Classical Bar Charting & Technical Analysis

Kenneth H. Shaleen

Section One - The ‘Bottom of the Chart’

Chapter 1 Rationale of technical analysis

Chapter 2 Volume Analysis

Significance
Healthy Price Trends
Blowoff Volume
Determination of Volume Parameters

Chapter 3 Open Interest in a Futures Contract

Significance
Healthy Price Trends
Idiosyncrasies

Chapter 4 General Rule for a Healthy Price Trend (on a daily futures chart)
Why Total Volume & Open interest is Used
Worksheet

Section Two - The Basics

Chapter 5 Price Scales

Arithmetic verses Logarithmic
Futures Continuation Charts

Chapter 6 Trending with the Trend

Trendline Construction
Trendline Analysis

Chapter 7 Support and Resistance

Section Three - Reversal Patterns

Chapter 8 Price Pattern Recognition - an Overview

Chapter 9 Head & Shoulders Top and Bottom

Chapter 10 Broadening Formations

Chapter 11 Double Top and Double Bottom

i
 Section Four - Continuation Patterns

Chapter 12 Symmetrical Triangles

As a Continuation Pattern
As a Reversal Pattern

Chapter 13 Right Angle Triangle

Ascending Right Triangle
Descending Right Triangle

Chapter 14 Wedge Formations

Rising Wedge
Falling Wedge

Chapter 15 Flags & Pennants - ‘half-way’ formations

Section Five - Gap Theory

Chapter 16 Gap Theory

Four Basic Types of Gaps
Ex-Dividend Gap
Island Formation

Section Six - Minor Trend Change Indicators

Chapter 17 Minor Trend Change Indicators

Key Reversal
Outside/Inside Range
Mid-Range Close

Section Seven - Other Forms of Technical Analysis

Chapter 18

Chapter 19 Mathematical Models - Trend Following

Chapter 20 Mathematical Models - Oscillators

Chapter 21

Chapter 22 Spreading and Spread Charts

© Kenneth H. Shaleen
ii CHARTWATCH 9/08
International Futures Research

Kenneth H. Shaleen is President of CHARTWATCH, an international research

firm to the futures industry. CHARTWATCH distributes weekly technical
analysis at www.chartwatch.com and produces a daily telephone market
update.

Mr. Shaleen has instructed technical analysis courses for the Chicago Mercantile
Exchange since 1976. These courses are also conducted in London,
Singapore, Malaysia, Hong Kong and other locations throughout the world.

Mr. Shaleen is the author of Volume & Open Interest (Irwin, 1991, 1997) and
Technical Analysis & Options Strategies (Irwin, 1992), three Chicago Mercantile
Exchange course book as well as numerous articles.

Mr. Shaleen Joined the Northern Trust Company in Chicago in 1968 as a

Management Science Analyst to develop computer models for the Bond and
Trust Departments. A move to futures research was made in 1972. After
supplying analysis for twelve years as Director of Research for two Chicago
Board of Trade clearing member firms, he formed CHARTWATCH in 1984.

Mr. Shaleen is a charter member of the Commodity Options Market (1982) at the
Chicago Board of Trade. He holds a B.S. in Civil Engineering from the
University of Colorado (1967) and an M.B.A. in Finance from Northwestern
University (1968).

Please direct all inquires concerning videos and

any of the services offered by CHARTWATCH to:

CHARTWATCH
Fulton House 1604
345 North Canal Street
Chicago, Il 60606 USA
Telephone: 312 454-1130
 www.chartwatch.com

iii

Foreword

Earning an MBA in finance at Northwestern University in June of 1968, I was

never exposed to one of the most active markets in the world - the Chicago
futures exchanges - basically just down the street from the downtown
Northwestern campus.

After reading Commodity Speculation with Profits in Mind by L. Dee Belveal

(Commodities Press, 1967), I started my first chart - Sept ‘68 Oat futures. This
course manual is an ongoing compilation and refinement of all the charts and
technical analysis I have encountered since this time.

The majority of charts in this manual are futures charts, although the first
Technical Analysis Course I conducted for the Chicago Mercantile Exchange in
the Summer of 1976 used the Edwards and Magee “bible” of price pattern
recognition: Technical Analysis of Stock Trends. Now the technical analysis
course has come full circle with Stock Index futures charts and individual equity
charts increasingly in evidence. This is the course manual.

The “live” charts assigned in the early CME Technical Analysis courses were
Live Cattle and Soybeans. These evolved to Deutsche Marks and Treasury
Bonds - as the volume in financial futures became dominant. The simple
creation of a chart also evolved with the advent of electronic “after hours” trading
and the proliferation of technical analysis software.

Much of what is contained in this course manual is the old fashioned “art” of
drawing trendlines to construct classic price patterns. A combination of older
charts and more up-to-date examples are co-mingled to show that human nature,
creating the price patterns, does not change.

To the 15,000+ students who have helped me analyze these charts - my thanks.

Ken Shaleen
Chicago
Sept 22, 2008
Classical Bar Charting
&

Technical Analysis

Prepared and Presented

by

Ken Shaleen

President, CHARTWATCH Inc.

