets of
304
• C A L C U L A T I N G M A R K E T PRICE
OBJECTIVES
then called 'wave A w i t h i n a larger correction,' you M U S T imply that the market will retmce the entire distance traveled by waves
Using Oscillators with the Elliott Wave Principle
FT U8
* 305
60
a-b-c
of your Zigzag pattern. W h y ? Because an a-b-c that is labeled 'wave A i n a larger pattern' has only two options: it must develop as a Flat or Expanded Flat. Period. N o other option. Flats always retrace all of wave A , and Expanded Flats should attempt to exceed the o r i g i n of wave A . So i f you don't t h i n k the entire distance w i l l be retraced, the next move must be called 'an x wave.' A n X wave is simply a wildcard that allows you to take one recognized corrective b u i l d i n g block and use it to l i n k another corrective pattern together. T h e different b u i l d i n g blocks should have some sense of balance and p r o p o r t i o n between the i n d i v i d u a l puzzle pieces. Y o u get a m a x i m u m o f three wildcards for any complex corrective pattern developing. T h e connectors can each be labeled w i t h 'X' or you can distinguish between t h e m by using X , T , and ' Z ' .
Figure 10.17 Source: TradeStation. © TradeStation Technologies
Bob Prechter introduced this latter m e t h o d i n later revisions of his
caused a lot of Lira to change hands along the way for some traders.
ond Zigzag w i l l also have a Fibonacci relationship between t h e m .
but this sure fits the character of a Zigzag. I ' l l bet it is, and it likely
m o n t h l y chart of the N i k k e i . Frequently the first Zigzag and the sec-
substructure of the decline marked between the arrows on prices,
for our DJIA chart now. T h e D o u b l e Zigzag i n Figure 10.16 is the
Bonds. I do not have data i n a shorter t i m e h o r i z o n to check the
plex patterns at another t i m e . M o r e likely we w i l l not need t h e m
L B T chart on the left is the m o n t h l y chart o f Italian G o v e r n m e n t
book T h e Elliott Wave Principle. W e w i l l look at these more c o m -
five-wave advance extended to complete wave a up, there is a clean
c o m m o n . G o l d displays an indicator divergence as does L B T T h e
where we left o f f two days ago. V i o l a ! L a Zigzag. W h i l e the first
ite Index w i t h the tops of waves a and c black arrows. T h i s is very
I can now wipe the perspiration f r o m m y brow and continue right
far right shows a negative reversal signal f o r m i n g i n the Compos-
the market was f o r m i n g a Zigzag right on cue for the book! So
T h e n the larger trend resumes. T h e L o n d o n F T S E futures o n the
trade? True, but that was not the only reason. I was excited because
similar. Two back-to-back five-wave patterns. Fast. D i r e c t . Sharp.
I started to j u m p up and down w i t h excitement. Why? A profitable
the 60-minute data of the September L o n d o n F T S E futures look
marked wave b i n Figure 10.18, w h i c h led to a Z i g z a g pattern,
develop. I n Figure 10.17 the weekly G o l d (futures G C ) chart and
W h e n the market on July 29 produced a three-wave decline,
Zigzags all look similar regardless of the market i n w h i c h they
306
• C A L C U L A T I N G M A R K E T P R I C E OBJECTIVES
DJIY-15 min
Using Oscillators with the Elliott Wave Principle • 307
one minor push over today s high will complete Elliott wave patterns developing in the S&P. The S&P will top at a major Gann target. The indicators in Figure 10.18 are at the same levels that marked the top of wave 4 up. We know Zigzags frequently form in markets that want to get the correction completed as soon as possible, so we have all the right ingredients for a sharp trend reversal. Do not hang around to look for a perfect Fibonacci target between waves a and c. Instead, calculate the next larger move, which should be down. I have removed the prior label at the low suggesting wave 5 and replaced it with an f to make a strong point about this market low. In the DJIA we have five waves down into that price low in Figure 10.18. The Expanded Flat that formed wave 4 up in the DJIA becomes wave A up in the S&P futures. Then wave B into the low and a clean five-wave advance expected to end tonight in the Globex night session will complete a large Expanded Flat. We will look at the S&P futures nearer the end as the Globex data will be more difficult. However, it means I have two Indices tracking and indicators that suggest further market weakness. Orthodox Elliott wave practitioners will calculate the next market
Figure 10.18 Source: TradeStation. © TradeStation Technologies
objectives by measuring the distance of the first five-wave decline and then projecting Fibonacci swings from the end of the correction. The 0.618, 1.00, and 1.618 relationships will be identified. However,
exceedfive-wavepatterns for Fibonacci swing projections.
time as most Zigzags are that distinctive and clean. Wave c in our
B down to the new low as occurs in the S&P. Always use wave Bs that
Zigzag. It was no major accomplishment to encounter one in real
thefive-wavepattern in the DJIA or thefive-wavepattern and wave
we completed the secondfive-waveadvance to end a perfect textbook
the entire distance traveled in the first part of the move. That means
of the first bounce from the low. In today's session for July 30, 1998,
traders who want a more accurate price objective should measure
three-wave decline to complete wave b down that did not retrace all
8,816, and the market high for wave c up is 9,041. The 0.618 and
the bottom of wave b. I do not think it will this time because just
In Figure 10.18 the DJIA high was 9,367, the price low at f was
Zigzag has not traveled a distance equal to wave a as measured from
308 • CALCULATING MARKET PRICE OBJECTIVES
1.00 price projections are 8,700 and 8,490. The first target is a high-confidence target. The 8,490 level horn 9,040 seems extreme.
Using Oscillators with the Elliott Wave Principle
309
D J I Y - 1 5 min 07/31/98 C=88807
There is also a Gann objective at 9,038, but we need to see how the market reacts to the first objective and what wave structure and indicator action forms prior to thinking about the second target. We are set for the next trading session. July 31, 1998 brought a decline from our Zigzag pattern. The first target we calculated of 8,700 has not been realized. However, with this new session we can see that our technical indicators do not display a price bottom. Both indicators have declined to prior oscillator extreme lows, and we know to expect a bounce in the next trading session followed by further selling as these indicators will not produce bottoms in their current positions. Wave structure appears to be accelerating as the market is moving in a direct pattern to each new low made throughout the day. The 8,700 objective should not be revised at this time. Notice an asterisk at the price low. I have changed the cycle position in Figure 10.19 from what we had been using in Figure 10.18. Why? We know cycles are not symmetrical, so keep the cycles in view on your chart current. This new period
'7/24
'7/27
'7/28
' '7/29 ' '7/30'
'7/31
Figure 10.19 Source: TradeStation. ® TradeStation Technologies
than this if I had tried. Indicator divergence, a negative reversal in
from today's action. An adjustment was indicated.
in this book to date. I could not have constructed a better summary
aligns the cycle to the actual price low and captures the early decline
second target. Time predictions are always a problem.
daily S&P chart has a major cycle low due to bottom tomorrow in
in Figure 10.20. It is not clear how long it will take us to reach the
closes into the lows of the session. A more serious problem is that the
first calculated on July 30 is being confirmed in the chart displayed
wave corrective rally. The market then declines to new lows and
tive at 8,700. The second target at 8,490 that looked crazy when
from today's session. We have a sharp decline followed by a three-
price low today is 8,785, which has still not realized our first objec-
market is in serious trouble in Figure 10.20 as we view the chart
both the Composite and RSI, price structure acceleration, and the
August 3, 1998: In a startling development, we see that the U.S.
sure from the top of the Zigzag pattern to the price low prior to
Figure 10.20 have every bearish signal present that we have discussed
We will calculate a new Fibonacci swing in Figure 10.20. Mea-
the daily time horizon. In the short-time horizon the indicators in
310
• CALCULATING MARKET PRICE OBJECIIVES
DJIY-15 min
Using Oscillators with the Elliott Wave Principle
• 311
H o w do we k n o w this? N o t f r o m E l l i o t t wave structure. T h e RSI has a negative reversal i n place, and the two peaks f o r m i n g the signal are marked N and R i n Figure 10.20. A l l the price-on-close levels that you need to make a price projection are recorded i n the top right corner. T h e f o r m u l a is i n the RSI chapter i f you need to t u r n back. T h e price projection f r o m this negative reversal is 8,703. T h e first target at 8,700 is now c o n f i r m e d , b u t so too is the second at 8,490. T h e amplitude for the RSI is greater t h a n the guidelines I gave you. B u t we also have a cycle low approaching i n the daily data, accelerating wave structure, a n d strong c o n f i r m a t i o n o f a target m u c h lower t h a n 8,700, w h i c h is 8,490. I f the market breaks 8,700, there w i l l be a problem. T h e amplitude i n the RSI is not a contradiction. It w o u l d be i f it were the only evidence present. I m e n t i o n e d earlier that wave b's are choppy i n short-horizon data, b u t they may f o r m capitulation lows i n longer horizon data. T h e only way to tell is the position o f your indicators. As a cycle low is approaching, it is less likely that the current decline w i l l be wave b d o w n f r o m the Z i g z a g pattern w i t h i n a larger Expanded Flat. O n l y t i m e w i l l tell.
Figure 10.20
A u g u s t 4, 1998: T h e second target was 8,491 f r o m
yester-
You should have defined the m i d p o i n t to be near the p o i n t where
the end o f a five-wave pattern f r o m the end of the Zigzag.
f r o m the h i g h i n Figure 10.21 i t is clear the decline is n o t over.
the close, this consolidation into the close is not a f o u r t h wave near
m e t h o d f r o m the m i d p o i n t o f the strongest s w i n g i n t h e move
t i m e p r o d u c i n g a back-and-fill correction prior to the sell-off into
however. Here is the bigger p r o b l e m : U s i n g the price projection
low has still n o t realized the 8,700 target. W h i l e the market spent
close was 8,490.9. W e got here m u c h faster t h a n I expected,
Today's 8,785
A G a n n target also resides at this level. Confirmation.
in the top left corner o f Figure 10.21. Target 8,490 to 8,491; actual
the Fibonacci target at 8,490 that we identified f r o m Figure 10.18.
T h e market d i d n o t bat an eye at 8,700. L o o k at the closing p r i n t
Fibonacci objectives are 8,642 and 8,491. T h e 8,491 target duplicates
30. F i g u r e 10.21 shows w h a t has transpired i n today's session.
today's three-wave corrective rally. W e find that the 1.00 and 1.618
day's price action t h a t c o n f i r m e d t h e 8,490 objective f r o m July
Source: TradeStation. © TradeStation Technologies
3 1 2 • CALCULATING MARKET PRICE OBJECTIVES
DJIY-15 mm
Figure 10.21
Using Oscillators with the Elliott Wave Principle • 3 1 3
DJIY-15 min
Figure 10.22
Figure 10.22 now shows the DJIA on August 5, 1998, after the
complete five-wave pattern, and the indicators do not display clear
a new low. The Gann target at 8,410 is the objective I would rely on as the Fibonacci projections below 8,490 cannot be confirmed at this time.
that developed yesterday on August 3. The current decline is not a
the second backward S crosses the price data. The backward S
Source: TradeStation. © TradeStation Technologies
Source: TradeStation. © TradeStation Technologies
marks only the 1.618 projection that was made from the decline
buy signals.
market has closed. The last chart suggested a minor bounce with consolidation did not develop until the market realized the Gann
into the close still needs a small fourth wave up and then a drop to
a decline toward 8,410. This was not quite right as a bounce or
I'll suggest a minor rebound upward on open as the last decline
3 1 4 • CALCULATING MARKET PRICE OBJECTIVES
Using Oscillators with the Elliott Wave Principle • 3 1 5
develop tomorrow, and the target would allow balance to occur from
of 4' in Figure 10.22.
(black arrow in Figure 10.22). Further weakness to a new low could
8,410, and then a rebound was attempted, which is labeled 'wave a
The same horizontal line of resistance is where wave 2 up ended
The market bottomed at 8,411.3. I'll take that since the target was
return to a horizontal line defining resistance for this indicator.
objective providing follow-through from the prior day's meltdown.
to measure a 1.00 and 1.618 relationship from current highs.
described and lean on Fibonacci analysis first and Gann second in
sure the distance traveled by wave 1 down, and use that relationship
projection method to use in which situation? I use all the methods
the early stages of the decline from the price high in this chart. Mea-
This may raise the question, how do you know which price
chart. There is also conflict present between the S&P futures chart
this example implies.)
the last leg down is a fifth wave to end wave 3 in the 15-minute
recent markets must allow for confluence zones to be wider than
Extending patterns in real-time are always tricky. It is possible that
is in a lower price range in 2011, the expanded trading ranges of
leg down called 'wave b of 4' is a five- or three-wave structure.
from the top of the target zone. (2nd Edition note: While the SP500
our first real conflict beginning to develop. It is not clear if the last
range between 8,409.4 and 8,411.7. The market actually rebounded
fifth wave down to complete the larger decline. However, we have
zone or cluster of Gann targets that actually fell within a narrow
The decline as labeled from the high in Figure 10.22 needs a
longer horizon time intervals. The 8,410 level was a major confluence
The advance to the high now marked wave a of 4 is a Zigzag pat-
and the DJIA tonight, and the two indices have to be watched.
horizon now.
know if the market should exceed the 1.618 target and trade near
closely tomorrow as it is extremely oversold in a 120-minute time
wave a. This objective has been marked in Figure 10.22. So we
complex pattern could be developing. We need to watch the market
can target the 1.618 relationship relative to the distance traveled by
the RSI through this trend line, it will be a warning that a more
Flat pattern develop. We know wave c up within an Expanded Flat
decline developing. However, if the market is capable of moving
the rally into the close, we thought that we might see an Expanded
resistance trend line, we can be more confident about a fifth-wave
so the last leg down into the low will be viewed as a b wave. Prior to
trend line. If the market produces an immediate reaction to this
that follows into the close is clearly a developingfive-waveadvance,
Figure 10.20. The RSI indicator is now approaching this same
wave orfive-wavedecline. That is a problem. However, the advance
negative reversal signal that gave us the 8,703 price objective in
into the low has wave structure internals that could fit either a three-
In Figure 10.22, notice that a trend line extends from the RSI
tern that is clearly a three-wave advance. The decline that followed
8,650, we may have to reconsider the final decline, which is now marked wave b of 4 as it is hard to tell what it is.
It is August 6, 1998. We have finally stepped off track as the market did not decline from the simple Expanded Flat pattern we may have correctly identified an Expanded Flat pattern, but this is
momentum peak and that wave c up has allowed this indicator to
suggested yesterday. Figure 10.23 shows today's trading session. We
You will notice that the Composite Index broke above a prior
316
• CALCULATING MARKET PRICE OBJECTIVES
O . I I Y H i min
[).IIY-I
mm
900
-89500
920
•89500
940
-90500 omem
c=65774
2
-87500 86500 111 Y
V*
8705 6646
•85500
880 i j - ^ I l t X reeo 840
84500
Using Oscillators with the Elliott Wave Principle • 317
choppy pullback that fights the sell signal the entire way down. The decline being described is called wave a' down in Figure 10.23. This is a warning because we know if our indicators move a distance much greater than prices, we are most likely on the wrong side of the market.
CMB Composite
'x - A ^ — / V r x
7, 7^
7
•120 '401
•70.00
'rai
'8/03
'8(04
'8/05
66.0
-10.