 Classical Bar Charting & Technical Analysis
Kenneth H. Shaleen

Section One - The ‘Bottom of the Chart’

Chapter 1 Rationale of technical analysis

Chapter 2 Volume Analysis

Significance
Healthy Price Trends
Blowoff Volume
Determination of Volume Parameters

Chapter 3 Open Interest in a Futures Contract

Significance
Healthy Price Trends
Idiosyncrasies

Chapter 4 General Rule for a Healthy Price Trend (on a daily futures chart)
W hy Total Volume & Open interest is Used
W orksheet

Section Two - The Basics

Chapter 5 Price Scales

Arithmetic verses Logarithmic
Futures Continuation Charts

Chapter 6 Trending with the Trend

Trendline Construction
Trendline Analysis

Chapter 7 Support and Resistance

Section Three - Reversal Patterns

Chapter 8 Price Pattern Recognition - an Overview

Chapter 9 Head & Shoulders Top and Bottom

Chapter 10 Broadening Formations

Chapter 11 Double Top and Double Bottom

i
 Section Four - Continuation Patterns

Chapter 12 Symmetrical Triangles

As a Continuation Pattern
As a Reversal Pattern

Chapter 13 Right Angle Triangle

Ascending Right Triangle
Descending Right Triangle

Chapter 14 W edge Formations

Rising W edge
Falling W edge

Chapter 15 Flags & Pennants - ‘half-way’ formations

Section Five - Gap Theory

Chapter 16 Gap Theory

Four Basic Types of Gaps
Ex-Dividend Gap
Island Formation

Section Six - Minor Trend Change Indicators

Chapter 17 Minor Trend Change Indicators

Key Reversal
Outside/Inside Range
Mid-Range Close

Section Seven - Other Forms of Technical Analysis

Chapter 18

Chapter 19 Mathematical Models - Trend Following

Chapter 20 Mathematical Models - Oscillators

Chapter 21

Chapter 22 Spreading and Spread Charts

© Kenneth H. Shaleen
ii CHARTWATCH 9/08
 International Futures Research

Kenneth H. Shaleen is President of CHARTW ATCH, an international research firm

to the futures industry. CHARTW ATCH distributes weekly technical analysis at
www.chartwatch.com and produces a daily telephone market update.

Mr. Shaleen has instructed technical analysis courses for the Chicago Mercantile
Exchange since 1976. These courses are also conducted in London, Singapore,
Malaysia, Hong Kong and other locations throughout the world.

Mr. Shaleen is the author of Volume & Open Interest (Irwin, 1991, 1997) and
Technical Analysis & Options Strategies (Irwin, 1992), three Chicago Mercantile
Exchange course book as well as numerous articles.

Mr. Shaleen Joined the Northern Trust Company in Chicago in 1968 as a

Management Science Analyst to develop computer models for the Bond and Trust
Departments. A move to futures research was made in 1972. After supplying
analysis for twelve years as Director of Research for two Chicago Board of Trade
clearing member firms, he formed CHARTW ATCH in 1984.

Mr. Shaleen is a charter member of the Commodity Options Market (1982) at the
Chicago Board of Trade. He holds a B.S. in Civil Engineering from the University of
Colorado (1967) and an M.B.A. in Finance from Northwestern University (1968).

Please direct all inquires concerning videos and

any of the services offered by CHARTW ATCH to:

CHARTW ATCH
Fulton House 1604
345 North Canal Street
Chicago, Il 60606 USA
Telephone: 312 454-1130

www.chartwatch.com

iii
 Foreword

Earning an MBA in finance at Northwestern University in June of 1968, I was never

exposed to one of the most active markets in the world - the Chicago futures
exchanges - basically just down the street from the downtown Northwestern campus.

After reading Commodity Speculation with Profits in Mind by L. Dee Belveal

(Commodities Press, 1967), I started my first chart - Sept ‘68 Oat futures. This
course manual is an ongoing compilation and refinement of all the charts and
technical analysis I have encountered since this time.

The majority of charts in this manual are futures charts, although the first Technical
Analysis Course I conducted for the Chicago Mercantile Exchange in the Summer
of 1976 used the Edwards and Magee “bible” of price pattern recognition: Technical
Analysis of Stock Trends. Now the technical analysis course has come full circle with
Stock Index futures charts and individual equity charts increasingly in evidence. This
is the course manual.

The “live” charts assigned in the early CME Technical Analysis courses were Live
Cattle and Soybeans. These evolved to Deutsche Marks and Treasury Bonds - as
the volume in financial futures became dominant. The simple creation of a chart
also evolved with the advent of electronic “after hours” trading and the proliferation
of technical analysis software.

Much of what is contained in this course manual is the old fashioned “art” of drawing
trendlines to construct classic price patterns. A combination of older charts and
more up-to-date examples are co-mingled to show that human nature, creating the
price patterns, does not change.

To the 15,000+ students who have helped me analyze these charts - my thanks.

Ken Shaleen
Chicago
Sept 22, 2008

iv
 Chapter One

Rationale

In the long run, the price of any freely traded item is determined by the interaction
of the fundamentals of supply and demand. But in the short run, the price of that
item could move in exactly opposite direction as dictated by the fundamentals.
W hy? W hat is contained in the determination of the price? Hopes, fears, and
moods. An these moods can be rational or irrational. The technical trader
follows the price only. Price changes produce the profit or loss; as a final result,
this is what is important.

It is difficult for a trader to isolate a technically derived market view from the
fundamentals. The question always arises: W hy is price moving the way it is and
why should it go to a particular price level that might be predicted by technical
analysis? A technician must try to divorce himself from these thoughts. This
does not mean that a technician does not know what is going on in the “real”
world. Often, a scheduled economic report will initiate the price move. Indeed, it
takes reversals of the minor price trend to create the handful of classic price
patterns that will be studied in this course manual. A fundamental event is
usually responsible for creating the minor price change. So a technical trader
should monitor the economic release calendar - and be prepared for the possible
price moves that will result. Often these minor price moves create the necessary
conditions for a classical bar charting price pattern.

The bulk of this course manual will be devoted to the “low tech” art of price
pattern recognition. There is nothing magic about these price patterns. They are
the interaction of supply and demand variables as perceived by the marketplace.
And human nature - reacting to economic events and shock variables - tends to
repeat itself.

Equity Markets verses Futures Markets

The techniques investigated are applicable to equity (stock) charts and futures
charts. Each of these markets has its own unique characteristics. These
characteristics are not difficult to understand - and do not pose any major
problems for the astute technician.

1-1
The Difference between Equity and Futures Charts

The main difference between equity charts and futures charts is that equity charts
contain only price and volume statistics whereas a futures chart contains the
added variable of open interest. There is a major conceptual difference in a
market that must have a short for every long at the end of a trading session i.e., a
futures market, verses the equity market where there are (predominately) longs
only, at the end of the trading session.(For purists, the concept of short interest in
the equity market will be detailed in Chapter Two, Open Interest).

The question often arises: W hy should a derivative market, the futures, be

analyzed when the underlying “cash” or “spot” market is the dominant market? A
futures market that has achieved a “critical mass” becomes a microcosm -
diminutive but analogous to the whole (cash market). The “basis”, the difference
between the cash market price and the futures price, is highly arbitraged and
produces a relatively stable (although dynamic) relationship between the cash
and futures market.

1-2
 Chapter Two

Volume

Volume provides a “filter” for classical high, low, close, bar chartists. Volume is
so important that when this internal statistic does not support the technical
conclusion derived from the analysis of price movements, the contemplated trade
is suspect and would not be initiated. In this regard, volume is often the
validating statistic that causes a technical trader to “pull the trigger” on a new
trade.

Definition: Volume is the number of futures contracts (shares for an equity

market participants) traded each trading session.

In a futures market, a contract is consummated only when both sides of the trade
agree to the price and quantity. At the end of a trading session, the total number
of contracts bought equals the total number of contracts sold. Therefore, the
following equality always prevails:

Buy Volume = Sell Volume = Total Volume

Published volume figures represent one side of a futures trade only. The phrase
“more buyers than sellers” is never true with respect to volume statistics. A more
representative phrase to explain the rise in prices during a particular trading
session might be “more potential buyers than sellers”.

The dissemination of volume (and open interest) statistics varies widely by

futures exchange. In an open outcry trading environment the final volume totals
are not available until after the close of trading. Electronic screen based trading
(and most equity exchanges) do show “on-line” volume. This is obviously a boon
to technical traders.

The CME Group Clearing House releases the volume (and open interest)
statistics prior to the start of open outcry trading the next morning in the U.S.
See the Appendix of this manual for examples of how to obtain the statistics
directly from the exchange website.

Note that once each week, a technician will be able to analyze the statistics
before trading begins the next day. This occurs with the Saturday morning
availability of the data for the Friday trading sessions.

2-1
Significance of Volume: Volume is a measure of urgency. It is the result of the
need for traders and investors to “do something”. And nothing creates more
urgency in a market than a losing position. Since any useful chart analysis
determines what the losers are doing, the technician will want to monitor this
sense of urgency, thereby assessing the health and strength of the prevailing
price trend.

The specific volume number is not important. It is necessary to classify the

trading session’s volume into one of three categories: low, average or high.
These three categories are not static; Their boundaries will fluctuate with
substantial changes in open interest and/or price volatility.

Ideal Healthy Price Uptrend: The ideal situation for a healthy bull market is
volume moving up as the bull market expands. A strong price uptrend is
characterized by greater volume on up days when prices close higher, than on
down days when prices settle lower. This relationship is shown schematically in
Figure 2-1.

Figure 2-1
Ideal Bull Market
Price versus Volume Interaction

2-2
Volume measures how anxious trader are to establish or close out their positions.
The ideal situation for a healthy uptrend in prices is for volume to increase on
rallies in price and decrease on selloffs. This configuration created the old adage
- “Don’t sell a quiet market after a fall” - because a low volume selloff is actually a
bullish technical situation.

Monitoring volume to identify price moves as counter-trend is important. Low

volume on price down days is telling the astute trader that there is no urgency on
the part of the longs or shorts to close out their positions - and thus the prevailing
major price uptrend should continue.

Ideal Healthy Price Downtrend: The ideal situation for a healthy bear market is
for volume to increase a prices move lower. A strong price downtrend is
characterized by expanding volume on days when prices close lower and
increasing volume on price up days. This concept is shown schematically in
Figure 2-2.