•35.00
10.(
•55.00 RSI Trend
'8)06
'm'
A rally is then attempted, and it is a very distorted looking advance that leads into a high near the close marked wave b'
- 20.00 RS1*AV5S
350
'7(30
' W V 'J/hV ' ' M 6 ' ' « ( ) ' ' 'Mil • '«;o5
up. The advance toward the close is what I would call a 'choppy mess,' and it is very well suited to be the character of a wave b advance in a short time horizon chart. Either the market is fighting the advance, though that is unlikely as it fought the decline as well into wave a' down, or the market action within waves a' and b'
Figure 10.23 Source: TradeStation. © TradeStation Technologies
is screaming at us to look for a cycle low approaching that we have
just one pattern within a larger move, and there are several factors that must be balanced.
not defined in Figure 10.22. We have witnessed the corrective patterns called 'Flats' and
cluster is exceeded.
breaks above the maximum oscillator peak, it would mean that
form a cluster. The trend resumes immediately after the x axis time
longer horizon charts such as weekly and daily. If the oscillator
degree triangles in markets. They develop when the time targets
in this time horizon. It is important to know the position of the
with Gann Analysis, it is Gann that is the best way to detect large
wave c up at the high of our 15-minute chart. It is a major top
cycle low to pass. (2nd Edition note: Now that I am much stronger
price high that completed the Zigzag corrective pattern marked
very common in small corrections when the market is waiting for a
time the Composite Index challenged this extreme was at the
range bound and moves sideways. Maybe we are due. Triangles are
test during today's session is marked with a black arrow. The last
third pattern is the triangle, which occurs when a market becomes
mum displacement in the 15-minute chart of the DJIA. The
basic patterns that it may use to create a corrective pattern. The
challenged the horizontal trend line drawn to define the maxi-
'Zigzags' develop in this decline. However, the market has three
In Figure 10.23 you will see that the Composite Index
(2nd Edition note: I make no cycle forecasts intraday now. Only use
declines rapidly while the price action produces an insignificant
position selected offers an alignment with three prior price lows.
zon chart data. From the black arrow the Composite Index then
adjusted from where it was positioned in Figure 10.22. The new
larger advance that would be seen developing in longer hori-
The cycle in the 15-minute chart of Figure 10.23 has now been
underlying strength is building within the market for a much
318
• C A L C U L A I I N G M A R K E T PRICE OBJECTIVES
Using Oscillators with the Elliott Wave Principle • 3 1 9
the market completes a structure with waves a-b-c-x-a-b-c.
forming wave c' down into the cycle low that should bottom eariy
we call it 'wave x.' Our complex pattern will not be satisfied until
complete a small Expanded Flat pattern that it started today by
second pattern with internals called a'-b'-c' is just a connector, and
would not be much help to us right now. The market will either
require at least three patterns to satisfy a larger correction. So the
small moves labeled waves a' and b'? Yes. The technical indicators
back-to-back corrective patterns. This is a problem. Markets always
market will do tomorrow. A l l this information from just the two
pattern is complete by forming wave c' down, we will then have two
identified. The same choppy price action is also telling us what the
care. Just know it is one lesser degree. When the second corrective
informative for us already by highlighting the cycle we have now
offer an underlined text feature, and in the heat of battle I do not
within that mess. However, this choppy action has been incredibly
with an underline to be formally correct. But TradeStation does not
torted price structure. There is not going to be a five-wave structure
degree because of the balance consideration. I should use a and b
wave a' into the high marked wave b ' is undeniably choppy and dis-
fore I know to immediately drop the wave count down by one lesser
In Figure 10.23 the price action that forms from the low marked
b' were the same degree as waves a-b-c in the first pattern. There-
bottomed. (2nd edition note: This remains true of cycles in general.)
called a-b-c. For these proportions it would be wrong to imply a' and
patterns that aggressively go nowhere until a cluster of cycles has
small moves in both time and movement relative to the structure we
place. In a long-term chart, markets may form extensive corrective
c'. The price action into the price pivots called a' and b ' are very
lows, or they will do all they can to waste time until the cycle low is in
pattern would be a smaller Expanded Flat with internals called a'-b'-
cycle low on open tomorrow. Markets will either decline into cycle
Flat pattern with internals marked a-b-c, and the second corrective
longer support.) It now suggests that the market will experience a
then have only two corrective patterns. The first is an Expanded
extremely helpful.) But this dialogue with a fixed cycle intraday I no
a'-b'-c' in Figure 10.23 form a very small Expanded Flat, we would
Gann time targets in a daily time interval. In fact, 2-day charts are
could continue upward to develop the advance we need to produce
our first complex correction.
a' into tomorrow's cycle low, that would be ideal. Then the market
from this price structure, we have now been politely introduced to
produce a wave c' decline that just breaks below the start of wave
further delay. As I do not think the market will make a new low
patterns back to back is of tremendous value now. If the market can
the cycle low is in place, after which it will advance sharply without
Knowing that we must always develop at least three corrective
in tomorrow's session, or the market is simply wasting time until
nearby market levels. Either way, the market should go up tomorrow.
suspect there is confusion about how to use it. If the waves marked
that would imply the market will develop an a-b-c advance from
introduced. People in the industry clearly hate to use the letter x. I
market may stall and go nowhere until the cycle low is in place. But
the use of an x wave in Figure 10.16 when a Double Zigzag was
a-b-c advance in the DJIA chart. Instead of a wave c' decline, the
pattern is becoming a complex correction. We first encountered
waves a-b-c. That would mean a fairly decent rally to develop another
We now have to use the dreaded x wave because the corrective
320 • CALCULATING MARKET PRICE OBJECTIVES
Using Oscillators with the Elliott Wave Principle • 3 2 1
terns. I'm not sure how we will end up.
cator signals present in Figure 10.23. The 15-minute DJIA chart
It is this indicator conflict that will create complex corrective pat-
that further gains are a high probability. There are conflicting indi-
zons in Figure 10.23, you will see that there is a directional conflict.
ing us a wealth of information, there is another reason to anticipate
oscillators in the 120-minute chart. If you study the two time hori-
While the small price action marked wave a' and wave b' is offer-
ing technical evidence will cost you money.
nals I am looking for in this indicator. You will find a black arrow
Trading too far ahead of the market's structure and its correspond-
crosses back through its zero midline. This is one of the main sig-
within my head sometimes rather than from what is on the screen.
dition has been alleviated by watching to see when this oscillator
most of my own trading errors. I trade from the ideas constructed
the RSI Trend oscillator to see when an overbought or oversold con-
minds. This timing difference is probably the primary source of
to discuss this indicator, but it is needed right now. I am looking at
take much longer to develop in the market than they do in our
been added called the RSI Trend. We have not had an opportunity
ahead, but we must not trade ahead of the facts. Patterns actually
suggesting that an advance could be near. Another indicator has
to jump ahead of the facts present within our data. We can think
Composite Index and RSI arc diverging in the 120-minute chart,
I knew this conflict was developing yesterday. But the key is not
is displayed along with a 120-minute bar chart of the DJIA. The
reason I continued to learn and improve.)
only as a lead signal. It is leading. We will follow it into tomorrow.
an evening report has never changed 15 years later. It is a major
Trend is an indicator derived from another oscillator, and it is used
thinking. (2nd edition note: Interesting, my objective of writing
attention when we discuss using oscillators on oscillators. The RSI
trading session as they are to inform clients about what I may be
example of how this oscillator is used, and we will give it more
are as much to keep me in control and focused within the next
the RSI Trend oscillator and our price data. We have just seen an
is losing sight of your original game plan. So the evening reports
120-minute DJIA chart. There is tremendous divergence between
through the trading day as the market action unfolds. The fault
duce another a-b-c advance will fit the technical picture within this
exposure because I continually have a stream of ideas developing
the idea we are developing of a complex correction that will pro-
S&P well from a one-minute chart in an effort to reduce capital
room to advance before it will cross the zero line again. That means
for waves a' and b', can be a fault as well. I do not trade the
ing the rally into the high. We can see this oscillator has much more
able to read small pieces of the market's puzzle, as we have done
that has developed a signal in conjunction with a fourth wave dur-
strict money management that keeps me in the P/L race. Being
is a confirmation signal. There is also an upward-pointing arrow
to trade. I struggle with this fine line constantly, and it is only
wave 2 up. That juxtaposition in wave count and oscillator position
On the other hand, we have to be ahead of the market in order
pointing down toward the RSI Trend oscillator that coincides with
under a Fibonacci 0.382 retracement in the 120-minute bar chart
previous overbought signals have been aligned with declining
Tonight we can see that the DJIA has advanced to a level directly
In contrast, the 15-minute chart is once again overbought. Our
322
• C A L C U L A T I N G M A R K E T PRICE OBJECTIVES
Using Oscillators with the Elliott Wave Principle • 3 2 3
indicators are clearly attempting to advance.
market is t r a n s m i t t i n g fairly easily tomorrow. I f the market unex-
new low is a m u c h lower probability because the 120-minute D J I A
a-b-c advance, so we should be able to read the innuendoes that the
w o u l d then be trading f r o m the alternate scenario. A decline to a
w i l l be favored. Even a stall into the cycle low w i l l lead to another
then challenge the current low i n the chart for Figure 10.23 and we
declines immediately f r o m this area of resistance, wave c' of x d o w n
tionship w o u l d mean that we are i n trouble as the market w o u l d
is a wait-and-see juncture at a critical market level. I f the market
monitor i f wave c' d o w n gets too zealous. A break below that rela-
and to a G a n n target that ahgns w i t h the m o v i n g average. T h i s
pectedly
reacts more strongly to the cycle i n Figure 10.23, t h e n
T h e chart prepared for Figure 10.23 is i n fact the chart used i n
W h i l e the target for the D J I A is 8,705, the target for the September
of these views that tracks w i t h the cycle low.
the view that a rally w i l l then develop is a h i g h confident outlook.
session tomorrow. N o damage has been done financially i n either
t i m e h o r i z o n signals. Predicting the market open is not as easy, but
and a key reversal swing back up c o u l d follow i n the same trading
tonight's client report to show the risk now present f r o m c o n f l i c t i n g
expect wave 5 d o w n as this w i l l be against the longer-horizon chart,
D i d I just suggest three-wave interpretations f r o m the same price data? Yes. T h e favored o p i n i o n is t h a t a m i n o r decline
S & P is 1,107 derived f r o m a G a n n W h e e l , w h i c h is an i m p o r t a n t objective f r o m the current price low.
war i n progress.
get so confused.
but rather day-to-day charts of w o r k i n g battle plans w i t h i n a larger
favored view that the market w i l l advance. T h a t is where people
confront t h e m . These charts are not perfect cataloged illustrations
that price objective. A weak open or n o t does n o t change the
not to give m u c h attention to larger-degree labels u n t i l forced to
target, and the data must still develop a three-wave structure into
wave units together and then see where we are at that t i m e . I tend
c o n t r o l o f the t i m i n g o n h o w i t w i l l advance t o w a r d the same
w i l l end at wave (5) d o w n , so I ' l l just label t w o back-to-back
alternative? N o . It is a l l o w i n g some r o o m for the market to take
the larger picture is g o i n g to become an extending pattern that
market w i l l ignore the cycle entirely and just advance. Is that an
a-b-c Zigzag pattern w o u l d then be wave (2) up. I a m not sure that
the 0.382 retracement i n this same chart. A variation is that the
the first five waves d o w n w i t h the n u m b e r ( I ) below wave 5. T h e
relationship of the measured move. T h e market is c u r r e n t l y at
where the Zigzag pattern is completed. T h e convention is to mark
0.618
T h e target is m a r k e d i n the 120-minute chart and is the
There is an asterisk i n the 120-minute chart that marks the h i g h
on open w i l l t h e n lead to a three-wave advance toward 8,705.
T h e alternative wave c o u n t is a market decline to a new price
T h e D J I A m a r k e t a c t i o n for A u g u s t 7, 1998
five-
is displayed
and 1,107.
exceed a 1.618 relationship w i t h wave a'. W e now have a target to
has realized a h i g h of 1,108.50. O u r o r i g i n a l targets were 8,705
scenario may be proven wrong. W e k n o w wave c' d o w n should not
has r a l l i e d to a market h i g h of 8,710.0 and the September S & P
But we must have a price objective that defines where our favored
i n F i g u r e 10.24. W e are d o i n g very w e l l today as the
low. T h a t is what is unexpected. W e k n o w it w o u l d be wave 5 down.
DJIA
324 • CALCULATING MARKET PRICE OBJECTIVES
DJIY-lb niiti
DJIY-DHily
Using Oscillators with the Elliott Wave Principle • 325
based on facts, not based on trying to fit structure into our analysis. Big difference, and the change can be seen in Figure 10.24. Figure 10.24 also displays the current chart for the daily DJIA. The daily chart shows that the current price low at 8,361.9 is neariy an exact 0.618 retracement. The current advance that has developed today to 8,710 is to the 0.382 ratio within this same mathematical grid marked on the daily chart. The reasons that these specific pivot levels have been used in both these charts to calculate Fibonacci ratios are fully described in the Fibonacci chapter. Why the price spike in the daily chart has not been used to calculate the Fibonacci retracement targets is also explained in the Fibonacci should not be used to create Fibonacci retracements.
Source: TradeStation. © TradeStation Technologies
chapter in great detail. This offers a real-time example of why spikes
Figure 10.24
Just one problem: Right market direction, right targets; wrong pat-
We have a new problem. We can see from the daily DJIA chart
wave (2) up having followed ourfirstfive-wavedecline. The current
wave b of 4, but a fifth wave. We are making a wave count adjustment
be within wave (3) down?' The asterisk in Figure 10.24 would be
addition, we are now forced to accept that the final leg down is not
is, 'Where will wave 4 up become too proportionally large to still
must look at a much bigger picture now to consider our options. In
or is complete. The question developing in the short-horizon data
0.618 ratio in the 15-minute chart from the price low means that we
to be addressing if the second five-wave unit is extending further
the market low into a major price target. The current rebound to a
100 X 3 and 50 X 3 for intraday decisions in the S&P We need
afive-wavepattern. So that means we have a Zigzag in place from
I sometimes use Point-and-Figure charts. Use the dimensions
that develops out of the triangle is not a three-wave structure. It is
to patterns in overseas markets.) To try and answer this question,
of the day toward our objectives. Here is the problem. The advance
I start looking at it in different ways and pay particular attention
become a triangle. A rally then unfolds throughout the remainder
Figure, I generally know something ugly is hiding within a chart.
the market to react in this manner to the cycle. Therefore, wave x has
an extending pattern? (2nd Edition note: If I switch to Point-and-
had considered as we knew the cycle was present that would allow
chart. Is the current price low wave 5 down or wave 3 down in
market has elected to stall into that cycle low. It was an option we
labeled this same decline as afive-wavestructure in the 15-minute
c' decline to complete a small Expanded Flat pattern. Instead, the
structure from the high marked with an asterisk. We have just
not get one. The cycle identified yesterday has not produced a wave
in Figure 10.24 that the entire decline is not a completed five-wave
tern. We needed a three-wave advance into these objectives and did
326
• C A L C U L A T I N G M A R K E T PRICE OBJECTIVES
Using Oscillators with the Elliott Wave Principle • 3 2 7
can go either way still, so we are not getting anywhere fast f r o m just
t h e m to our next chart.
the 0.618 retracement objective does not ovedap wave 1 d o w n . W e
to label the charts w i t h the higher degrees. O K , I ' l l give i n and add
up is choppy, and this bounce is clean and simple. T h e rally into
(3) down. I just painted myself into a corner, and I a m now forced
look at the rebound relative to the decline f r o m the h i g h . Wave 2
low is not too large yet to still be considered wave 4 up w i t h i n wave
the asterisk. It fits the guideline between waves (2) and (4). N o w
bounce i n our daily chart that we k n o w is a Zigzag f r o m the price
T h e t i m e frame or corrective price structure that constitutes 'too
looking at structure i n these charts.
out, you m i g h t say, i n the short-horizon charts.
the same degree. I n other words, we do not k n o w i f the rally that
be attempted u n t i l additional analysis is considered. W e are tapped
w i t h i n the same ' u n i t ' or cycle, it is less likely a correction w i t h i n
market opinion? N o o p i n i o n . 'I'he p o i n t is that no o p i n i o n should
taken over twice as m u c h t i m e to develop as a more simple pattern
further work i n the shorter h o r i z o n charts. But what is the current
oping is an i m p o r t a n t issue. As a general guide, i f a correction has
W e are forced to look at the bigger picture before we can do any
large' and out of proportion relative to the whole structure devel-
the sharp rally that developed today w h i c h became the strongest
satisfy an entire correction?
cent decline now i n place. O u r indicators even warned us about
a of (4) is suggested. But is today's simple Zigzag rally sufficient to
significant top. W e haven't missed a beat throughout the 10.7 per-
just a simple Zigzag f r o m the b o t t o m . T h a t is why the idea of wave
f r o m a pivot level most people never suspected w o u l d become a
rection. Wave (4) is a larger degree and m i g h t develop more than
However, we have been tracking a 1,006-point decline i n the D J I A
wave (3) decline, or wave a up w i t h i n a developing wave (4) cor-
W o n ' t we miss a t r a d i n g day i f an o p i n i o n is not offered? Yes.
developed right to our target today is wave 4 up i n an extending
finish.
ability that a more complex pattern w i l l f o r m i n position 4. Easy.
format has meant a workload that frequently takes u n t i l 3 a.m. to
simple pattern develop w i t h i n wave 2, there w i l l be a higher prob-
ahead. I ' m tired because describing the evening analysis i n this
W e need to k n o w that now. T h e guideline implies that i f we have a
day. Today's session for August 7 is a Friday, and we have a weekend
to anticipate alternating corrective patterns between waves 2 and 4.
the short-horizon price action go where it w i l l w i t h o u t us on M o n -
quick recap: T h e E l l i o t t Wave Principle provides us w i t h a guideline
a day o f f now to evaluate where we stand i n the bigger picture. L e t
suspect you may have forgotten a point or two f r o m that chart. So, a
retracement since the entire decline began. W e can afford to take
It has been so l o n g since we first discussed Figure 10.2 that I
M o n d a y w i l l be reserved for longer h o r i z o n work, and then
ture. W e can pick it u p f r o m there.
Zigzags frequently develop when the market wants to get the cor-
a big move, and then come back to the short-horizon price struc-
quick rebound characteristic of a Zigzag correction. W e also k n o w
we can continue. It is called pacing. Take a mental ' t i m e - o u t ' after
W h e n we look at the daily chart i n Figure 10.24, we see a sharp
nowhere i n particular. W e have not missed a n y t h i n g i n the market
rent advance is about h a l f the t i m e the market required to rally into
It is now Monday, August 10, and the D J I A and S & P are d r i f t i n g
rection over w i t h so that it may resume the larger trend. T h e cur-
328
• CALCULATING MARKET PRICE OBJECTIVES
so far. It would appear that the market is taking a day off as well. Until now we have focused only on the DJIA. The September S&P
Using Oscillators with the Elliott Wave Principle * 329
SPU8-240 min
futures contract must use the Globex night session for both wave structure and price projections now. The S&P chart has an important indicator signal in this chart. In Figure 10.25 is the 240-minute chart for the September S&P 500 futures market that was captured today about 2 p.m. on August 10. The structure labeled in this chart suggests that the decline from wave (2) is not a complete five-wave pattern. Waves (1) and (2) have been added to make this discussion easier to follow. But I still resist implying that this entire decline will develop into a five-wave pattern because in the very big picture this will become a fourth wave down. It will likely end after a very large Zigzag pattern from the high is in place. (Why? This is the fastest way for the market to get the correction over with as the larger trend will return to record new price highs.) The proportions developing fit nicely. We would then label the entire decline waves A-B-C. So do not assume I am a bear for much longer. Wave 4 up, to my sense of balance, is on the verge of being too large to proportionally fit the decline from the top of wave (2). However, it has not exceeded twice the time consumed to develop wave 2 of (3) so we are still within acceptable boundaries. If this advance is wave 4 up, it should be satisfied now at the 1,108.50 high
Figure 10.25 Source: TradeStation. © TradeStation Technologies
that was made on Friday.
Therefore, the market should advance provided that the indicator
has declined from the last peak (double asterisk) to a support level
market if we were short. The market is not following the indicator.
interpret this indicator? The Stochastics indicator in Figure 10.25
what we have used to suggest we would be on the wrong side of the
is telling me what the next market move will be. How would you
isk). Meanwhile prices have held their ground. Why, that is exactly
this view is given some confidence by Stochastics. This indicator
created by placing moving averages on the indicator (single aster-
My sense of balance needs wave 4 up to end right now, and
330 • CALCULATING MARKET PRICE OBJECTIVES
holds the support average. So why am I comfortable with the idea that wave 4 up has a top in place now? Because viewing Stochastics as a buy signal right now is not a cor-
Using Oscillators with the Elliott Wave Principle • 331
pay attention to this signal. While the Composite Index diverged with prices near the bottom, both the RSI and Stochastics oscillator did not diverge. This is important for Stochastics. This is a trend-
Stochastics formula.
Index and RSI with moving averages as the basic palette from which
there are no similar patterns present. This signal is unique to the
is war! Grab anything you know works in war. I use the Composite
go along with it. If you look at the RSI and Composite oscillators,
and we have not spent a lot of time discussing Stochastics. Fair? This
A 'poop.' It offers us an immediate signal and a price projection to
Wait. This is not fair because we have not been tracking the S&P,
market condition is what George Lane calls a 'pop.' The reverse?
indicator and how Stochastics responds within a trending market.
displacement that developed wave (2). This pop out of an oversold
down to the single asterisk. It does not consider the character of this
popped to the top of the screen and has formed a peak at the same
tomorrow considers only the indicator travel from the double asterisk
know which will be correct). Again we see that the indicator has
up for one specific signal. The first interpretation to suggest a rally
the high that became wave (2) up (or wave B up as we still don't
ning of this book and to monitor this formula when it is correctly set
condition, it pops up to the top of the screen. Look at the pop into
sons: to monitor the Stochastics Default Club discussed at the begin-
At the first chance this indicator is given to alleviate the oversold
not trade from Stochastics as you are seeing, but I watch it for two rea-
it is clearly spending the majority of its time below the 40 level.
off track at this juncture and I did not think to discuss it eariier. I do
up' where it becomes a flat line at the bottom of our screen, but
developed now because it would have caused a lot of people to step
ing market in this time horizon, and Stochastics has not 'locked
rect interpretation for this indicator. I am very relieved that this signal
tion to equality and two times the first measurement. The market
techniques that we have covered that would be a fit for the problem.
2.618 targets. In this chart from the top of wave 4 up is a projec-
But when I need clarification, there is something in our arsenal of
formed the Stochastics pop by projecting 1.00, 1.618, 2.00, and
cators just to confuse me. I can do that very nicely without any help.
first.) This price decline can then be used from the high that
well when they are needed. Not before. I do not need a slew of indi-
(You can use the Stochastics bottom, but let's be conservative
formulas, and various price projection techniques are called upon as
take the corresponding price move to make the price projection.
The methods we covered for reverse-engineering, other indicator
travel in the Stochastics indicator preceding the current pop, and
present now is one of those I must be on the lookout for at all times.
projection from the Stochastics signal. Measure the distance of
watched for in other indicators. The signal in Stochastics that is
I know I covered this earlier, but here's how to make the price
to analyze a market. I know there are specific signals that must be
Now we have a Stochastics signal present that is like being hit
will make a new low. zon charts so that we can determine how the current market decline
feels like to be hit by the market's bat, and I have been taught to
In addition to this signal, it is important to view the longer hori-
in the head with a baseball bat. I cannot ignore it. I know what it
332 • C A L C U L A T I N G M A R K E T P R I C E O B J E C T I V E S
Using Oscillators with the Elliott Wave Principle
• 333
the A m e r i c a n market as it d i d i n this decline. T o r o n t o was already
ger horizon analysis.
a n d f r e q u e n t l y tracks w i t h a t i m i n g displacement compared to
displayed i n m y Internet report for August 10, 1998, to address l o n -
it has a h i g h c o r r e l a t i o n , develops a very clean wave structure,
fits w i t h i n a larger picture. T h e f o l l o w i n g chart i n Figure 10.26 was
market action. L o o k at the lows that f o r m at waves iv and 4. T h e
end of this chapter.
M u c h t i m e was devoted to setting Fibonacci ratios based o n prior
was a second wave. W e w i l l r e t u r n to a big-picture outlook near the
to calculate the Fibonacci retracements i n the Toronto Exchange?'
wave practitioner, because the b o t t o m of the 1987 low i n m y view
Someone is going to ask me, ' W h y d i d you use the price spike
sions i n context to a G r a n d Supercycle top. Sorry. N o t this E l l i o t t
I m i g h t add that the G e r m a n Dax Index is the leader at this t i m e .
assumes that I a m m a k i n g a statement about the relative subdivi-
the first F i b o n a c c i target before t h e A m e r i c a n Indices. G l o b a l l y
waves ( I ) and (2). T h i s situation is n o different. Everyone quickly
how the market is f o r m i n g a fourth-wave decline that has realized
wave (1) up to start the move. T h e 120-minute D J I A chart also used
5 developed i n t h e S & P . So t h e T o r o n t o Index is used to show
to understand that I label every s u b d i v i d i n g five-wave pattern as
waves i n t o new highs. T o r o n t o f o r m e d a second wave up as wave
in the m i d d l e of Figure 10.26. D o n ' t j u m p to conclusions. You need
d e c l i n i n g w h e n the A m e r i c a n markets were developing their fifth
Suddenly all the E l l i o t t wave traders have been lost to the chart
not ignore the spike this t i m e . T h e decline that has developed has
It is interesting relative to t h e A m e r i c a n stock indices because
market tells us i f we should use the spike or not. That's w h y I d i d
I n Figure 10.26 the m i d d l e chart is the T o r o n t o Stock Exchange.
now realized the 0.382 retracement. T h e market is suggesting that D.IIY Monthly 08/31/88 C=85748 -232
lOHY Monthly 08/31/88 C=852370 - 1 1 7 ^ •800000
•90000
/ n
/ l/ll/
/
/
V
ACY Daily 08/10/98 C=566B7 -29
JompulerTecTinology J
-60000 -54000
•750000 •700000
Composite Index 70,00
•70000
'/—TIIT' /
1
ijll
I
l/
kGOOOO
tv
fl J
iJ//
•60000
III /
•50000
- ilifl' Jr '/ (1) III
•40000
•30000
'92 '93 '94 '95 '96 '97 '98
J
•650000
RSkAvgs •60.00
•600000
4
•550000
/1 ^

•500000
rf iJJ
1
'
' j ' 'j '
AOY Daily
•400000 •350000
'J
this was the right calculation. O n the far right is the daily A M E X C o m p u t e r Index and A M E X O i l and Gas Index. As these t w o sectors make u p more t h a n 10 percent of the total S & P 500 Index, they have a strong influence on
now declined to a very long-term trend and to the top of the previ-
•44000 AMEXOIIand Gas
'J
the direction of the S&P. B o t h indices display incomplete patterns. O n the left i n Figure 10.26 is the DJIA. It shows the market has
'v
08/10/98 C=40815 -779 -4BO0O
•460000
W^
f
'M
92 '93 '94 '95 '96 '97 '98
ous f o u r t h wave. The Fibonacci
grids within this chart show how the
Toronto Index was used to define the start of a large third wave. W e have balance as t h e m i d p o i n t s overlap i f we align t h e A m e r i c a n
|||
Index w i t h Toronto's. So we have a short-horizon o p i n i o n that a
'A
new market low w i l l develop. But m y guess is that we w i l l f o r m a
Figure 10.26
Zigzag decline w i t h i n a large f o u r t h wave f r o m the h i g h . W e w i l l
Source: TradeStation. © TradeStation Technologies
334
• CALCULATING MARKET PRICE OBJECTIVES
Using Oscillators with the Elliott Wave Principle • 335
pattern up? I do not know. I can only try to stay one day ahead of
forward and begin to look at the longer horizon.
and wave structure into the advance. Will it just be a corrective
to see if the Stochastics signal is correct or not. Then we can move
should rebound tomorrow. I need to look at the indicator positions
track only one more short-horizon chart in our real-time scenario
horizon outlook to consider.
tics pop correctly warned that market weakness would be seen
scenario will be dropped now because we have the much longer
ure 10.27 shows today's action for August 11, 1998. The Stochas-
the market. I think tomorrow will be an up day. Our day-to-day
This is a good place to end the short-horizon journal. Fig-
cator signals can be used to create a wave count.
in the price projection section of this book is diverging. The DJIA
At this point you have a very good idea of how the day-to-day indi-
reverse. The momentum extreme indicator we discussed earlier
timing as I have to see the footprints made along the indicator trail.
in Figure 10.25 is now present in the daily DJIA chart in the
picture for us every step along the way. For this reason I never know
tive oscillator is clearly diverging. The Stochastics pop described
I cannot tell today, but the indicators will tell us and will paint a
that this latest decline is occurring into a cycle low. The deriva-
triangle? Will the market develop five waves down from the high?
10.25 (lower target at 2x). The daily chart is being used to show
rection? Will a large wave b rally form the second leg up within a
objective has been exceeded slightly from that projected in Figure
X up follow after this Zigzag decline that will lead to a larger cor-
confluence levels defined by Gann and Fibonacci. The Stochastics
What will the market do next in the bigger picture? Will wave
today. New price lows have developed in the S&P and DJIA to
-PY-Daily
Always keep in mind that balance and proportion are of much greater importance than having all the internals perfectly cataloged. If the balance and scale is not right, fix it. This is the same as making adjustments to cycles as required. The longer the time horizon, the more time you should devote to the balance, proportion, and symmetry of your wave scenario. With practice the bal-
var Derivative Osci
^,jylll>l>.,yjll.,,,yy.
H||ii^r^'^!i**|||
ance becomes more than just playing the right notes on the piano. We were fortunate to have a market reaction to a key objective that could be turned into a choreographed dance. It can happen. Not
E)(tremeKy2 -2665,55
always. But it is worth the effort to study all you can because you can
,i^k.,k,J*.J,.n
now understand what it is like to fall in step for even a brief period. F
M
A
M
J
Figure 10.27 Source: TradeStation. © TradeStation Technologies
Once you experience this flow and harmony, it becomes an addiction. You will easily become motivated to work hard to find the next sonata that may grow beyond just the notes within a chart.
3 3 6 • CALCULATING M A R K E T PRICE OBJECTIVES
Using Oscillators with the Elliott Wave Principle * 3 3 7
triangle. It was not a three-wave structure if you wondered about it
over the last few days.
made more difficult to see by a sprawling fourth-wave contracting
has its own strengths and weaknesses. We have seen both illustrated
rally may have bothered some people as thefive-wavestructure is
the ElHott Wave Principle is clearly a viable technical method that
developed. In Figure 10.18 there is a Zigzag rally. Wave c in this
more potential than you first thought? I cannot convert anyone, but
ket scenario, and I tried to squeeze past the first triangle when it
and your accountant's. Might not the Elhott 'Piano' have a bit
We actually had two triangle examples in our real-time mar-
The right notes in perfect harmony can become music to your ears,
some people.
have tremendous problems with this pattern because it is common
fourth wave was indeed a triangle, and that might have confused
contracting triangle in a short-horizon DMK/$ chart. Most people
have had to compress the data to make it very obvious that the
most common way this pattern develops. Figure 10.28 displays a
is actually wave c' of four in this small five-wave advance. I would
Figure 10.24. The pattern that developed in the DJIA is in fact the
three-wave decline that is just prior to the high marked wave c up
tunity to discuss both patterns. We encountered a small triangle in
Zigzag was complete, confirming the pattern. In Figure 10.18 the
However, the daily S&P 500 data will provide us with ample oppor-
at the time. Also the market cleariy fell to new price lows when the
There are two patterns remaining that we have not discussed.
a three-wave structure. Case closed. No exceptions. If any of the
the orthodox high.
single leg you label as one of the swings within a triangle must he
frequently produce a wave b second leg that breaks the origin of
is called wave a-b-c-d-e. That too is easy. Here's the catch. Every
illustrate this pattern. However, markets are permitted to and will
develop within a triangle pattern. That part is easy. Each swing
triangle pattern in Figure 10.28 is the pattern most books use to
industry to illustrate a common error. Suppose that five swings
show a perfect pennant formation. A dotted line in the contracting
triangles. I say this because there are so many examples in our
that most illustrations in books that show line charts nearly always
There is in the industry a lot of misunderstanding about
for wave b to exceed the origin of wave a. The problem might be
waves in your triangle has five waves, it is not a triangle. The most Contracting Triangle
Expanding Triangle
abused market in recent years was the weekly chart data for the U.S. T-Bonds. Most people used the outer dimensions of a contracting price structure to suggest that a contracting triangle was forming. If you go back to look at these wave interpretations, you will find afive-wavepattern in the first wave position. That should have been an early warning that this market was not developing a bullish contracting triangle. When the second wave became a five-wave pattern, it ruled out the possibility that the market
Figure 10.28
338
• C A L C U L A T I N G M A R K E T PRICE OBJECTIVES
Using Oscillators with the Elliott Wave Principle • 3 3 9
f o r m i n g on market news. W h i l e it can happen, it does not have to
a triangle.
Expect this because it happens often. A lot is said about wave Es
was f o r m i n g a bearish c o n t r a c t i n g triangle. Bonds d i d n o t f o r m
a t i m e projection for resolution of the pattern. T h e intersection of
pattern for their first wave.
this pattern to end. It is the m a x i m u m a m o u n t of t i m e allowed, not
is h o w people got caught on the w r o n g side: they used a five-wave
these two exterior trend lines converge marks the t i m e allowed for
T h e resolution f r o m wave B up w o u l d be a wave C decline. T h a t
highs and higher lows to mark a projected intersection. W h e r e
decline, and then a bearish triangle w i l l develop to f o r m wave B up.
be complete is to extend the converging trend lines of the lower
is a distinct five-wave pattern, it can be used as wave A , let's say for a
be the cause. O n e way to determine where the triangle pattern must
W h e n the first leg of what is thought to be a contracting triangle
I n a triangle the first leg d o w n is generally sharp. T h a t is why the
these two trend lines is the apex of a pennant chart pattern.
thrust must occur before the market exceeds the apex of the pat-
make a new h i g h .
w i l l explode i n what we call a ' t h r u s t ' out of the triangle. T h e
that the market wants to preserve. Wave B up is then p e r m i t t e d to
ket for the next move. A t the c o m p l e t i o n of the pattern, the market
triangle i f a Z i g z a g decline f r o m the h i g h is the correct pattern
T h e triangle is a c o i l i n g pattern b u i l d i n g energy w i t h i n a mar-
current decline i n the D J I A may develop into only a long-horizon
of your o r i g i n a l expectations.
is well k n o w n that wave D may not break below wave C . D o not
w i l l produce a thrust out of the pattern i n the opposite direction
T h i s is also displayed i n Figure 10.28. I n a contracting triangle, it
that is just one leg o f f is c o i l i n g i n the opposite d i r e c t i o n , and it
wave a. I f it does, you have to consider an e x p a n d i n g triangle.
leg of your triangle is not part of the o r i g i n a l structure. A triangle
a contracting triangle, wave C may not break the t e r m i n a t i o n of
to be f o r m i n g . There may be a larger triangle, or perhaps the first
it comes back to f o r m another leg i n the complex structure. I n
solidation, the pattern is not developing the triangle you believed
most t i m e to develop, and every t i m e y o u t h i n k it is complete,
tern. I f the market crosses t h r o u g h the apex and it is still i n a con-
Wave C w i t h i n a t r i a n g l e is very sneaky. I t usually takes the
2.618 swing relationships. T h e n project these ratios f r o m the ter-
determine stop placement.
the market i n wave B, and t h e n calculate 0.618, 1.00, 1.618, and
C i n a bearish triangle). Use another price projection m e t h o d to
can make a price projection. Measure the distance traveled by
page exposure for stops below wave C is horrendous (above wave
O n c e the market has developed a thrust out of this pattern, you
put your stops just below wave C for this very reason! T h e slip-
we address the longer horizon analysis.
kets are frequently a last leg d o w n that w i l l throw o f f your t i m i n g .
degree best fits this fourth-wave correction w i l l be deferred u n t i l
T h e y are hard to identify. Be aware that wave Es i n financial mar-
not considered the larger picture for this market. D e c i d i n g w h i c h
f o r m their o w n contracting pattern, w h i c h can be very confusing.
i n Figure 10.29 is simply called a ' f o u r t h wave' because we have
relationship relative to wave B. Wave Es, on the other h a n d , may
m i n a t i o n of wave E. T h e resolution of the triangle interpretation
Wave D is not as tricky because it frequently forms a Fibonacci
340
• CALCULATING MARKET PRICE OBJECTIVES
Using Oscillators with the Elliott Wave Principle • 341
can be defined. I would also argue about the placement of wave b up in this chart. But whether or not this structure is a triangle will have no impact on the larger issue. The debate over this pattern was ended once and for all when the decline that developed in April and May 1998 moved the RSI only to the 40 to 45 range for this indicator. That is a decline within a bull market based on the range rules defined in Chapter 1. Case closed. If you viewed indicator signals as more important evidence than wave structure, the only conclusion that you could derive would be that the market needed to develop afifth-waveadvance from the June lows. It should be noted that those in favor of the triangle scenario were Figure 10.29 Source: TradeStation. © TradeStation Technologies
forced to use the choppy shructure into the high marked wave '5?' as the end of the advancefiromthe resolution of the triangle. When a fifth wave cannot exceed the third wave in a developing five-wave
10.29. We will discuss this analytic technique in the next chapter.
at this wave structure now in Figure 10.29.