Figure 2-2
Ideal Bear Market
Price versus Volume Interaction

2-3
Even in a long-term bear market, prices will not continually decline. Adverse
moves against the direction of the major trend will result in price rallies - some
lasting several days, or erratically, over several weeks. If no urgency develops
for the shorts to cover their positions, volume should decline. This is the
proverbial low volume rally. A low volume rally is bearish. Ergo, “Don’t buy a
quiet market after a rise”.

Blowoff Volume - A Warning Signal: There is one amplification of the ideal

price versus volume configuration that is of paramount importance and should not
be overlooked. Losing positions, especially speculative ones (as opposed to
legitimate hedges), often create conditions that lead to excessive volume. The
urgent need to close out losing positions produces “blowoff “ volume.

Blowoff volume is volume of an extremely high magnitude which is a warning

signal that the price trend is in the process of exhausting itself.

Prices often move violently in the opposite direction after such blowoff volume.
Figure 2-3 illustrates the theoretical relationships between blowoff volume and
price activity. Figure 2-7 is an actual example of this sign of exhaustion in the
Corn futures market.

Figure 2-3
Theoretical Examples of Blowoff Volume

One caveat is in order. Extreme high volume is the warning signal which
indicates at least a temporary trend change. This signal does not have to
coincide with the exact extreme price day. Often blowoff volume will occur one
trading session before the ultimate high or low price posting.

2-4
 Determination of Volume Parameters

Figure 2-4

A bar chartist will scan back an ‘appropriate distance, whether two weeks or two
months, depending on the specific conditions prevailing on each chart.
Hypothetical horizontal lines are drawn, representing the threshold levels of “low
and “high” volume.

One third of the volume readings should fall into each of the three categories

2-5
 Volume Analysis

Figure 2-5

2-6
 Volume Analysis

Figure 2-6

Note the increase in volume on price down days

and reduction in volume on price rallies

2-7
 Example of Blowoff Volume

Figure 2-7

2-8
 Chapter Three

Open Interest
Futures trading is a zero sum game: for every dollar in - there is a dollar out.
Admittedly, the exchange clearing house and member firms scoop a little off the
top; but for every open position in a futures market there has to be an opposite
position.

Definition: Open Interest (O.I.) Is the summation of all unclosed purchases or

sales at the end of a trading session. Similar to volume, the published figure
represents one side of the transaction only. Confusion concerning the definition
of open interest can be avoided by remembering this simple equality:

Long O.I. = Short O.I. = Total O.I.

Futures technicians monitor the total open interest figure. This number represents
the summation of open positions in all the contract months traded for the
particular commodity. The reason for using total open interest is explained in
graphic detail in Chapter Four.

How Open Interest Changes: Open interest changes from one trading session
to the next, fall into one of three categories: 1) Increase, 2) Decrease, or 3) No
change. Each of the three situations will be examined in a hypothetical example.
For this illustration, it does not matter whether prices moved up or down. W hat is
necessary is that a “significant” price change occurred. W hile no specific
definition of what constitutes significant will be given, it is safe to assume that the
definition begins at more than five minimum price tics. The technical
ramifications of these changes will be apparent later in this chapter when
detailing the ideal healthy price uptrend or downtrend.

Example:

Prior Day’s Total Open Interest = 180,000

Answer the question: W ho is getting in or out of the market?

Case One: Open Interest Increases

Total O.I. now at 183,000 3,000 new long contracts opened

a change of +3000 3,000 new short contracts opened

3-1
 Case Two: Open Interest Decreases

Total O.I. now at 178,000 2,000 long contracts sold out

a change of -2,000 2,000 short contracts bought back
(covered their existing shorts)

Case Three: Open Interest Unchanged

Total O.I. now at 180,000 In this situation, the trader would not
no change know exactly what changing of
positions occurred.

Significance of Open Interest: There are three reasons why technically based
futures traders monitor open interest.

Open interest:

1) Indicates the existence of a difference of opinion.

2) Provides “fuel” to sustain a price move.

3) Determines if the losers are being replaced.

There is nothing that creates a market more than a difference of opinion. This is
reflected in a willingness to take an open position. Open interest is a reflection of
this important concept. If a market was at equilibrium and the entire trading world
knew this - the open interest in a futures listed on that commodity would be zero.
There would be no need for hedgers to lay off unwanted risk, or speculators
placing positions trying to profit from a mis-priced market.

The analogy of fuel to the market is like that of fuel to a fire. If the fuel is removed
from a fire, the fire will go out. If fuel is removed from a price trend, the trend will
change. Fuel in a futures market is provided by the losing positions. W hen open
interest declines, fuel is being removed and the prevailing price trend is running
on borrowed time. For a healthy, strong price trend (either up or down) to
continue, open interest should increase, or at least not decline. This is so
important a concept that remembering the word fuel as a surrogate for open
interest will place a trader ahead of 80 percent of all futures traders worldwide!

3-2
Technicians do not care if a losing position is being financed by meeting margin
calls and throwing more money at the market, or if a loser steps aside and new
blood comes in to take the loser’s place. W hat matters is that the funds are being
posted at the clearinghouse. The losers are necessary to pay off the traders with
the correct market judgment. W hen the losers decide that they “don’t want to
play the silly game anymore” and leave the market, open interest will decline.
Obviously the losers pay the price for their misjudgment, but what is of
importance to the technician is that declining open interest means the prevailing
price trend has become very unhealthy.