The RSI has an upper and lower volatility band displayed in Figure
not. In Figure 10.10 the data is compressed. Let's take a closer look
on its own. Elliott wave structure by itself is deadly in my opinion.
ary 1998. The debate concerned whether a triangle had formed or
cators made the more accurate call compared to using wave structure
over the price structure that developed from August 1997 to Janu-
case. Both interpretations correctly called for a decline. But the indi-
there was a great debate within the S&P Elliott wave community
structure, it is called a failure. It failed to make a new high in this
When we discussed Figure 10.10, we briefly mentioned that
triangle pattern: the diagonal triangle, or rising wedge.
called wave b' up. That could end the prior advance. Therefore,
To discuss the first wave, we need to look at a distant cousin of the
on the right, and you will have a five-wave pattern into the high
'Where is the first wave and from what price low does it begin?'
daily S&P 500 futures chart. Move wave a' over to the adjacent low
If afive-waveadvance developed, it raises a significant question:
There are several ways to label this triangle in the middle of the
develop in wave positions 4, B, and x only, the diagonal triangle
not matter how you interpret each wave. A valid three-wave swing
steam, so to speak. While a contracting and expanding triangle will
Each wave within this structure is a three-wave pattern, and it does
at the end of a larger trend when the market is running out of
with much better proportions, but it does not change the outcome.
tern that formed in a financial market. These patterns will occur
first leg. It will not change the debate. I can give you a triangle
In Figure 10.30 we have an actual bullish diagonal triangle pat-
this triangle can have a sprawled-out first wave a down or a sharp
342 • CALCULATING MARKET PRICE OBJECTIVES
Bullish Diagonal Triangle
Using Oscillators with the Elliott Wave Principle • 343
Bearish Diagonal Triangle
h ttp ://ww w. aeroin ves t .com
Aerodynamic Investments Inc ^ F i g u r e 10.30
will develop only in wave 5 or C. The internals are similar in both
'J
Source:
ture. Waves 1, 3, and 5 in a textbook environment will develop
F i g u r e 10.31
patterns since each swing or leg will develop a three-wave struc-
'J
>
's
'o
W
b • • '98 ' V • W
'A ' W ' V
'j
TradeStatlon. © TradeStation Technologies
Zigzag patterns. Do not get hung up looking for Zigzags in a real
chop your trading account into shreds so that you end up trading the
returns to the origin of the wedge.
pattern will save a trader a lot of money. This pattern in real life can
wedges will end. Once the pattern is complete, the market rapidly
declining slope between each successive impulsive wave within this
that you use a guide to help determine where these rising or falling
for the fifth wave relative to the third. The three lines displaying a
when the exterior trend lines converge. What is more important is
your third wave in the wedge is not complete. The same can be said
choppy, they will do. The wedge shape should be quite apparent
forward. If your third wave has a slope that is steeper than your first,
market environment. As long as the price structure internals are
market declines back to the wave iv in June 1997, which is the start
declining slope of each line gives the impression that they are falling
You can see that the resolution from this pattern is sharp, and the
of these lines. They are progressively becoming less steep, and the
asterisks to mark waves that form a rising diagonal triangle pattern.
connect the start and end of each impulsive wave. Look at the slope
structure to the left in 1997 that defines wave v of 5. There are five
triangle. You will find three lines in the bearish diagonal pattern to
explanation. We will come back to this in a moment. First look at the
been horizontallyflippedso that I could illustrate a bearish diagonal
ter interpretation. This chart shows a five-wave pattern that needs
data displayed in the falling wedge that ended a market decline have
chart as displayed in Figure 10.29. However, Figure 10.31 is the bet-
prematurely thinking it is complete. Here's a useful guide: The same
In Figure 10.31 we are viewing the same daily S&P 500 futures
ronment because an error can easily be made in the third wave by
final resolution poorly. The slope is the key to surviving this pattern.
Tracking these patterns can be very tricky in a real-time envi-
3 4 4 • CALCULATING M A R K E T PRICE OB)ECTIVES
of the wave v diagonal that later ends in August 1997. This is the pat-
Using Oscillators with the Elliott Wave Principle • 3 4 5
three-wave structures for each wave in a termination pattern. How-
wave building blocks are cataloged. The 13 patterns are:
interpretation shown in Figure 10.31. A diagonal triangle forms
pattern is quietly mentioned when all the patterns of the Elliott
It is pattern number 5 in this list that is the key to the chart
every single one except number 5 in the list that follows. The fifth
Patterns 12 and 13 are just complex corrections that use connected corrective building blocks together in a group. We will look at these in a moment.
There are 13 Elliott wave patterns, and we have discussed or used
tern that always marks the end of a trend and leads to a correction.
ever, there is another wedge pattern. It is the diagonal triangle type
strates a diagonal triangle in the first wave position that ends into
Leading diagonal triangle (wedge)
only one of two positions; wave 1 or wave A. Figure 10.31 demon-
4. Ending diagonal triangle (wedge)
except this pattern must be a distinct wedge, and it is permitted in
3. Truncated fifth wave or failure
wave breaks the top of wave 1. Is that not a rule violation? Yes,
2. An extendingfive-wavepattern
2, or the leading wedge. The internals are 5-3-5-3-5, and the fourth
1. A simplefive-wavepattern
5.
bonds because bonds generally are slow to establish a new trend.
9. Expanded Flat
However, you will find that this pattern develops more often in
8. Flat
rencies and equity indices and should be used only in hindsight.
7. Double Zigzag
the market high of December 1997. It is a rare pattern within cur-
6. Zigzag
10. Contracting triangle
In Figure 10.31 the RSI indicator has three oscillator lows marked
the prior range are demonstrated once again. However, the indica-
13. Triple threes
new low, so the guidelines we discussed that an oscillator will test
12. Double threes
1 through 3. The low at point 3 is equal to point 1. Point 2 makes a
11.
Expanding triangle
tor level cleariy warns that the bull market is leading into a fifth Pattern 3 in this list is the fifth-wave failure when wave 5 is
wave, which negates the triangle.
The minimum is three complete patterns, and the puzzle pieces
a failure for Figure 10.29.
any three complete corrective patterns that follow one another.
situations are often forming fifth-wave failures. We had to consider
these last two patterns to be highlighted. A double three is just
new high as S&P futures fail to form a new high. Futures in these
10.23, but it turned into another Zigzag. So Figure 10.32 allows
in intraday data. The reverse is also common: The Cash makes a
nario. We were close—I thought we had one developing in Figure
when the Cash S&P makes a new low when the futures do not
complex patterns did not develop in our real-time market sce-
used more often than truncated. It is a common trading signal
In the list that cataloged the 13 Elliott wave patterns, the two
unable to exceed the termination of wave 3. The term failure is
346 • CALCULATING MARKET PRICE OBJECTIVES
Double Threes
Triple Threes
Using Oscillators with the Elliott Wave Principle • 347
a standard two-dimensional board. However, the analysis work is not two dimensional. It is synonymous with playing chess on a threedimensional board. I would use the indicator signals from underlying components within the S&P 500 Index in conjunction with global equity indices and the key financial markets of Currencies, Bonds, and Gold. The balance between indicators and wave structure must balance in my three-dimensional game of chess. When they align and you see the pattern as a three-dimensional model, you have
Figure 10.32
something to get excited about. I will not go into great detail about
will be constructed from Flats, Triangles, and Zigzag patterns. Figure 10.31 shows another real-life example as wave 4 is a complex double three. A common error made within our industry can be discussed from the double three chart in Figure 10.32. If wave x up were higher than wave b, which is at the price high, would it still be wave X? No. How do you know that the market is not forming a Flat or Expanded Flat? Waves a-b-c would be labeled wave A, and the x wave would become wave B in a larger pattern. You would then have to see afive-wavepattern develop in the market to complete the Flat or Expanded Flat pattern. Wave x's into new highs or lows are usually wrong wave counts. I'm not saying they will never occur, but I have not used an x into a new high or low for any time horizon or financial market within the last six years. That means it is far more rare than the elusive leading diagonal triangle.
how to take the information we have just covered up to yet another level by showing you a three-dimensional model. It could easily be done, but that's why people pay me for my market opinion, so I'm not going to be stupid and give everything away But I did promise that I would show you why I had a bullish outlook for equities by utilizing the Wave Principle when others had a different view. First, you know I construct wave counts from indicator signals, not just from price data structure. That alone will lead to conclusions and market interpretations that are different from those of conventional Elliott wave practitioners. But I also construct my larger market opinions from components within an Index rather than depending totally on indicator signals developing from the Index itself Let me show you. The S&P 500 Index is made up of 500 stocks, but most stocks within the Index have no weighted significance in terms of the actual movement of the Index itself In fact, there are numerous S6'P Sectors within the Index that have no weighted significance on the
Long-Term Analysis
future movement of the Index. In Figure 10.33 you will see that 19 cent within the S&P 500 Index. In addition, 12 stocks make up 21.27
of chess, I would refer to that market decline as a game of chess using
S&P Sectors contribute an underlying weighted value of 58.58 per-
If the S&P decline we tracked together in this chapter were a game
348
•
C A L C U L A T I N G M A R K E T PRICE OBJECTIVES
I AEKODYNAMIC INVESTMENTS INC htti>://www.«erolnve«tc<>
r
Sector
5 33%
A|>ill 199B US Equity Index Summaiy
SW SOD IndeK (Sector Subset) % »f S i P (stock) % of S&P I
Financials 16.03% Mjr. Reg Bank 5.08% Money Ctr Banks 2.99% Techrology 15.68% Computer (hrdw 8. Soft) B,B1%
Consumer Staples 14.38%
Beverages (non)
2.92%
Consumer Discr. Foods 2.57% Household Prds 2 14% Tobacco Health Care 10.39%
Capital Goods 9.74%
143%
Russell 2000
Nasdaq 100
Using Oscillators with the Elliott Wave Principle • 3 4 9
The underlying components of the Index change. A market leader with significant weighting within the Index today may be less influential in a year's time. Therefore, to base all your analytic work on the Index itself is the same as swimming in an ocean and never
13.40% IBM 1.44% INTC 2.17% IV1SFT 2.27%
1 52%
MO
1 27%
PG
2 02%
KQ
68.94%
looking for a shark or barracuda passing under the surface that could ruin your day.
2.80%
When you consider that Energy and Computer Technology
16.00%
alone contribute an underlying weighting of 25 percent to the S&P 500 Index, you can see why Figure 10.26 had the A M E X
9 90%
0 17% 4 05%
Composite Indices for Computer Technology and Energy on my 3.08%
in your index, you have a higher probability of being right. So I
Hth Care (Drugs) 4.41%
computer screen. If you know 25 percent of the underlying weight
BioTechnology Health Care (Div)
Aerospace/Def 1.21% Elect. Equip 4.21%
Manuf. Diversified 2,29% Consumer Cyclicals 9.27% Autos 1.77% Producer Durables n/a Energy 9.00% International Integrated B.41% Communications 5.96% Industhals Basic Materials
Utilities 3.06% Transportation Other
Telephone
PFE MRK
1.06% 1.59%
analyze 20 S&P Sectors that have nearly a 60 percent weighting and align their wave patterns and indicator signals to construct a
216% 1.62%
mN RD
3.04%
GE
wave interpretation for the Index. I use the same approach then
underlying weightings of these individual indices. Finally, you can
3.90% 060%
for the German Dax and Nikkei. I then have three opinions from
4.10%
8 22%
3 72%
19.11%
weight the Global indices themselves. Yes. It is a lot of work. But it is worth it to try and stay on the right side of the market. Bond markets require a similar approach, but the yield curve is
Chemicals 1.67% Gold Mining 0.35% Mat. 8. Processing
9.87% 6.41%
Elec. Companies 2.38%
weighted and then the Global rates. Currencies become a puzzle Q.65%
1.17% 21.27%
50.58*
100.00%
1.22% 100.00%
1M.0fl%
between crossrates, Bonds, and Gold. The markets where I have been caught dead wrong are usually markets where I am either unable to look at underlying charts or I failed to do so. I will never trade the
Figure 10.33
markets I refer to as 'roots and hooves.' We do not get along at all.
S&P sectors that offer current data up to August 7, 1998. The four
also an interesting record for the 2008 crash and third leg to come.)
picture for Bonds and U.S. Stocks. Let's start with a few sample
torical reference because of the Tech crash a couple years later. It is
approached for longer-horizon work, we can take a look at the big
While this is old information now, it is kept in the book for his-
W i t h your understanding of how the underlying analysis is
percent of the total S&P 500 Index in April 1998. (2nd Edition note:
350
• CALCULATING MARKET PRICE OBJECTIVES
Using Oscillators with the Elliott Wave Principle • 3 5 1
sectors in Figures 10.34 to 10.37 are wave interpretations that my clients know have been unadjusted wave scenarios since 1993. It explains clearly why I have been so bullish. In the bigger picture I am still looking for the end of wave (3) up. Wave (4) is then expected to form a massive contracting triangle. Why a triangle? We are due. The market has not used that puzzle piece in the really big picture. It will at some time. Within 22 Sectors that I monitor, only four have needed a significant wave change in four years, excluding adjustments for the developing extension. We may well see a fifth-wave blowoff to end wave (3) up in the future, but the entire rally is not a blowoff since 1987 to end a Grand SuperCycle fifth wave in my opinion. It should be very clear now that my wave interpretations for the S&P 500 Index have been derived from aligning key wave (2) sector declines that have the strongest weighting within the Index itself
Figure 10.34 Source: TradeStation. © TradeStation Technologies
Figure 10.35 Source: TradeStation. © TradeStation Technologies Inrjinnfil Hnnkr. Monthly
r Ruqlunul Bank;; WL'oklv
Figure 10.36 Source: TradeStation. (c) TradeStation Technologies
352 • CALCULATING MARKET PRICE OBJECTIVES
Using Oscillators with the Elliott Wave Principle • 353
Moody's Aaa Bond Price
Figure 10.37 Source: TradeStation. ® TradeStation Technologies
to 'construct' a wave scenario for the Index. Keep in mind that the weighting changes within the sectors change and must be monitored for future wave interpretations. Corrections hke the present
80
90
00
10
20
30
40
60
60
70
<5) 80 90
IMllMlllllllllllllMlllllllllllllllllllllllllllllllllllllllllllrillllllllllllliiiiliMilinrliiiiliiiiliiirliiiilii.ili
one have and will change leadership in the underlying Index. This is not an orthodox way to apply the Elliott Wave Principle. But, as I
Figure 10.38
mentioned earlier, this is war. Grab anything you know works to win a battle! This approach has been working in a global environment as well. The last two sector charts in Figures 10.36 and 10.37 have
When do I look at the Stocks listed in Figure 10.33? When I need to see what is brewing under the surface within each sector.
received a single email to show a reader had made a connection to
corrective since the introduction of Bond futures at the CBOT
Edition was likely never studied by readers 15 years ago. I never
10.38, this will raise a few interesting questions. The entire rally is
repeated at the start of this chapter in Figure 10.39 for the 2nd
est rate-sensitive Index. When you look at the Bond chart in Figure
to the next chapter. (2nd Edition note: But the last chart that I
the way, Major Regional Banks is a composite sector that is an inter-
for review without further comment. It is time for me to move on
thumbnail method. The market will answer you if you are right. By
Now the big picture. I'll leave the final charts in your hands
the internal wave counts deliberately removed so you can use the
354 • CALCULATING MARKET PRICE OBJECTIVES
DJIA- Q U A R T E R L Y September 1962 to August 1998 AERODYNAIMIC INVESTMENTS INC http:/f«mw.acroinvest.com Elliott W«VB Interpretation Connie Brown, CMT ®
PART 5
A u g 1998
DJIA- YEARLY 1932 to April 1997 AERODVIMMC INVESTMENTS INC (£) May 1997 As Published in T e d i n i c a l Analysis of Slocks and Commodities Magazine, June 1997
Triangle intended scaled to X-axis for forecast.
not to be the time
Figure 10.39
the coihng pattern suggested for the years ahead. Now if you missed the revised forecast at the beginning with an expanding triangle pattern, I'll flag your attention to it as the last leg down in Figure 10.40 will be a life changing event for many people.)
NEW METHODS FOR IMPROVING INDICATOR TIMING AND F I L T E R I N G PREMATURE SIGNALS
chapter 11 VOLATILITY B A N D S ON OSCILLATORS By the end of Part 3, you will feel certain that no conventional indicator is safe in my hands. 1 will invert it, smooth it, compound it, disassemble it, superimpose it, and even imbed it—all in an effort to understand it. —Technical Analysis for the Trading Professional, First Edition
I
t never occurred to me that some people would tabulate all the indicators I referenced throughout the book to see i f they
could identify the Composite Index. The Composite Index was the only imbedded formula. Momentum has been added to a fast RSI to break the limitations of normalization when an oscillator must travel in a fixed range of zero to one hundred. I never use volatility bands on an oscillator that has unrestricted movement. Volatility bands prevent detrended averages from defining historical extremes, as was demonstrated in chapter 8. Many consider common indicators such as the M A C D , Stochastics, and the RSI as sacred entities. Most believe an underlying formula must never be altered. Not even the conventional display • 357
358
• IMPROVING INDICATOR T I M I N G AND F I L T E R I N G P R E M A T U R E SIGNALS
Volatility Bands on Oscillators
• 359
now before we move forward.
indicator plotted i n an unconventional manner. Therefore, i n Part
data. I t h i n k his f o r m u l a is easier to apply as w e l l . So let me fix this
because you may not have considered l o o k i n g at a conventional
I had described o m i t t e d his use o f t w o volatility bands o n price
W o u l d you not change it? Chances are h i g h that you do not k n o w
give you the TradeStation f o r m u l a he used o n prices. T h e m e t h o d
out a signal to you more clearly i f one line is plotted as a histogram?
asked me to change i f this book went into a second edition was to
of an indicator should be changed. But what i f Stochastics shouts
T r u e Average Range Convergence Bands).
more importantly, discard or m i n i m i z e the elements we do not like.
Please call t h e m S T A R C Bands. It was their original name (Stoller
you that we can m i x and m a t c h the elements we like and t h e n ,
dors w i l l be able to m o d i f y it to fit their standard specifications.
T h e t i m e has come to push the boundaries of convention, to show
W h i l e the f o r m u l a below is i n TradeStation format other ven-
T h r e e m y mission is to challenge your perceptions about indicators.
T h e last three chapters w i l l , I hope, give you greater flexibility and broaden your options. I n this chapter we look at a m e t h o d w i t h w h i c h to address a
[LegacyColorValue = true]; input:av(6), atrlen(15), factorl (2), factor2 (3);
top2 = mav -I- (factor2 *atr);
used only w h e n the question arises i n the market. It is monitored
topi = mav-h (factorl *atr);
to answer that one particular question. T h e m e t h o d or indicator is
mav=average (c, av);
are sometimes answered by indicator formulas expressly designed
atr=average (truerange,atrlen);
oversold can a market become?' Specific questions of this nature
vanatr (0), mav (0), topi (0), top2 (0), botl (0), bot2 (0);
specific market problem. T h e problem is, 'Just how overbought or
for only one specific signal at one critical market juncture. T h e rest of the t i m e you m i g h t ignore the indicator or m e t h o d entirely.
botl = mav-(factorl*atr); bot2 = mav - (factor2*atr);
For example, the m o m e n t u m extreme histogram displayed i n the if topi >0 then plot2 (topi, 'StollerHi2');
m e t h o d we are about to look at now.
if top2>0 then plotl (top2, 'StollerHi');
reverse-engineering chapter is one m e t h o d that compliments the
Using volatility bands w i t h indicators rather than on prices can answer the question, 'Just how overbought or oversold can a market become?' It w i l l not answer the question every t i m e , but w h e n this m e t h o d steps forward and makes a statement, i t should be respected.
if botl >0 then plot3 (botl, 'StollerLo'); if bot2>0 then plot4 (bot2, 'StollerLo2'); T h e AvgTrueRange is used to smooth out price bars w i t h volatility that is higher or lower than n o r m a l . TrueRange is defined as the larger of the following:
T h e f o r m u l a for volatility bands that I use was given to me by M a n n i n g Stoller. Sadly we have lost h i m , b u t one of the things he
• T h e distance between today's H i g h and todays Low.
360
• IMPROVING INDICATOR T I M I N G AND F I L T E R I N G P R E M A T U R E SIGNALS
Volatility Bands on Oscillators
* 361
• T h e distance between today's H i g h and yesterday's Close. • T h e distance between today's L o w and yesterday's Close.
08/21/98 C=1107r 0
T h e value for the L e n g t h i n p u t parameter should always be a positive whole n u m b e r greater t h a n 0. You w i l l find more about this i n the book by W i l d e r , Welles, Jr. New Concepts Trading
Systems,
in
Technical
McLeansville, N C : Trend Research, 1978.
M a n n i n g Stoller used this f o r m u l a only o n price data. I t is an alternative to Bollinger Bands because extreme price moves do not exceed S T A R C bands. There is only one signal that I look for i n S T A R C Bands. T h e signal is w h e n the market data pulls o f f the volatility band and then fails to reach the band i n a retest attempt. Years later I realized this was a signal similar to what George Lane used w h e n he said to wait for divergence w i t h lower v o l u m e to
93
'94
'95
'96
'97
'98
Figure 11.1
enter. H e therefore needed a t h i r d divergence signal to gain per-
a good wake-up call.
they are no different t h a n how someone who favors volume uses it
d o i n g analysis at all hours of the n i g h t as well. T h i s one signal is
mission to execute an order. I f you really t h i n k about these signals
as c o n f i r m a t i o n . I never display volume because I read oscillators i n a way that they deliver the same message.
T h e signal itself is very simple. L o o k at the upper and lower extreme ranges for the RSI. W h e n the RSI is near an extreme
high
the RSI and outer bands i n this chart are not used because other
easily missed after an exhausting week of t r a d i n g w h e n you are
interpretive relationships you m i g h t extract elsewhere
especially at e m o t i o n a l market extremes—not ones that can be
i n Figure 11.1 that have been h i g h l i g h t e d by black arrows. O t h e r
c o u l d be near, and you may be missing i t . ' I need blatant signals,
touch the outer band on a second attempt. There are four signals
attention and look at other methods n o w because a trend reversal
have to display divergence or create a reversal signal—just fails to
ticular signal that nearly always demands respect and states, 'Pay
is it. Period. N o t h i n g more to add. T h e RSI does not necessarily
But it produces one par-
and fails to challenge or reach the outer band a second t i m e . T h a t
w i t h volatility bands. T h e reason I like this f o r m u l a is that it gives
signal occurs w h e n the RSI attempts to resume the former trend
G e r m a n G o v e r n m e n t B u n d futures. A 14-period RSI is plotted
low the strong move w i t h a pullback f r o m the band. T h e market
for use w i t h oscillators. Figure 11.1 is a weekly chart for
fication
or low and is t o u c h i n g the volatility band, the indicator w i l l f o l -
I had applied M a n n i n g Stoller's volatility bands w i t h a m o d i -
very few signals
that warrant attention.
between
362
* IMPROVING INDICATOR T I M I N G AND F I L T E R I N G P R E M A T U R E SIGNALS
Volatility Bands on Oscillators
» 363
ing average envelopes or moving standard deviation bands from a
extremes for this signal and keep it simple.
reason that I do not care for band formulas that operate on mov-
methods we have covered offer better timing. Just monitor the
Look at the upper band relative to the RSI. The RSI violates the upper boundary only four times in a five-and-a-half-year period. That is a correct setup placement for this band. Now look at the lower band. There are two square brackets to mark an area where the RSI travels outside the boundary of the lower volatility band for a period of time. You do not want this to occur. The RSI may travel in extreme moves along the boundary of the band, but not outside of it—at least not until you are familiar with the character of this signal. It is important for volatility band formulas to accommodate separate coefficients for the upper and lower bands. It is for this
simple moving average. The formula that creates the band displacement in Figure 11.2 is derived from an average of the true range of prices to accommodate data gaps. Each band then applies a separate coefficient so that the user has independent control of the upper and lower perimeters. The modified formula in Figure 11.2 continues to use Stoller's original 6-period average with a I5-period average true range. The coefficients that I start with are 2.1 for the lower band and 2.3 for the upper band. I find that one pair of bands is sufficient when you use Stoller's formula on RSI. The same weekly chart for the German Bund futures is dis-
Indicator: RSI+Vol Bands
played in Figure 11.3. The lower band is now using a coefficient of
Input: Coefdwn(2.1),Coefup(2.3); Plot 1 ((Average ((RSI (Close44)),6))+(Coefup* (Average (TrueRangeCustom((RSI(Close,14)),(RSI(Close,14)),
LGB Weskly
09/21/96 C=11077 0
(RSI(Close,14))),15))),'Plotl'); Plot2((Average((RSI(Close,I4)),6))-(Coefdwn*(Average (TrueRangeCustom((RSI(Close,14)), (RSI(Close,I4)),(RSI(Close,14))),15)));Tlot2'); Plot3((RSI(Close,14));Tlot3'); If CheckAlert Then Begin If Plotl Crosses Above Plot2 or Plotl Crosses Below Plot2 or Plotl Crosses Above Plot3 or Plotl Crosses Below Plot3 or Plot2 Crosses Above Plot3 or Plot2 Crosses Below Plot3 Then Alert=True; End;
'83
Figure 11.2
'94
'95
'96
'97
'98
Figure 11.3
364
• IMPROVING INDICATOR T I M I N G AND F I L T E R I N G PREMATURE SIGNALS
3.1 instead of 2.3. The 3.1 was selected by appearance alone knowing that the RSI should rarely exceed the band. The upper band has
SPMonthly 08/31/98 C=106040 -190
Volatility Bands on Oscillators » 365
LFT-Monthly 08/31/98 C=55200 -3460 -110000
been 'tweaked' by using a coefficient of 2.5 instead of 2.1.
55000
-7GO00
boundary as the band in the areas marked by square brackets. At
-90000
The lower band now shows the RSI traveling down the same signals 1, 2, 3, and 4, the signal forms with an M or W pattern in
45000 .r'l*ii[l ,.^.'•'^,.I.'H'
50000
the RSI, showing the second peak or trough failing to reach the RSI+Vol Bands
70.00
remain fairly consistent for an individual market, however.
80.00
pattern will vary considerably. The personality of the W or M will
•90.00
outer band. The width of the oscillator pivots within the M or W
p ''Vw'
35O00
RSI*Vol Bands 85.00
Signal 3 shows the second peak well under the rise of the band.
75.00 B5.00
BO.OO
The first peak did not touch the band. This is an example of
55.00
50.00
how you would examine the relative displacement of RSI to the outer band. Figure 11.4 contains a very strong signal because of the double
'92
'93
'94
'95
'96
'97
45.00 '93
'94
'95
'96
'97 '
Figure 11.4
bottom that RSI formed. But RSI has a major flaw that will be
together in a monthly chart, it was a massive buy signal.
bull market. This is the range rule concept that was demonstrated
for that time horizon. When you consider that all this was coming
to 45 marks the end of a correction within the context of a larger
prices and allows the indicator to break into a new higher range
ing a bottom between a range of 40 and 50. The zone between 40
indicator to hold this level generally explodes into a third wave in
examples you will experience using these bands. The RSI is form-
defines resistance for a bear market. A market that can allow an
Index and German Dax. These signals are the cleanest, strongest,
that the RSI declined to the 65 level, which is the upper zone that
S&P 500 Index and is then repeated in the monthly London FTSE
a choppy corrective pattern. But it is more important to recognize
extreme trend reversals. The signal first develops in the monthly
and the price just fights the indicator all the way down by forming
addressed in the next chapter. RSI often misses the big picture
in Chapter 1.
Figure 11.5 is used to show that Bollinger Bands applied to prices
deviation bands on prices offer an analytic tool, but timing is a
the price fails to follow. The indicator falls nearly to the lower band.
displayed with prices. In the monthly FTSE chart the standard
the importance of not staying short when the indicator declines and
RSI. This would be true even if the band formula on the RSI were
a question mark. This one signal has been highlighted to reinforce
are very different in character from the band formula applied to the
There is a sell signal marked in the London FTSE data that has
3 6 6 • IMPROVING INDICATOR TIMING AND F I L T E R I N G PREMATURE SIGNALS
TY-Monthly 08/31/98 C=114'17 +'31
Volatility Bands on Oscillators
• 367
10 Year Treasury Note
A ^Hf|
A
4/V
•118'24 •106*^08 •93*24
RSI+Vol Bands
Figure 11.5
problem. The market is able to track the upper boundary for several years. The Bollinger Band narrowing in conjunction with a choppy price decline, which corresponds to signal 3 in the RSI relative to the upper volatility band, is of value. But, if timing is important, Bollinger Bands alone will offer limited help.
'87
'88
'89
'90
'91
'92
'93
'94
'95
'96
'97
'98
Figure 11.6
other indicators, but how much value it really adds is questionable.
stronger signal. More is not necessarily better.
system. To simply buy or sell based on this indicator pattern alone
divergence between the RSI and Composite Index that offers the
ity bands. It is essential that they be used with a price projection
Signals 1 and 2 display the pattern that is watched for, but it is the
Let me strongly caution you about the RSI signal within volatil-
would be financial suicide. Please keep in mind that these signals
are based on an indicator that sees only a closing price! This is a perfect example of a method being only an analytic method and not a trading signal. Figure 11.6 is a chart showing the 10-Year Treasury Note market in a monthly time horizon. The band formula can be applied to
L
chapter | 12 THE COMPOSITE INDEX
et's get the criticism of the first edition right out of the picture up front. What's the formula?!
The Composite Index will diverge where the RSI often fails to warn a trend reversal is approaching. The reason the Composite will pass on the warning is that it has a momentum indicator imbedded within the RSI formula. In English the steps you need are the following:
Step 1: create a 9-period momentum indicator of a 14-period RSI
Yes, it is momentum on RSI, not price. This is an
indicator on an indicator. Step 2: create a very short 3-period RSI and then a simple 3-period moving average of the 3-period RSI to smooth it slightly. Step 3: add Steps 1 and 2 together to create the Composite Index. Step 4: Create two simple moving averages of the Composite Index using the periods 13 and 33. Step 5: Plot the Composite Index and its two simple averages as three lines in a separate frame under the price data.
369 •
370
• IMPROVING INDICATOR T I M I N G A N D FILTERING PREMATURE
SIGNALS
The Composite Index • 3 7 1
presentation. That sends me crazy.
informed historian about who was who, who did what, and how
do go it seems everyone has something they want to sell within the
together called CompuTrac. T i m is perhaps our industry's most
think it is the same to watch from home on their computers. If you
created by T i m Slater. T i m put the wodd's first software program
developed the tool for which they were so well known. Now people
speakers. This was a major annual event that was an exciting week
tion was invaluable as you could see how the originators used and
Manning was my boss. Manning was one of the original ' T A G '
they had the freedom to openly discuss their work. The informa-
the World Trade Center, North Tower and I worked on the 104th.
value was being with people you never had access to in person and
me, it was Manning Stoller. Manning worked on the 105th floor of
do not go to these venues or just offer basic introductions. The
If anybody deserves the credit for developing this formula with
they tell us about a market.
result, most of us with something to say, or something different.
gence patterns at the extreme ranges are just the beginning of what
lectual rights over to the conference coordinators these days. As a
oscillator is its horizontal displacement from the zero line. Diver-
because of content concerns. As a speaker you must sign the intel-
your learning. As example, I learned that the greatest value of any
occur so far as I am aware, partially because of cost, but primarily
opportunity. Without the question there is no forward movement in
what happened to 'never'. But sadly such conferences no longer
from a question. If you can recognize the question, look at it as the
' I ' l l never go to one of those Elliott or Gann lectures.' Ha! Look
tions' taught a serious life lesson for this business; never walk away
1 recall looking at the weekly schedule to set my plan and stating,
oscillators on my own. He was a great teacher because all his 'ques-
special. For those of us new to the industry it was overwhelming.
this problem he gave me a better understanding of how to develop
room to room. You knew you were witnessing something incredibly
be able to see them. I was sent away to solve the problem. Through
at the crossovers and there was an excited buzz that vibrated from
small windows that looked like postage stamps. He also wanted to
merchandise! The halls were filled with people swapping rooms
easily. He once wanted 10 indicators within the maximum eight
keynote speaker at dinner. No one was allowed to sell product or
Manny knew I could program the custom user interface of TeleTrac
two speakers for four hours before dinner. Then there would be a
end up with something that begins to look like a very fast M A C D .
listened to two speakers in four hours before lunch and another
ple you can put the Derivative Oscillator into the RSI frame. You
of technical analysis. They all gave their speeches twice and you
bine indicators into groups to solve the display problem. As exam-
week before the world's most recognized names in the business
could not zoom in/out or move. Manning taught me how to com-
few. Total greenhorns like me could immerse themselves for a solid
display more than eight tiny windows on a single screen. They
Welles Wilder, John Murphy, and John Bollinger to name but a
a Microsoft Windows platform. As a result, the software could not
Lane, Robert Prechter, Jr., Gerald Appel, Richard W. Arms Jr.,
Trac, a product once under the Telerate corporate name, was not
at the Annual TAG conference were industry giants like George
How the Composite Index was born was out of necessity. Tele-
people came into the industry and contributed to it. Other speakers
372
• IMPROVING INDICATOR T I M I N G AND F I L T E R I N G P R E M A T U R E SIGNALS
The Composite Index • 3 7 3
opportunity to learn. It all begins with a question.
the people and their indicators. When an individual began to claim
answers today. You have to know and if you do not, recognize the
within the technical analysis industry to know the stories behind
these questions pass by unanswered because you do not know the
the business and retired (as a trader) did a great deal of research
versus fifth wave endings? (There will be divergence.) Don't let
Ed Dobson who used to own Trader's Press and has since sold
individual, with Manny's approval, for his opinion. I was likely try-
imbed formulas by adding them together.
naive about how this business functioned. I sent the formula to an
with Ed the story of how he gave me the challenge to learn how to
When I found momentum worked best on RSI, I was incredibly
ownership of the development of this oscillator, Manning shared
So the next time I flew down to his office, I did the programming
ods were effective, but using momentum worked best with RSI.
code it into TeleTrac, because he was extremely computer phobic.
spread differential will move in an unrestricted range. Both meth-
at a new formula he had developed. It was mine. He was unable to
the original formula. When you detrend two moving averages, the
months later he sent the same formula back to me to take a look
add the spread of a very short and semishort moving average to
ego needed the assurance I had done something of interest. Several
less range, without cluttering the screen, add momentum or just
ing to obtain approval from a mentor as I looked up to him and my
To change a normalized formula so it can flow freely in a limit-
was never in question.
oscillator in Figures 8.10 to 8.13. Then you might ask, 'Do markets
roots of an ownership dispute that emerged years later. It, however,
in the DJIA? We looked at this application of a simple detrended
on his computer and went back to New York. But I left behind the
during the first leg down in the 1929 crash versus the 2008 decline
we first sent it to him. I programmed several other setups he wanted
would like to know where an oscillator was capable of extending to
to add the formula to his TeleTrac. I said he was a turkey because
Why do you want to change the range RSI can move? Maybe you
be a royal pain in the assets. But it is something to consider if
the divergence signals compare to Elliott wave third wave endings
important nearly 20 years later. So intellectual property rights can
oscillator can travel? (No.) Is it the same for every market? How do
arise. Therefore this is a long process, but in this example becomes
Are the most significant signals always at the extreme ranges the
delay in this timing is to ensure no other applications or claimants
Do oscillator highs form sharp peaks but complex W patterns at lows?
cations for five years before they even begin to process them. The
side.) How does an oscillator compare in the DJIA at historic tops?