Ideal Healthy Price Uptrend: In an uptrending market when open interest is

going up, both sides are increasing their positions. The additional short sellers
may be existing shorts adding to their losing positions, or new short sellers
entering the market (thinking prices are too high). Longs may be adding to their
profitable positions, or new longs may be joining the bull bandwagon. W hat is of
the most importance to the longs is the fact that the losers are being replaced and
more fuel to sustain the upmove is entering the market. This is a healthy price
uptrend and prices should continue to work higher as long as open interest does
not begin to decline. This is shown in schematic fashion in Figure 3-1.

If open interest is declining, the underlying support is suspect. Liquidation of both

long and short positions is occurring. Less conviction concerning probable price
movement together with less fuel to sustain the price trend produces a definite
warning signal of an impending trend change.

Precautionary measures should be taken such as moving protective stop orders

(hopefully to realize profits) closer and placing orders to liquidate existing
positions. Potential hedgers should begin to implement any previously planned
strategies to protect inventory, lock in profit margins, etc.

Figure 3-1
Ideal Bull Market
Price versus Open Interest
Interaction

3-3
Ideal Healthy Price Downtrend: The ideal situation for a healthy price
downtrend (price expected to continue moving lower) is open interest increasing
when prices close lower and open interest decreasing on trading sessions when
prices close higher. This is shown in Figure 3-2.

There is often difficulty in finding the ideal technical situation of open interest
increasing during a downtrending market, especially in the metals, grains and
livestock futures. This is because “commodity” speculators are by nature bullish.
They would much rather buy something than be short. Thus a short position in
futures can be taken if other indicators, such as price patterns, suggest lower
prices - even though the ideal situation of increasing open interest is not present.
This concept will be illustrated in many of the charts in this course manual.

Figure 3-2
Ideal Healthy Price Downtrend
Price versus Open Interest
Interaction

Short Interest (Equities)

The concept of short interest in the U.S. or any equity (stock) market) is
completely different from open interest in futures. In any equity market that
allows short sales using borrowed stock, a short interest situation might exist.
Short interest is the number of shares that have not yet been purchased to cover
short sales. The borrowed stock must eventually be returned to the lender. Two
schools of thought prevail as to what short interest implies:

1) Traders expecting a price decline are initiating short sales;

They could be correct,
or
2) By definition, the shares sold short must eventually be bought back:
A bullish situation.

Increases in takeover activity and arbitrage cloud the issue of how to interpret
short interest statistics from an equity market.

3-4
 IDIOSYNCRASIES

Questions quickly arise when a new trader is exposed to the concept of how open
interest should ideally interact with price changes in a futures market.

1) W hat happens to open interest as a market reverses price direction?

2) How can open interest continually increase in the direction of the new price
trend? This would mean that open interest would have to continually rise!

Markets in Transition: The answer is that normally open interest does decline as
a new price trend initially gets underway. This is because the participants with
the correct (profitable) positions are liquidating (realizing that the trend is
changing) and the losers are closing out their positions (confused and not
realizing that the trend really is finally starting to go their way).

Open interest often declines to a “steady-state” condition - a level of open

contracts that often approximates the level that prevailed before the start of the
previous price trend. As the new price trend becomes more apparent, open
interest should then begin to increase - fueling the price move to new highs or
lows as the smart money rides the trend and the skeptics fight it.

Open Interest as a Coincident Indicator: On many occasions, open interest acts

as a coincident indicator with price. This occurs most often in the traditional
physical commodity futures during a bull market. Total open interest rounds
upward as a price bottom is being made and tops out as price peaks.

In general, any time open interest begins to act erratic after experiencing a steady
increase, price is, at best, moving net sideways, and more than likely trying to
change direction.

A special case in which open interest often acts as a coincident indicator in both
bull and bear directions is the CME Group currency futures. Total open interest
has acted in a coincident fashion with price in so many instances that it cannot be
attributed solely to chance. The changes in open interest in the foreign exchange
(forex) futures accurately reflect how the dominant price making force is changing
its market view.

3-5
 Ideal Healthy Price Uptrend - and Early Warning Signal

Figure 3-3

3-6
 Ideal Healthy Price Downtrend

Figure 3-4

3-7
 Historical Charts

Figure 3-5

3-8
 Historical Charts

Figure 3-6
Ideal Healthy Price Downtrend

Lean Hogs - Feb 1999

Note that after the January 5 price decline, a vicious short covering rally ensued.
Technical traders cannot simply use the fact that an price rally is short covering -
to initiate short sales. A price reason must be present.

3-9
 Historical Charts
Figure 3-7
Open Interest as a Warning Signal

3-10
 Historical Charts

Figure 3-8

13-11
 Historical Charts

Figure 3-9
Open Interest in the Transition from a Bull to a Bear Market

3-12
 Historical Charts

Figure 3-10
Open Interest Increasing in a Bear Market

Note the transition from a bear to bull market was initially accompanied by a
ecrease in open interest. This is usual.

3-13
 Historical Charts

Figure 3-11 A Figure 3-11B

Open Interest as a Coincident Indicator History does Repeat Itself

Deutsche Mark Sept 1998

Note how the liquidation in total open interest corresponded to the price declines.
W hen the bulls finally had the courage of their convictions and were willing to hold
long positions overnight, price was able to move up.