Library of Congress (in Washington) holds formula copyright appli-
giant gap down because all the panic executions exhaust the down-
on this formula in my name. I have had it for a very long time. The
the start of World War I? (It is a bottom when they reopen on a
posite Index, here's news; there is a Library of Congress Copyright
when the exchanges are closed for an historic period of time like
a copyright. As example, to the individual trying to steal the Com-
charts? Maybe you would like to see what happens to an oscillator
does not need to be. If you develop a new formula you can obtain
established? W i l l an oscillator test a prior extreme made in intraday
Intellectual property has become ugly in this business. But it
test prior extremes?' (Yes.) What happens when new extremes are
374
• IMPROVING INDICATOR TIMING AND F I L T E R I N G PREMATURE SIGNALS
The Composite Index • 3 7 5
trade. I t was a pattern i n the b e g i n n i n g that steady equity g r o w t h
years later.
o n . I was always overleveraged because I was an analyst l e a r n i n g to
f o r m u l a becomes popular that you may need to defend your work
courage and incredible confidence about t h e methods you lean
you find a f o r m u l a that solves a problem. You never k n o w i f your
market should not go as you do i n the forecast itself
investors heard 1 was w r i t i n g a book, he was very upset and threat-
exposure. T h a t means spend as m u c h t i m e calculating where the
six months o l d at that t i m e . W h e n m y largest and most i n f l u e n t i a l
dive i n and expect to excel u n t i l you k n o w h o w to manage the risk
it d u r i n g the e d i t i n g process. M y fledgling hedge f u n d was about
are learned just as indicator patterns are studied. You just cannot
ter was first w r i t t e n w i t h the f o r m u l a clearly disclosed. 1 removed
w o u l d be h i t hard by a single mistake. Risk management techniques
W h y d i d 1 not release the f o r m u l a i n the first edition? T h e chap-
mass psychology is slower to catch o n and move the market. O u r
maturity.
cash flow changes approaching and then j u m p i n too early because
do a n d make i t y o u r passion. P e r i o d . These things c o m e w i t h
t h a n we w i l l see t h e m unfold i n real life. W e often understand the
f o r m u l a to attract investors . . . o n l y results. Be good at w h a t y o u
It was also experience that taught me we see things faster i n charts
Holy Grail. A F u n d does n o t need t h e mystique o f a m e t h o d or
was m a t u r i t y that taught me not to be i n the markets all the t i m e .
not k n o w t h e n , what I k n o w to be true now. N o formula is the
needed permission to execute at a Fibonacci price target. But i t
M o s t were just angered. For this I a m sorry, it was w r o n g , but I d i d
Oscillators added t o t h e probability o f b e i n g right and gave t h e
Some f o u n d the clues and emailed m e to c o n f i r m they were right.
correct leverage, and where I was w r o n g before I was stopped out.
p u l l e d the f o r m u l a f r o m the book and clues were inserted instead.
w h e n a move w o u l d happen. Fibonacci taught m e price targets,
w o u l d have been finished before 1 had a chance to b e g i n . So I
i n g and t h e size o f the move. G a n n analysis gave m e a sense o f
out o f the F u n d i t w o u l d have caused a stampede for the door. I
T h e E l l i o t t Wave Principle gave m e a sense o f what was c o m -
ened to leave i f I revealed m y methods fully. I f this investor p u l l e d
the way.
tially terrified to fail h i m and choked. T h e industry knew M r . T u l l i s
But there is no question the lessons come w i t h a few scars along
left m y f u n d that I began to take o f f and do well. Maybe 1 was par-
w i l l contribute to a l o n g trading or analyst career for you as well.
agement that was given to me. But it was not u n t i l after E l i Tullis
then beginners fail to believe what they see. Realizing these things
for all they taught m e about markets and the support and encour-
indicators do not lag; our need for c o n f i r m a t i o n causes the lag, and
I regret deeply 1 c o u l d never pay back these particular investors
gave Paul Tudor Jones his start. T h e opportunity presented to m e
L e a r n i n g each m e t h o d took t i m e a n d y o u m u s t understand
to j u m p into t h e m i d d l e o f a strong t r e n d w h e n I began o n m y
because I had no price projection m e t h o d of confidence. You need
scared out of positions too early. I d i d not have the heart of a l i o n
nowhere. I t was partially inexperience and extreme over t r a d i n g
out the correct risk management skills I w o u l d overtrade and get
differential was filled w i t h fear. M y P/L w o u l d just go sideways to
that t r a d i n g a n d analysis skills are very different animals. W i t h -
versus t h e knowledge level I had was so mismatched the spread
376
• IMPROVING INDICATOR TIMING AND F I L T E R I N G PREMATURE SIGNALS
The Composite Index • 3 7 7
the formula:
mulas. The detrended oscillator w i l l set the horizontal level of
indicator will require two functions to create the formula. Here is
detrended moving average and don't look at the normahzed for-
range movement described in the first chapter. In TradeStation the
own. Jumping into a strong trend is easy. Switch to a an intraday
resistance for you to sell or buy within the strong move. I w i l l forever be in your debt, Mr. TuUis, because you were kind when
TradeStation:
Plot3(average((plotl),33),'Plot3');
thrust me forward into a career that is now a quarter century old
Plot2(average((plotI),13),'Plot2');
You toughened me up and gave me a kick i n the behind that
PIotl(RSIM09 + RSI3,'Plotl');
it was undeserving and mercilessly tough when it was needed.
RSI3 = AVERAGE(RSI (CLOSE,3),3)
have to know what's going on in the world. For traders, Mr. Tullis
The second function is written:
thing about technical analysis; the practitioner rarely retires. We
RSIM09 = MOMENTUM(RSI (CLOSE,14),9)
quickly, yet the passion for markets has not dimmed. That is the
The function RSIM09 is written:
and still going strong. It is hard to believe the time has passed so
said it best: 'A trader will always be a trader as long as they have $100 in their pocket.'
I have passed the formula on i n eSignal, C Q G , and MetaStock
a standard menu item, they will do so i f clients show a need. The
of the Composite Index story.
Bloomberg Press. You must ask your software vendor to add it as
information to keep you busy for a very long time. Now, for the rest
book Breakthroughs in Technical Analysis, David Keller Editor,
to describe. But it also means you are about to be given enough
But it has been in the public domain now since the release of the
reservations about. That gives me a balance and a freedom hard
it has been custom programmed for institutions using Bloomberg.
little to prove for the sake of ego after 25 years, and nothing to have
(see Appendix A) formats to the vendors and many users. I know
As a private trader there is no reason to hold back on you now,
I gave it the name the Composite Index because it was in fact
formula is now a standard free offering in Market Analyst.
limitations with Bloomberg.
can detect a trend change when the standard RSI formula alone
Fibonacci Analysis to see how these are created due to contract
is currently relative to n-pcriods ago. Therefore, the modified RSI
charts will have price projection boxes. Please refer to my book
easier to read. The Composite Index is able to compare where it
methods on the screen are the methods I use most often. Several
cators within indicator formulas to try to make the display much
them near August 17, 2011. This is useful to know because the
me how to graph indicators on indicators. I began to imbed indi-
The charts in this chapter were working charts as I was using
the composite of two existing indicators. Manning Stoller taught
and weekly time horizons. I have added a small line over the
because the Composite Index does not give the characteristics of
Figure 12.1 shows two charts of the Cash S&P500 in monthly
fails to develop divergence. However, it does not replace the RSI
378 • IMPROVING INDICATOR TIMING AND FILTERING PREMATURE SIGNALS
The Composite Index • 379
oscillators to help you see the pattern that was the reason for needing the Composite Index at first. In the monthly chart on the right the RSI shows no divergence with the price data. The Composite Index is diverging. In the weekly chart on the right, the RSI fails to form divergence again relative to price. It is the Composite Index that diverges. In fact the RSI forms a new momentum peak high while the Composite diverges. But the signal that forms in RSI becomes extremely strong and bearish when it swings up toward the 57 level which is where its moving averages cross over. It is important to notice that the fast-moving average (shorter period) is crossing down through the slower average. Often at this very juncture I find a sharp move follows in the market. In this case it was a serious break. Always be aware of the length of the longest bar within the trend. This is a Gann statement I have found to be true. When the length of the longest bar is broken in a trend reversal, the old trend is finished. It doesn't mean a deep retracement cannot happen or even a slight new high may follow, but for purposes of trading and trend analysis, the prior trend is finished. Another important Gann statement is that the extremes that mark the end of a swing are the end of the swing, but not the beginning of the new trend. The trend begins from the first secondary retracement of the new trend. This is from Gann's stock course. It is important for price projection methods to know this because it changes where the start is located. So Figure 12.1 shows two charts with the Composite diverging with price. Is that reason to sell? Not if that is all you have. But these signals are developing into a price target that can only be seen in the monthly chart. The line at 1,361 is a Fibonacci confluence zone, or major objective. The lines in the weekly chart that seem to
3 8 0 • IMPROVING INDICATOR TIMING AND F I L T E R I N G PREMATURE SIGNALS
The Composite Index • 3 8 1
cut through the swings at the high are the starting ranges defined for Fibonacci support targets. The ranges all start from the origin of a third wave decline. So price target comes first and permission always comes second from the oscillators. Figure 12.2 is a close-up of the same weekly Cash S&P500 chart. This chart has vertical lines so it is easier to match indicator peaks with price. This is a very common setup that occurs i n any time horizon. Divergence occurs first in the Composite Index. A second divergence occurs in the Composite Index. Two divergence signals in the Composite are rare. Usually one will do. But the final price high rarely occurs at the divergence signal. It may end the fifth wave of wave iii of 3. The second momentum swing up will often top wave V of 3. Then a final momentum push into the underside of the averages tops wave 5. The breakdown begins and then it is the oscillator pushing up into the crossover of the averages, or a negative spread that aligns with a third wave decline. Both oscillators provide information that I consider to be extremely important. Notice the horizontal lines drawn through the oscillators. The double line in the Composite Index is more of a narrow band rather than a precise line. As you monitor the Composite Index at this resistance zone the indicator is giving you a probability. Scan across the chart from right to left. Every oscillator pivot marked with a carrot or A symbol is at that level of resistance/support that the market has respected i n the past. Every market swing has a reversal at this level of displacement in the Composite. It is a highly probable scenario that it will happen again when there is a double test at the same zone. There is no way of telling how much of a reaction. That comes from other methods.
382
• IMPROVING INDICATOR TIMING AND F I L T E R I N G PREMATURE SIGNALS
The Composite Index • 3 8 3
occurs, but the trigger is from the oscillator.
tal level each time as well. What if there is no history to compare
daily chart on the right it is a clear illustration of how a target zone
become support. The market respects the indicator at this horizon-
confluence zones for objectives is important, and by adding the
In the RSI the two A and v signals show how resistance can
German D A X Index. The oscillator divergence in the monthly
odds are pretty good it will run you over again!
Figure 12.3 is the monthly, weekly, and 88-minute chart of the
work within this specific market. If it doesn't respect it in the past,
your probability of being right, you must look at markets offshore.
out how an oscillator tells us something more about how it wants to
Cash SP500 weekly chart in Figure 12.2. I f you want to increase
and the market fails to respect it most of the time? You just found
Figure 12.3 was screen-captured at the same time as the
to? It is not a trading signal. What if there is a history at this level
higher probability?
1,144. I have just photo-captured the screen in a daily time hori-
build wave structure. What else can we use to help build a case of
topped at 1,208.47. It is in a fast market condition and currently
is only wave a of iv. That is why I use longer horizon data to help
a position keeps nagging me at the corner of my eye. The market
than wave 2. The momentum top just before wave iv to the left
S&P500 continues onward to the 1,208 target. As I am writing this
the same degree and it is extending. Wave iv tops just a tad higher
lows in German markets. (We are about to discuss this.) The Cash
wave decline. The amplitudes at 2' and 4' 4 warn they are within
sell and there was an incomplete Elliott Wave pattern into those
are second and fourth degree retracements i n an extending third
the market arrived at 1,188 the oscillators did give permission to
88-minute chart are two Composite pivots marked 2' and 4'. These
retracements from the 1,101 low were 1,188.39 and 1,208.21. When
screen that leans on the longer horizon views for trend. I n the
was 1,101.54. A rally then developed toward 1,208.47. The two target
the market fell off the highs. The 88-minute chart is the trading
ence zone at 1,101.04. The actual swing low that began from 1,347
marks resistance. The market fell to the target zone drawn before
zone. In the chart is a line that marks the top of the target conflu-
how the Composite Index pushes up into a horizontal line that
down in the weekly Cash S&P chart falls to a Fibonacci confluence
chart is the same as the Cash SP500. The weekly chart shows
Before moving on, let me point out something. The price melt-
exchanges. They went out two years further than what was posted
and the fifth wave down is unraveling now. Developing Fibonacci
of 2008 I was checking the open interest in futures from the
the oscillator. The nearby high into the 1,208.21 target is wave 4 up
sends me crazy in O i l . But the real reason is that in November
the pivot marked as the top of wave 2 on price. It is marked; 2nd; on
2011. I only trade the December contract. The monthly rollover
The Composite Index has topped near the same displacement as
Figure 12.4 shows the month, week, and 240-minute December
charts. It just became a whole lot more interesting and profitable.
will return to a series captured three days ago at the same time.)
chart shows 1,144. There is a three-day separation between the two
to the equity swings. O i l is a good one to use. (Now all the charts
chart will show a different last bar price of 1,178 and this daily
Use a market that is not an equity index, but has a relationship
zon and will add it to Figure 12.2. But it explains why the weekly
TradeStation . . . i s | j i -
Figure 12.3 Source:
_
H X
'
TradeStation..
TradeStation . . .
A e r o d y n a m i c I n v e s t m e n t s Inc., © 1 9 9 6 - 2 0 1 1 , D a l l y M a r k e t R e p o r t , w w w . a e r o i n v e s t . c o m
TradeStation. © TradeStation Technologies
Figure 12.4 A e r o d y n a m i c I n v e s t m e n t s Inc., © 1 9 9 6 - 2 0 1 1 , D a i l y M a r k e t R e p o r t , w w w . a e r o i n v e s t . c o m Source: TradeStation. © TradeStation Technologies
386 • IMPROVING INDICATOR TIMING AND FILTERING PREMATURE SIGNALS
The Composite Index • 387
best is the simplest.
The only player in town would be China taking on a triple hedge.
average that has been detrended within its formula. Sometimes the
the front contract for 2008.1 knew it had to be government activity.
have to scale in. Every oscillator in the world is based on a moving
shock. The open interest six years out for 2012 was equal to that in
time. It will also become my answer in flash crash scenarios when I
by most professional data services. What I saw at that time was a
any market, or time horizon.
ever so slowly. I've traded December ever since and my poor dys-
be catching this one pattern as the easiest to detect in the chart of
and Bonds. It was too good not to be them, and the Chinese move
and RSI into the price low marked wave v of 3? By now you should
inflation. It was a hedge on the freefall valuation of the U.S. Dollar
Do I need to point out the divergence between the Composite
They moved the Oil market so it hedged future purchases against
lexic brain doesn't have to worry about delivery dates any more.
U.S. Treasury Bonds are markets to always be aware of whether
is an estimate, but better than simple trend lines to create channels.
always be revisited and the market reacts quickly to such tests.
diagonal without all the details demonstrated in the Gann chapter. It
by the oscillator. When new extremes form, the old range will nearly
same slope. You are actually starting to learn Gann analysis on the
lator high. Sell it. That is a test up to the previous range established
bars or gaps for the first to set the angle and create parallel lines of the
equal to the displacement one peak behind the most extreme oscil-
ated by a method described much cadier in this book, using strong
The Composite oscillator aligned with wave C is at a displacement
axis within RSI. Price is testing a channel. The channels were cre-
historic market high. The weekly chart is a more advanced signal.
where the asterisk shows the market testing support on the horizontal
as the oscillator is at the same displacement as accompanied the
These are signals that follow the strong signal in the monthly chart
The Composite signal in the monthly chart should be easy to read
tive spread, not as clear in the RSI when the prior range is tested.
viding multiple ranges with the same start near the low marked B.
spread between the moving averages on the Composite have a posi-
115.48 is a cap of a confluence Fibonacci zone derived from subdi-
that could be missed, but the value in the Composite is the fact the
made back in 2007. It is still on my screen. The weekly chart shows
low behind it. The Composite has a slight upward slope to this signal
The monthly chart shows 115.56 as a mid-point in a swing projection
tor low in the RSI swings up and then RSI tests the prior oscillator
The 115 price target shows up in both the monthly and weekly charts.
charts are very strong buy signals. In the daily chart the July oscilla-
moves are so big I trade short horizon against my monthly charts.
240-minute time horizons. The asterisks in the monthly and daily
monthly charts? Remember the flash crash in the DJIA? These
you trade them or not. Figure 12.5 shows the monthly, daily, and
Why do I use such short intraday charts against weekly and
tor configuration. Figure 12.6 shows the Cash S&P500 in a weekly
see it tops under a resistance line that has been around for a long
to return to the Cash SP500 charts with a slightly different oscilla-
from price data. It is my lifeline at points that lack clarity. You can
several markets developing a strong signal at the same time, I need
referenced before. This is a plain Jane detrended moving average
Now that you have a sense of how the S&P500 was only one of
There is an indicator in the bottom of the weekly chart I have
TradeStation .. Is-ili-l
:_:,n'X, ' TradeStation.. | s - | l i - |
jO
xr
TradeStation..
|S-|li-|
;jn:;X
Figure 12.5 A e r o d y n a m i c I n v e s t m e n t s Inc., © 1 9 9 6 - 2 0 1 1 , Daily IVIarket Report, w w w . a e r o i n v e s t . c o m Source: TradeStation. © TradeStation Technologies
TradeStation Chart..
[sj^ij
PINX - Weakly U1178.81 0.00 -0.00% B : l ..
ftT**^^
Trfn r'™'! ® 1 9 9 6 - 2 0 1 1 , Daily M a r k e t Report, w w w . a e r o i n v e s t . c o m Source: TradeStation. ©TradeStation Technologies
390
• IMPROVING INDICATOR T I M I N G AND F I L T E R I N G P R E M A T U R E SIGNALS
The Composite Index • 3 9 1
and daily chart w i t h the Composite Index and detrended simple m o v i n g average. I t is i n the simple detrended m o v i n g average that I need to make a few points. T h e weekly chart shows two arrows under the oscillator. T h e extreme move to the right has pressed the oscillator to the same extreme that developed at the price lows i n early 2009. T h i s is i m p o r t a n t . A reaction should be expected f r o m such an alignment. You w i l l notice a horizontal line does not mark the 2008 m o m e n t u m low. T h a t is because it was not the historic m o m e n t u m low. I always mark these extremes because w h e n they come back into frame they are i m p o r t a n t considerations. T h e nearby reaction i n the daily chart was a significant move regardless o f the t i m e frame you favor. W h a t other markets should we be t r a c k i n g and always aware of? C h i n a . But what i f your software vendor does not provide the Shanghai Index? N o excuse. There is always a way. I n Figure 12.7 are three Funds i n the m o n t h l y charts. F r o m left to right they are the Templeton Dragon F u n d , the Fidelity C h i n a Region, and the Templeton C h i n a W o r l d . Use f u n d N e t Asset Values ( N A V ) as an index. You can also use exchange traded funds (ETFs). T h e symbol for C h i n a is F X l . A l l are possible alternatives to m o n i t o r i n g other markets w h e n it m i g h t be d i f f i c u l t for you to obtain the actual exchange data. T h e divergence signals and resistance levels are all valid. I n all three funds, what level is the RSI b e g i n n i n g to rollover f r o m i n these charts? W h a t is the level that defines a retracement i n a bear market? I f you are unsure, go back to scan Chapter 1. Figure 12.8 is the screen layout I favor for Asian indexes. F r o m left to right are the equity indexes for C h i n a , H o n g K o n g , a n d Japan. A l l three indexes were i n trouble before the break i n N o r t h
The Composite Index • 393
American indexes. Then Europe was ahead of North America. I am always surprised so many traders still do not watch any other market than their own. Sometimes you can see a blue-print before it happens and still people do not want to watch a market they cannot directly trade. The Shanghai Composite in Figure 12.8 is data obtained from Commodity Systems Inc. (CSI). TradeStation allows third party data again so I download at night to update all the charts with a single button. I have no need for intraday data. The charts in Figure 12.8 are working screens. It shows actual targets, wave counts in development and oscillator levels of interest that were defined from action in the oscillators years ago. Old extremes do repeat in future data. For example, the weekly China Shanghai Composite in the far left of Figure 12.8 has a remnant arrow still on the chart on the far left Composite Index. That arrow marks an old buy signal and the swing that followed. Notice the arrow above the same horizontal line in the more recent data. The same level marks a significant failure in prices. The Nikkei on the far right contains the most subtle. There is a horizontal dotted line into the chart high. That same oscillator displacement marks the start of a decline in the recent data. Don't always look for big signals. It is perhaps the most common problem I encounter when coaching traders; you don't look at the internals. Often it's not about the big picture. Figure 12.9 shows charts for six Funds, all Corporate Debt BBB. These charts are going to be the next crash and I wanted to understand more than just what was in these images. When I started to research some background information I found many are caught in a corner trying to obtain gains in an environment of zero short-term rates. So what are they doing to boost returns? They are buying as
394
• IMPROVING INDICATOR TIMING AND F I L T E R I N G P R E M A I U R E SIGNALS
The Composite Index • 395
3-dimensional chess ahead of the masses. The seed of the idea came from something I read. It was Gann's idea and I read it in his 1941 edition of How to make Profits Trading in Commodities. You won't find it in the second edition. But the markets see this coming now and it explains why there is a stampede rush into Swiss Francs. The first edition of Gann's book is so rare that typing errors are corrected by hand on correction tape. In this edition he freely talks about the growing war in Europe. Do you recall what was written in Chapter 2 when cycles were the focus of discussion? The entire reference can be found in that chapter regarding World War II. It was before Pearl Harbor and the United States was not in the war when his book was released. The section that makes us think about current stresses on the Euro reads: This war is no different than any other war. It is being fought for commercial supremacy to determine WHO is to dominate the comFigure 12.9 Aerodynamic Investments Inc., ® 1996-2011, Daily Market Report, www.aeroinvest.com Source: TradeStation. © TradeStation Technologies
merce of the worid in the future. Germany wants to dictate prices and dominate the commerce of the wodd, and England wishes to do the same. Russia would like also to do the same thing.
much as 50 percent Foreign Denominated Bonds! That's a ticking time-bomb. Suddenly you understand why the next chart follows as I move into currencies. Anyone who thinks a stock trader need not track currencies is in for one rude awakening down the road. The EURUSD is likely to break apart, as most believe Germany will not bail out Italy. However, I had suggested an uncoupling nearly five years ago when it was unheard of at that time because fundamentals could not begin to point to such a problenL But I wasn't trying to think through global cash flow issues and some wizard playing high-level
I read this paragraph and see history is repeating itself once again. Germany tried to dominate commerce through Europe with the Euro currency and failed. Now that it has failed they will want out. I think it is just a question of when and how, rather than 'if' So what will happen to all those funds with Foreign Denominated Bonds and exposure to the Euro coming unglued? Right. They are toast. The collapse will be a huge shockwave in the United States because people think they are investing in American Dollar denominated corporate bonds. Many people looking for returns on
396
• IMPROVING INDICATOR TIMING AND F I L T E R I N G PREMATURE SIGNALS
The Composite Index • 3 9 7
BBB rated funds will likely be living on fixed incomes and they are just trying to keep out of trouble and find something better than a zero return with rates so low. As you read this, some may think, how is this related to technical analysis? Technical analysis is the means to see what is approaching in the future. We are problem solvers, slayers of dragons, and sleuth hunters of risk exposure others never saw coming. We are really good at it too. But being a master of technical indicators is not the end result we seek. Technical analysis methods become our laser swords of choice. They should be the means to thinking through how events could unfold. It is not about becoming the fastest detective in the world to recognize a contracting triangle! No. Don't lose sight of why we do this. Constantly ask what will follow. I ask who is going to do what to whom?! I just want to be on the strongest side of the rope being pulled. Figure 12.10 is the daily USDCHF chart. The dollar has been in a nose dive while the Swiss Franc is rallying. Spot currency charts are read as the first currency in the name is the trend. So the U.S. Dollar is in freefall as the Swiss Franc strengthens. The daily USDCHF chart has seven vertical lines to help you examine the oscillator positions relative to the price data. The two
Figure 12.10 Aerodynamic Investments Inc., © 1996-2011, Daily Marltet Report, www.aeroinvest.com Source: TradeStation. © TradeStation Technologies
oscillators are the Composite and 14-period RSI. This gives me the
has just completed an Elliott wave corrective flat pattern. The
forms a double bottom. Who cares? That's old news. It worked, but
If the market were at line 7, what could be told? The price data
and state divergence is present in the Composite while the RSI
us nothing. If a test is coming the downtrend is still incomplete.
extreme line 6. Most analysts will start at the bottom of this swing
low after the vertical line in April. The divergence, however, tells
Index to a level equal in displacement to the same oscillator at
low in this chart. The prior range will be tested that is the trough
recent swing up. The dollar has rallied, pushing the Composite
bottom to the far right in the RSI made a new oscillator extreme
Read charts right to left. The first detail of interest is in the most
can it tell us something about the market now? YES! The double
opportunity to explain the use of these oscillators in a strong trend.
398
• IMPROVING INDICATOR T I M I N G AND F I L T E R I N G P R E M A T U R E SIGNALS
The Composite Index • 3 9 9
say 'go' on line 4.
ing. M o v e f u r t h e r left to l i n e 6. W h i l e it m i g h t be a new oscillator
swing up into the rectangle b o t h t i m e horizons w o u l d switch to
line of leftover support line. Either way, the market is back and fail-
one was not present i n the shorter h o r i z o n . T h e n w i t h the second
four upended. Price at hne 7 is testing the underside of a resistance
T h i s is a 'no g o ' i n this t i m e h o r i z o n . It does not mean, however,
Composite at l i n e 7 was tested just to the left of it where wave a of
side o f its o w n m o v i n g averages. ( T h i n k very small pivot.)
tal line of resistance. Sell first and ask questions later. T h e failure
pivot i n RSI just to the right o f line 2 that presses up into the under-
pullback and t h e n fails on a small push back up to its o w n h o r i z o n -
tance line w h i l e RSI is caught b l i n d . RSI finally clues i n w i t h the
line 6 is the same displacement as seen at line 3. RSI makes a small
L i n e 2 is a strong sell. T h e Composite is diverging under a resis-
h i g h , it is a corrective zigzag rally. T h e peak i n the Composite at
not change. T h e trend was expected to continue because the price
become overbought at lower price levels.
pointed to one of the setups o f f a low. But the cycle of analysis does
between the peaks o n lines 5 and 6. T h i s means the market has
to execute into those dates. There are other signals because I only
i n g negative reversal signals relative to the oscillator peaks that fall
are G a n n cycle time/date targets. I was watching i f I had permission
RSI are c o n f i r m i n g the market is weak. B o t h indicators are f o r m -
along this amplitude displacement. Each of the lines 1 t h r o u g h 7
h i g h 6 cannot be exceeded. M o v e to line 5. T h e Composite and
Finally, line 1—a l i n e i n view because o f a prior track record
i n the RSI just to the right of line 6 offers a lower risk entry as price
M o v e over to line 5. T h e RSI is back to the level that marked line
objectives had not been realized.
the M i d d l e East. However, the DJIA was lagging all t h r o u g h 2007.
shows no f o r m a t i o n to b o t t o m .
tant. C h i n a s cash flowed into India, South A f r i c a , Australia, and
between lines 5 and 6. It is a double b o t t o m relative to the RSI that
global leader. N o t currently, b u t the correlation remains impor-
nals. N o t i c e the oscillator lows i n the Composite Index that f o r m
Australia was cleady the benefactor o f China's b o o m and it was the
and 5. I have no interest i n the price projections f r o m these sig-
interest to the Shanghai Composite is the Australian market. W h y ?
5 and the oscillator peaks just to the left that fall between lines 4
i n Figure 12.11.1 align these three markets together, but the greatest
negative reversal patterns between the oscillator extremes o n line
There is a tremendous a m o u n t of information i n the three charts
1. T h e Composite is under resistance. B o t h oscillators are creating
b i r t h o f Fort K n o x . W e held 5 b i l l i o n o f a total 9 b i l l i o n i n the
at l i n e 4. T h e RSI at l i n e 3 has no history at this displacement.
the U n i t e d States demanded payments i n gold b u l l i o n . It was the
at a new h i g h and the market w i l l weaken, b u t return as we saw
History repeats itself differently each t i m e . After W o r i d War I
L i n e 3 shows oscillators i n a h i g h risk scenario. T h e Composite is
C h i n a would change the entire direction of the crisis approaching.
used it to show customers and it was the entry level for this trade.
is c o m i n g f r o m Europe, b u t d o n ' t fall asleep because a crack i n
correction fits i n a rectangular box shape i n the price data. I had
question was where w o u l d the trigger come from? So m u c h news
M pattern a n d it completes an E l l i o t t wave flat correction. T h e
T h e economic war started l o n g before the 2008 m e l t d o w n . T h e
M o v e over to l i n e 4. T h e Composite o n l i n e 4 shows a small
Chapter THE PRINCIPLES OF DEPTH P E R S P E C T I V E APPLIED TO TWODIMENSIONAL CHARTING
M
y first job out of school was working for Kodak Canada. I then transferred to Eastman Kodak working as a professional
photographer and then as a sales representative and on to corporate life in Rochester. I have a secret. I am a Master Photographer and was an apprentice of Ansel Adams. As Kodak struggles on the brink of bankruptcy I am reminded of all that the company taught me and the opportunities it offered. I left Kodak when a hard fought battle was lost not to introduce digital photography before Japan. Traditional films were just too profitable and they shelved the digital capabilities that would have saved them from the near bankruptcy position they find themselves to be in now. It is sad what has happened to great company. But they paid for my life changing experience of 12 weeks with one of the greatest
403
404
• IMPROVING INDICATOR TIMING AND F I L T E R I N G PREMATURE SIGNALS
The Principles of Depth Perspective Applied to Two-Dimensional Charting • 4 0 5
as they relate to chart analysis, it is necessary to understand how we
and lines of insignificance.
omitted in our charts. To understand depth perception problems
beginner sees a silhouette. The artist sees lines of varying strength
subtle cues to create a solution for the missing plane that has been
living form trains you to see energy lines and momentum. The
is called the median plane. Our minds will unconsciously look for
between objects and from front or behind. A class for drawing a
reconstruct and recover the third dimension, which is depth. Depth
Drawing a group of objects will train you to see space proportions
our environment in two dimensions. Our brain will always try to
training the right side of the brain is just as important as the left.
a chart? Anatomically we can see only a visual representation of
usually suggest they take an art class. Sounds odd, doesn't it? But
dimensional, but how does this fact change our ability to interpret
traders are struggling to see balance and proportion in charts I
factor when we look at a two-dimensional chart? The wodd is three
ence on nearly everything else in my life that followed. Now when
Why is depth perception so difficult, and why does it become a
photography masters in our history. That experience had an influ-
An area of research that Kodak spent millions on is to onderstand
are able to see depth at all.
to see—to see correctly. That means a good deal more than merely
based on research dollars spent.
you see it. Learning to analyze a chart is really a matter of learning
We really do see differently. So don't be offended; I am being factual
that all six of these informational cues may have an impact on how
to do the tasks that developed the right side of the brain more often.
angle. When you look at a chart, you will be very surprised to know
hyper-focus and strength so they could catch the food! Women had
elevation, interposition, clarity and aerial perspective, and visual
ers and trappers who brought in the food. They developed skills of
Some of these depth cues are linear perspective, texture gradient,
Why? Men didn't need these skills because they were the hunt-
able to us. The subtleties our mind searches for are depth cues.
that men in general do not see depth perception or proportion well.
Generally, there is a great deal of information about depth avail-
how men and women see the world differently. They discovered
principles will have an impact on traders.
of men tested in the 1970s were nearly blind to the blue spectrum.
and then look at a few chart formations to demonstrate how these
Munsell Color Test. Men cannot see blues. Period. Over 70 percent
looking with the eye. We will look at all six depth perception types,
Every employee from groundskeeper to executive was given a
So don't create advertisements that are predominantly in the blue color spectrum. Most colorblind issues are found in men. I don't know why. Why did Kodak need to know? We used this knowl-
Linear Perspective
an observer. The greater the distance, the closer the lines will appear
billboard image selection and much more.
parallel, such as railway tracks, converge upward as they recede from
target audience. The issues about depth perception would impact
familiar. It is common knowledge that lines or edges that are truly
Color Films and I used images that avoided weaknesses within our
Linear perspective is perhaps the one with which you are most
edge in our advertising. I was the brand manager of Professional
406
• IMPROVING INDICATOR T I M I N G AND F I L T E R I N G P R E M A T U R E SIGNALS
The Principles of Depth Perspective Applied to Two-Dimensional Charting
• 407
until they merge as one into the distance. The mathematical system for creating the illusion of space and distance on a flat surface origi-
One Point Perspective r14200
nated in Florence, Italy, in the early 1400s. It was an architect, Leon
•13400
Battista Alberti, who first wrote down the rules of linear perspective.
•12600
Technical analysts need to understand the mathematical components of linear perspective because these depth cues have a major influence on how we will see signals in a chart. The example of
•100.00

receding railway tracks is the simplest mathematical system called one-point perspective. When we look at charts, we will be applying two- or three-point perspective based on the chart patterns we are
iConvergence Lines
H
/I/
•80.00 •60.00 •40.00 •20.00
Horizon Line
viewing. So let's review one-point perspective to ensure that every-
i VP^^^*^^.
one understands the fundamentals.
•100.00 •80.00
Vanishing Point
-20.00
cookie or a two-dimensional chart, there will be a horizon line.
-40.00
Whatever you are looking at, a three-dimensional chocolate chip
•60.00
The basics of one-point perspective are illustrated in Figure 13.1.
Figure 13.1
or look out from a seashore while sitting on the beach, the hori-
'0-
If you look out at an ocean's horizon from the cockpit of a plane,
• 'N
• • b
'98
'F
V
'A • • 'M-
•
'j
'j- •
'A
zon line will always be at your eye level. This invisible plane cuts called convergence lines.
always be at eye level regardless of where you are looking. If you
be connected to suggest a line leading to the vanishing point are
through everything, including your chart! Your horizon line will look down on the trading desk, the horizon line does not change as your eye level remains at the height of your eyes.