Also observe the interaction of price and open interest at the price highs. The
open interest line resembles the Head & Shoulders Top that formed on the price
chart!

3-14
 Chapter Four

General Rule for a Healthy Price Trend

Volume and Open Interest

Should Increase as Prices Move in the Direction
of the Major Price Trend

According to this rule, the most bullish condition on a futures chart is price moving
up on increasing volume and increasing open interest; The longs are in control
and the price uptrend is expected to continue. On a daily basis, this rule implies
that on price up days (when quotes close higher than the previous trading
session), volume should expand and open interest should increase.

In a strong bear market when quotes close lower, volume will expand and open
interest will increase. This ideal healthy price downtrend does not often occur in
those markets that the public prefers to trade from the long side. These include
traditional agricultural commodity futures and metals. Conversely, markets such
as interest rate futures (U.S. Treasury Bonds in particular) often do exhibit the
ideal bear market characteristics of volume and open interest up on price down
days. The most bullish and bearish technical situations are shown schematically
in figures 4-1 and 4-2.

Figure 4-1 Figure 4-2

Ideal Healthy Bull Market Ideal Healthy Bear Market

4-1
Why Total Open Interest is Used

4-2
Volume and Open Interest Analysis - Worksheet

4-3
 Ideal Healthy Price Uptrend

Volume Should: Increase on price up moves

Decrease on price down moves

Open Interest Should: Increase on price up moves

Decrease on price down moves

4-4
 Ideal Healthy Price Downtrend

Volume Should: Increase on price down moves

Decrease on price up moves

Open Interest Should: Increase on price down moves

Decrease on price up moves

4-5
 Ideal Healthy Price Uptrend

Volume Should: Increase on price up moves

Decrease on price down moves

Open Interest Should: Increase on price up moves

Decrease on price down moves

4-6
4-7
4-8
Weak Price Uptrend - Downside Reaction Likely

4-9
4-10
 Chapter Five

Price Scales

One of the first decisions a prospective bar chartist must address is what type of
price scale to use. The two choices are arithmetic or logarithmic. First, a
discussion of each type of scale - and then a recommendation.

Arithmetic scales are constructed with each small square on the chart actually
being a square. This referred to as rectangular coordinate chart paper.

Logarithmic scales increase in a doubling of price in equal price distances.

Since time is recorded in linear fashion, the result is semi-log chart paper.

Classical bar chartists quite often use the vertical height of the price pattern to
determine upside or downside measuring objectives. Obviously, the choice of
price scale will generate much different price objectives.

An upside target on a semi-log scale chart will be much higher than the same
price pattern measured on an arithmetic scaled chart.

Downside measuring objectives will not be as low on a semi-log scale chart as

compared to the target from the same bearish price pattern as reflected on an
arithmetic scaled chart.

This quandary obviously needs answering.

What gives futures trading its allure?

Answer: The high leverage or ‘gearing’.
It does not take a doubling or tripling of price to produce sizable percentage gains
or losses. In forty years of examining futures charts on a daily basis, Ken Shaleen
has found that futures charts plotted with an arithmetic scale do meet the
standard vertical height measuring objective dictated by the pattern. Note that
this objective could be calculated mathematically - and arrive at the same price
target as the ‘graphic’ method of taking the height and physically moving it over to
the breakout level.

The ‘bottom line’ is that the use of an arithmetic price scale on a daily futures
chart is perfectly acceptable.

5-1
In general, individual equities (stocks) move slower than anything listed as a
futures contract. And the equity market is dominated by ‘buy-and-hold’ type
investors who do not use margin accounts to leverage their holdings. This type of
market participant also tends to be on the long side of the market. They are
looking for individual stocks that are increasing in price at a good rate.

An upsloping trendline on a semi-log scale chart is increasing at a constant

percentage increase. This is the stock that an equity investor wants to find. They
will have an easier time finding such a buy-and-hold candidate on a semi-log
scale chart. In addition, the universe of equities to choose from is huge
compared to the 60-80 actively traded futures contracts.

The ‘bottom line’ is that individual equity traders should find semi-log scale charts
very useful.

Two exceptions

There are several situations in which a futures chartist might find changing the
price scale from arithmetic to logarithmic, helpful.

1. After a large price move has occurred on a daily chart. This would happen if a
valid Flag pattern forms. This is a very dynamic price pattern. The measuring
objective in a Bull Flag pattern can be far surpassed using an arithmetic scale.
Determining the graphic objective using a log scale will result in a much more
aggressive (higher) target.

2. Looking a long term (weekly or monthly) continuation chart of the nearest-to-

expire future over a long time period (years) might be better viewed with a log
scale for price. This is especially true for the stock index futures - where a long
term bull market existed. And, the price blowoff to the upside in the physical
commodity futures in 2007-08 can be kept in context by using a log scale.

5-2
 Example of Semi-Log Scale

Figure 5-1 London Gold - monthly

5-3
 Comparison of Arithmetic and Logarithmic Price Scales

Gold - Monthly Chart

nearest-to-expire future

Figure 5-2A Figure 5-2B

Price drop off the March 2008 high (1033.90 on the near-by future) ‘seems’ much
larger on the arithmetic scale chart.

5-4
 Chapter Six

Trading with the Trend

A ‘trader’ is different from an ‘investor’.

The trader is operating in a very short time frame - down to seconds and minutes.
For the day trader, scalper, market maker, it is the speed of execution and small
commissions that produces their profits.

The investor desires to hold the position and therefore has a much longer time
frame. A position trader is an investor who wants to ride a price trend. They have
the best chance of making a profit by identifying the direction of the major price
trend, ascertaining its health (strength), and trading with it.

Even a casual observation of historic charts indicates the presence of many

sustained price trends - up or down. Sometimes these trends last several years.
Yes, there might be some sizable moves against the major trend, but the long
term trend continues.

Trendline Construction

Defining a price trend is easy; trading with it is not so easy. Higher and higher
price highs and higher and higher price lows is the basic definition of an uptrend.
Figure 6-1 is a plot of this definition. Each upmove extends to new high price
territory while the selloffs do not decline as far as the price drop on previous
selloffs.

Figure 6-1 Price Uptrend

6-1
Downtrends are characterized by lower highs and lower lows. See figure 6-2.
In a downtrend it is possible to construct a straight line tangent to two price highs.
Typically these are the highs of the price bars on whatever time frame chart
(minutes, hours, days, etc.) Is being used. This line is referred to as the
downtrend line.

Figure 6-2 Price Downtrend

In a bull market, the uptrend line is the one drawn across (tangent to) the low of
the price bar at each of the relative price lows. This is shown twice in figure 6-3.

It only takes two points to determine a straight line. If a third point lies on any
trendline, a significant charting event occurs. The trendline takes on much
greater significance. The trend direction has much greater validity.

Figure 6-3 Uptrend Lines

6-2
Any trendline, when initially drawn, should not cut thru prices. This is the cardinal
rule of trendline construction. The line must be tangent to two relative highs or
lows. Subsequent price action (after the trendline has been drawn) may cut thru
prices - and obviously has to when the trend changes.

A three point trendline is infinitely greater in its technical significance

than a two point line.

A trader strives to find the technical situation where a market is residing above or
below a three point trendline. The trend will continue until a close beyond the
trendline occurs. Even then, the trend does not completely reverse; it changes
the trend to sideways. Violating a three point trendline is cause for exiting an
existing position. It is not enough evidence to completely reverse direction.
Addition technical corroboration is needed. Figure 6-4 shows a three point
downtrend line on the Dec 2008 Crude Oil future.

Figure 6-4
Three Point
Downtrend Line

Note that the 1 st

trendline was only
constructed using
two points. It did
not hold.

6-3
A chartist never erases a properly drawn trendline once it has been constructed.
This is especially true for shorter trendlines used to delineate the boundary lines
of an orthodox price pattern. Reference to the chart months or years later will
allow the technician to see what the prevailing price configuration implied - even if
the ensuing price move proved the analysis incorrect.

Knowledge of the technical personalities of each market will be gained by review

of the success or failure of each identifiable price pattern.

Another aspect of trendline construction is the use of a line constructed parallel to

the trendline. It is known as a parallel objective line or a return line.
Some chartists construct this line as a reasonable price target for the extent of
the next move. In bull markets, this upper line is often exceeded as the price up
move goes into a greater vertical angle of ascent. Still, it is a ‘reasonable’ line to
place on the chart.

Trend Channels

Although not as common as most traders might think, markets sometimes do

trend within a well defined channel - where the trendline and parallel objective
line contain the price swings.

The weekly continuation chart of the nearest-to-expire British Pound future in

figure 6-5 is a good example. Violation of the three point upsloping trendline
with a weekly close below the line would signal that the major price uptrend has
stopped - and turned the market to neutral.

Figure 6-5 Trend Channel on a Weekly Continuation Chart

6-4
 Chapter Seven

Analysis of Support and Resistance

Locating support and resistance is probably the most important part of any
classical bar charting analysis. Properly locating the price levels where support or
resistance would be expected, allows the trader to:

1. Set entry points for new positions.

2. Place protective stop orders.

3. Determine if a price trend is present.

4. Locate where the prevailing trend, if any, would change.

The concept of support and resistance is one of the most talked about, but least
understood aspects of technical analysis. In the process of defining support and
resistance, the following observation is useful:

Ask 100% of all the would-be ‘technicians’ in the trading world where overhead
resistance is located on the following line graph (Figure 7-1) . . . .

Figure 7-1

. . . . . and the overwhelming majority would state that the highest point on the
graph is resistance.

This seems logical because the previous price rally did stop at that level once
before. But, if a simple price high was overhead resistance, the price upmove
would turn down when approaching that level again. An uptrending market would
never materialize. And Double Top formations would be much more prevalent.
Markets tend to trend most of time, rather than reverse. The ‘double’ formation is
not a common reversal pattern.
7-1
Giving a price high on a chart a name, a technician should simply refer to it as a
“Benchmark High”. This is illustrated in Figure 7-2.