In Figure 13.2 a Stochastics signal has been extracted from a larger chart. Above it is a moving average signal. Is the slope of the average
this visual cue to create a converging line to mark a vanishing point.
swings as points to define a vanishing point. All the points that can
vanishing point is located within this chart. As a result, our eye uses
the reference points for our mind to then try to use the indicator
frame, making it easier for us to unconsciously determine where the
a chart the vertical and horizontal lines of the chart grid establish
Stochastics peak adds a point that is closer to the edge of the chart's
at the vanishing point. Verticals and horizontals stay the same. In
first Stochastics oscillator peak as additional information. The new
moving away from or to a viewer seem to recede on the horizon
ing the averages alone? Why is this difficult? Our eye will use this
lel to the viewer recede. In one-point perspective only lines that are
with the Stochastics indicator more steep than the chart display-
The vanishing point is the point to which all lines that are paral-
4 0 8 • IMPROVING INDICATOR TIMING AND F I L T E R I N G PREMATURE SIGNALS
FXUSJYMU L-Daily
The Principles of Depth Perspective AppUed to Two-Dimensional Charting • 4 0 9
We know our brains are extremely complex computers. When we are presented with an abstract chart, our minds try to reference known variables. The placement of an object defines twopoint linear perspective. All we need is one corner and our mind will try to complete the shape. It is the angles of the object that define where the vanishing points are located. In Figure 13.2 our minds will assume it has located one of the angles for our abstract puzzle to define depth. The vanishing point leads to the right in the chart. Sometimes one vanishing point in two-point perspective will not be on the page. Our mind will approximate where the point falls beyond our chart dimensions. Distortions are minimized when our mind establishes where the horizon line is located. Some charts will have a horizon line defined. That actually adds to our problem as the visual cue becomes stronger. An example would be the zero line for a histogram. When the object we are looking at does not have a perpendicular or parallel point,
Figure 13.2
our mind's eye will find a 'corner' within the abstract from which to extrapolate a converging line to a vanishing point. That is what
lyzing within this chart.
Figure 13.2 becomes more important than the inner peak that is a
becomes an influential point to all other information we are ana-
edge of a chart toward the vanishing point. The Stochastics peak in
the first Stochastics peak that is not a trading signal but which
Orthogonal lines are visual rays that connect points from an outer
point. Therefore, Figure 13.2 has a dominant point predefined at
indicated as being filtered by a broken line for the Stochastics travel.
start of a convergence line in Figure 13.3 becomes the dominant
exceeds this convergence line. The peak is filtered out as noise and
is forming in Figure 13.2. Each of the corners that marks the
It does not matter that Stochastics forms a peak at a later time that
to a third vanishing point. The third vanishing point will always
ing one-point perspective is insufficient for this chart.
horizon line but there will be an upward or downward recession
was drawn in Figure B.l, shows that the mathematical system defin-
or lows. Not only will there be the two vanishing points along a
fact that the vanishing point leads to the far right, not to the center as
come into play that is normally used only with extreme heights
peak in Stochastics serves as the origin for an orthogonal line. The
Because we are dealing with an abstract image, a third point can
broken line because the first is closer to the outer frame. The first
4 1 0 • IMPROVING INDICATOR TIMING AND FILTERING PREMATURE SIGNALS
The Principles of Depth Perspective Applied to Two-Dimensional Charting • 4 1 1
Two Point Perspective •14200 •134GG •12600
observed angle
observed angle
W ' ' D ' ' ' 9 8 '
'F'
'M
'A
-'M-'J
'J
^
first and then allow that pattern to recede, allowing us to view the lighter pattern.
squiggles will provide points that lead to a projected third vanish-
Figure 13.4
Figure 13.3
be 90 degrees to the horizon line. Many complex technical chart ing point.
The chart in Figure 13.6 can visually trick you. RSI TrendZ
a filled histogram pattern. Be aware of the visual impact when you
to point out the similarities of distortion.
The only difference is that different areas have been selected to plot
his work Sunmoon in Figure 13.5 and then return to look at a chart
clearly that RSI Trends 2 and 3 are in fact the exact same signals.
master of three-point perspective traps is M.C. Escher. Consider
However, the indicator on the bottom of the chart shows you more
trap itself applies to three-point perspective (Figure 13.4). The
and RSI Trend3 appear to give different signals at points A and B.
It is easy to catch you visually in a depth perspective trap. The
processed as an image that has height and a third vanishing point.
the correct vanishing points. Generally we see the darker image
to horizontal, and everything else that follows in your chart will be
Our mind alternates the dominant image in an effort to determine
add a histogram. The vertical lines of a histogram are 90 degrees
In Escher's work, there is frequently a play on depth of field.
412
• IMPROVING INDICATOR T I M I N G AND F I L T E R I N G P R E M A T U R E
SIGNALS
The Principles of Depth Perspective Applied to Two-Dimensional Charting • 4 1 3
FXUSJYMUL-Daily 08/19/96 C=14485 +15
14800 14200 13600 13000 12400
90.00 60.00 30.00 RSI Trend ^ •4.00 -600 RSI Trend 4.00 E-6.00
Figure 13.5
Figure 13.6
closely packed together as d e p t h that has a n increasing rate o f
this indicator evaluation.
size. O u r mind's eye sees objects that become smaller and more
be used is Trend2, and we w i l l discard Trend3 w h e n we continue
pivot—to try to determine the rate at w h i c h t h e elements change
an orthogonal line of convergence. Therefore, the pattern that must
process any element—such as a dot (parabolic), indicator peak, or
nificant because it is closest to the outer frame of the chart and is also
case i n Figure 13.6. W h e n linear lines are absent, our m i n d w i l l
T h e histogram i n RSI TrendZ at p o i n t A becomes even more sig-
change. T h i s rate n o r m a l l y depends o n the slope o f the g r o u n d . W h e n there is no g r o u n d w i t h i n a chart, our eye may see the jux-
Texture Gradient
taposition of certain indicator patterns as depth cues. You w i l l not
clarification.
w h e n there are n o straight lines present i n a n image. T h i s is the
you to move your vertical l i n e over t o a peak or l o w for f u r t h e r
plane is called the texture gradient. T h i s cue becomes i m p o r t a n t
be caught by this p r o b l e m , b u t i t is likely to be one that motivated
A n o t h e r d e p t h cue t h a t we use to define o u r missing m e d i a n
414
• IMPROVING INDICATOR T I M I N G AND F I L T E R I N G P R E M A T U R E SIGNALS
The Principles of Depth Perspective Applied to Two-Dimensional Charting * 4 1 5
Elevation Objects that fall between us and our horizon w i l l normally appear higher up the farther they are f r o m us. T h e railway tracks receding into the vanishing point w i l l always move upward to the vanishing point. T h i s upward movement becomes a visual cue f r o m which our eye w i l l travel w i t h i n a chart. Elevation cues are a given fact and help to set up a visual problem when combined w i t h interposition cues.
Interposition N o r m a l l y texture appears to be more dense i n a distant object than i n an identical b u t closer object. However, some depth cues w i l l be stronger t h a n others. T h i s is the case w h e n one object obscures part of another. W e frequently obscure another indicator such as an average w h e n m u l t i p l e studies are used as an overlay i n a chart. T h e partly obscured object w i l l be viewed as farther away. However, your m i n d may again trick y o u , as illustrated i n Figure B.7. T h e farther object appears to be closer because the cutout gives us a pseudo-interposition. It looks as t h o u g h the smaller card is i n front of the larger w h e n , i n fact, the smaller is positioned b e h i n d the
Figure 13.7 than its counterpart o n the right. T h a t is because shadows are also depth perception cues. W h e n you draw a trend line on a chart, you can shift the proportional dimensions w i t h i n it. T h e y w i l l be relative i f the line is long enough across your chart. But short lines can contribute to visual errors or distortions.
Clarity and Aerial Perspective
ment of light's wavelength frequency spectrum.
boundary of the smaller card o n the left w i l l make it appear larger
i n a higher p r o p o r t i o n t h a n reds that fall w i t h i n the infrared seg-
the same. However, adding the w i d t h of one pixel line to the inside
frequencies f r o m ultraviolet to blue are scattered i n our atmosphere
on the left f r o m the image on the right, so all four cards are i n fact
the fact that far objects appear bluish to us because the wavelength
these t w o smaller cards are identical i n size. I created the image
aerial perspective does play a part i n chart analysis. T h i s refers to
your precision ratio compass or proportional divider—to prove that
become a factor i n a t w o - d i m e n s i o n a l chart, but its near cousin
b o t h illustrations? I suspect you w i l l have to use your PRC—that is,
us than o n identical objects that are up close. T h i s alone does not
you tell i f the card that is actually at the back is the same size i n
It is harder to see the details on objects that are farther away f r o m
larger card w i t h a cutout. Before m o v i n g away f r o m Figure 13.7, can
416
• IMPROVING INDICATOR TIMING AND F I L T E R I N G PREMATURE SIGNALS
Because we are used to associating blues as distant, these cues can be used to our advantage or disadvantage in a chart. TradeStation
The Principles of Depth Perspective Applied to Two-Dimemional ChartiriP
•
417
your nose or sitting at the end of a table, the size of the object in terms of visual angle is independent of distance.
get creative, and that is when distortions occur.
charting capabilities. It is essential that the technical information
derived from distance cues, our interpretive abilities may begin to
(color intensity) need to wake up and move out of the dark ages for
visual angle. As it knows it must define the missing plane of depth
that offer only colors of equal intensity (brightness) and saturation
chart is that frequently the only cue our brain has to work with is
the recessive one when viewing the technical study. The vendors
size is referred to as size constancy. How this relates to analyzing a
impact of the color you use for your signal line. It should not be
object can look the same size regardless of changing retinal image
ately fall a perceived distance behind price. Be very aware of the
on the back of our eye is changing all the time? The fact that an
data. Select a soft gray for a Fibonacci line and watch it immedi-
can judge the size of an object when the visual angle of an object
nacci lines do not have to compete for attention with your price
angle plus perceived distance can determine height. So how is it we
markings that we may add to any chart. Use them wisely. Fibo-
Figure 13.8 shows us mathematically why one degree of visual
offers 16 colors that traders can use for their drawing tools and
are separated horizontally, each eye processes different views of
can use force you to view needless noise.
nal disparity. The term refers to the fact that, because our eyes
ties. Get after your vendor to make changes if the only colors you
One of the depth cues of greatest interest to us as traders is reti-
on your chart be compatible with your depth perception capabili-
Visual Angle
V i s u a l Angle
This is the sixth depth perception cue that will have an impact on our ability to analyze a chart. This cue requires anatomical considerations and elementary physics to relate lens equations to something useful that we can apply. Visual angle is a depth cue affected by pupil size and lens power. Everyone sees differently, and that is why it is as important to know our personal biases as it is to know the indicator patterns and underlying formulas themselves. Visual angle refers to the size of the image on the retina. When
tan a = 1/57 = 0.0175 tan'^ 0.0175= 1°
we look at similar objects, the one farther away has a smaller image on the back of our eye. Whether your computer screen is touching
Figure 13.8
likely the eye you first wanted to cover is your weaker eye.
angles and integrates them into one three-dimensional image. Over
text. One eye will handle the task easily, the other will hurt. More
brain views two-dimensional views from slightly different visual
which eye of yours is stronger or weaker. Cover one eye and read
image that forms in our eyes when we look out the window. The
visual processor and thereby reduce error. It is easy to determine
depth. Our two-dimensional image of a chart is the same as the
the back of my right eye. The purpose is to capitalize on a stronger
ity between the two pictures on the back of our eye that gives us
covering less surface area than that which is being projected onto
any object. What does disparity doforus? It is the sHght dispar-
The Principles of Depth Perspective Applied to Two-Dimensional Charting • 419
418 • IMPROVING INDICATOR TIMING AND FILTERING PREMATURE SIGNALS
10 percent of humans are stereoblind. There are great differences
Knowing how we process visual information helps us understand work in the same way.
on other depth cues that are monocular. The aerial perspective we
why one trader may like one indicator over another although they
in our stereovision ability. If you are stereoblind, you will depend
price data. The bottom of the chart contains two histograms.
ish, will be monocular. Therefore, we have all developed different
Figure 13.10 is Stochastics with averages plotted under the
discussed, which makes distant objects appear less clear and bludepth cue dependencies that match our anatomical abilities. One of my eyes is stronger than the other, so I always offset my primary trading screen to the right. Figure 13.9 will show you why
08/19/98 C=14402 -68
making this change produces a smaller visual angle in my left eye.
Retinal Disparity
Figure 13.9
Figure 13.10
as only an isolated case in the chart. Signal 8 offers perfect timing
disguise its origin for just a short period of time.
there is information at this peak that we are missing when it is seen
The middle histogram with averages is called the XYZ Trend to
The Principles of Depth Perspective Applied to Two-Dimensional Charting • 421
420 • IMPROVING INDICATOR TIMING AND FILTERING PREMATURE SIGNALS
We are evaluating the combined signals in the XYZ Trend and the study called the RSI Trend as they compare to Stochastics.
between the RSI Trend signal and the Stochastics indicator as we have seen at other points in this chart.
above zero.
complex top in the other indicators. At signal 2 this is indeed what
a histogram. It is extremely noisy with all the long bars plotting
momentum to attempt another advance that would create a more
this idea to Stochastics itself? Figure 13.11 shows Stochastics as
displays strengths may warn that the market will have sufficient
the spread of two RSI indicators, what would we gain by applying
firms signal 1 in Stochastics. However, the fact that pattern a
If using a histogram is useful when plotting the averages from
At points a and A, divergence is present. Pattern A clearly con-
I gain anything by changing how I am presenting the Stochastics
an extremely important observation to make that the RSI Trend
We can fix this visual problem, as shown in Figure 13.12. Did
occurs. The XYZ Trend is now diverging as well with prices. It is is failing just as it tests the zero line. We must later test other market moves that occur in a histogram where the pivot becomes the zero line.
FXUSJYMUL-Daily
At signal 3 the reverse is developing from what we saw at signal 1. This mirror image between buy and sell signals is important. At signal 3 the RSI Trend is rolling up, but the XYZ Trend is suggesting that the market may have sufficient momentum to still attempt another bottom. At signal 4 both histograms are rolling upward. Points p, q, and r and P, Q, and R have been discussed in a prior chart. Signal 7 is much improved by the RSI Trend. However, notice that the moving averages at signal 2 in this indicator had a negative spread where the shorter period average has crossed below the longer period average. In signal 7 the averages are positive. This is an observation that would require back-testing to define statistically whether it is important or not. (It is.) In the XYZ Trend, the oscillator is not diverging prior to the intervention. However, the indicator is at a historic high peak within this chart. Extremes would have to be back-tested in this indicator to see if
Figure 13.11
4 2 2 • IMPROVING INDICATOR T I M I N G AND FILTERING PREMATURE SIGNALS
FXUSJYMUl
The Principles of Depth Perspective Applied to Two-Dimensional Charting • 4 2 3
takes practice and patience. T h e r e are so many skills that people b r i n g to technical analysis that it is no wonder we learn so many different and interesting ways to t h i n k t h r o u g h market problems. I thank you for the o p p o r t u n i t y to update this book and share a few new lessons learned along the way. I leave w i t h you a Z e n Story to offer a final thought:
A rich man, fond of felines, asked a famous Zen ink painter to draw him a cat. The master agreed and asked the man to come back in three months. When the man returned, he was put off, again and 100.00 RSI Trend
W
• 'A
'M-
'J- ••'J
again, until a year had passed. Finally, at the man's request, the master drew out a brush, and, with grace and ease, in a single fluid motion, drew a picture of a cat—the most marvelous image the
:4.00 -6.G0 W ' ' D 9 8 '
'A
man had ever seen. He was astonished; then he grew angry. 'That drawing took you only thirty seconds. W h y did you make me wait a year?' he demanded. Without a word, the master opened up a
Figure 13.12
cabinet, and out fell thousands of drawings—of cats.
study? I w i l l let you decide that one on your o w n . However, it is clear one can have independent control of some of the attributes we observed i n the Stochastics study. So what is the mystery indicator called the ' X Y Z Trend'? It is a simple spread o f two averages again on price that define the histogram. T h e n a 9- and 45-period m o v i n g average of this histogram was added. O n c e you t h i n k you have an idea w i t h merit, give it some t h o u g h t on what w i l l be the best way to display your new indicator. Sometimes an unconventional way w i l l add to its value. Just a little different food for thought. Perhaps you had not k n o w n that perspective enters into our business. I t does because the m i n d works i n three dimensions at all times. O u r craft
424
• IMPROVING INDICATOR TIMING AND FILTERING PREMATURE SIGNALS
CSX
ft.rp
The Principles of Depth Perspective AppUed to Two-Dimensional Charting » 425
^f-P CSX
Aerodynamic Investments Inc., © 1996-2011, Connie Brown, www.aeroinvest.com
ferp
<'X.P
llll
Aerodynamic Investments Inc., © 1996-2011, Connie Brown, www.aeroinvest.com
426
• IMPROVING INDICATOR TIMING AND F I L T E R I N G PREMATURE SIGNALS
The Principles of Depth Perspective Applied to Two-Dimensional Charting • 4 2 7
428 • IMPROVING INDICATOR TIMING AND F I L T E R I N G PREMATURE SIGNALS
CREDITS A l l charts with the few exceptions noted below by TradeStation (a subsidiary of TradeStation Group, Inc.), 8050 SW lOth Street, Plantation, Florida 33324, USA. Web site: www.TradeStation.com Figures 2.12, 9.4, 9.5, and 9.8: Market Trader Gold, A i r Software, Inc., 115 Caya Avenue, West Hartford, C T 0 6 I I 0 , USA. Web site: www.alphee.com Figures 9.6, 9.9, 9.10, 9 . I I , and 10.39: Market Analyst—Astro Version, Market Analyst International Pty L t d . , Brisbane, Australia. Web site: www.market-analyst.com Figure 10.38 (The Moody's Aaa Bond data): Prepared by Grant Noble, Topline Investment Graphics [(303) 440-0157] Figures 9.1, 9.2, and 9.3: Microsoft Excel, Microsoft Corporation
Aerodynamic Investments Inc., © 1996-2011, Connie Brown, www.aeroinvest.com
. 429 •