Figure 7-2

Similarly, a price low on a chart (Figure 7-3) should be referred to as a

“Benchmark Low” and should not automatically be labeled as support. e.g.,

Figure 7-3

Market makers in equities, foreign exchange dealers, and scalpers on an open

outcry futures floor do notice that price often temporarily stops in the area of a
previous price high or low. The operative word is temporary. Position traders do
not operate as close to the market as these highly active traders. It is usually
impossible for all but the most active day traders to make money by ‘fading’ a
market as it reaches a benchmark high or low.

The daily and 10-minute charts (Figures 7-8 & 9) of the Dow Jones Industrial
Average illustrates how a benchmark low can act as temporary support. The
June 16 low of 8570 on the daily high low close bar chart was well known and
easily seen by all traders. Look at what happened during the trading session in
which the price decline again approached 8570 seven weeks later.

7-2
The supposed “support” was only slightly violated with the 8562 low on a 10-
minute price bar (Figure 7-9). The next time down on the 10-minute chart the
cash Dow stopped exactly at 8570. The selloff had reached “support”! (not
really). The next time down the Dow moved thru 8570 like a knife thru warm
butter.

If an aggressive short term trader wants to label a benchmark low price as

possible temporary support - fine. This trader must realize that if a healthy price
downtrend trend is in progress, this low will be taken out - usually quite
dramatically.

W hat then, is a more classical definition of support and resistance?

First, support, if any, resides below the current price level. Think of support in
two words - underlying support. Resistance, if any, is above the current price.
Resistance is always - overhead resistance.

The chartist’s “bible” Technical Analysis of Stock Trends, by Edwards and Magee
is still the best source of a definition. Specifically, . . .

A former price high (or top) should be underlying support in the ensuing price
uptrend. To find support, a chartist looks down and to the left on a chart and
locates the closest former price high. See Figure 7-4.

Figure 7-4 Support is: a Former Top

7-3
A former price low (or bottom) on a chart should be overhead resistance in the
ensuing price downtrend. To find resistance, a chartist looks up and to the left on
a chart and locates the closest former price low. See Figure 7-5.

Figure 7-5 Resistance is: a Former Bottom

Traders are obviously interested in what will happen as quotes ‘test’ a benchmark
high or low. It is important not to harbor a preconceived notion that a price rally
has to stop at a price level that it (temporarily) stopped at before - or that a price
sell-off has to bounce off a price level that stopped a sell-off previously.

Relating volume to this concept:

The lower the volume on the price sell-off to test support, the more likely
the support will hold.

The lower the volume on the rally to test resistance, the more likely the
resistance will hold.

The strength (ability to halt the price move) of the support or resistance, in theory,
should be directly proportional to the amount of dealing previously done at that
particular price level. Quotes may have to eat into support or resistance until they
get to a level where a sufficient amount of prior trading occurred - to finally
produce the friction necessary to stop the price move.

7-4
Putting support and resistance into practice:

Traders wishing to initiate a new long position would place the order to buy at a
price slightly above the support price. A good definition of “slightly’ is two
minimum price fluctuations.

Traders wishing to initiate a new short position would place the sell order at a
price just below (2 tics) the resistance price.

And, the price should not end in zero. Make any buy order end with a price of 1 or
6. Make any sell order end with a price of 4 or 9. The public gravitates to
common, all too common, prices ending with zeros.

Protective stop orders should be entered far enough into support or resistance so
the price move would have to inflict an inordinate amount of damage to the
support or resistance before the trader would get knocked out of the market. It is
far easier to make this statement than to quantify it. Every market has it’s daily
‘noise’. The stop must be placed beyond this noise.

Some markets, the stock index futures in particular, are notorious for eating farther
into support or resistance than any other market. The proliferation of
undercapitalized day-traders makes for this phenomena.

Another important note: Simply violating support or resistance (with a close into it)
does not automatically change the direction of the prevailing price trend 180
degrees. It changes the trend to neutral. Additional technical evidence must be
present to warrant entering a new position.

Locating support and resistance on a chart is an age-old technique. It does not

require a sophisticated computer package. It is an excellent place to begin the
analysis of any new chart.

7-5
 Example of Underlying Support

Airtouch Communications
Common Stock
Daily Chart

Figure 7-6

7-6
 Example of Overhead Resistance

Figure 7-7

7-7
 Dow Jones industrial Average (cash) - Daily
Figure 7-8

7-8
 Dow Jones industrial Average (cash) - 10 Minute
Figure 7-9

7-9
Blank

7-10
 Chapter Eight

Price Pattern Recognition

Overview

The high, low and last price for a specified time period are the basic inputs for a
classical bar chartist. In specialized forms of charting, such as candle charts, the
‘opening’ price is also recorded on the graph. In the move toward 24-hour
electronic dealing, there is some question about the benefit of the next day’s
open when it happens a nanno-second later than the previous day’s close.

Volume, if available, is an important aspect of any chart - especially a daily high,

low, close bar chart. Open interest on a daily futures chart is also a helpful aid in
assessing the validity of a price pattern.

This course manual primarily contains examples of futures charts. Reader

desiring more illustrations of individual equity charts are referred to the classic:
Technical Analysis of Stock Trends, by Edwards & Magee.

Price Pattern Recognition

Price patterns that tend to appear regularly on a classical bar chart are one of the
most valuable tools for a technical trader. These patterns indicate the high
probability direction of the major price trend, often carry specific measuring
objectives - and, most importantly, alert a trader where things are going wrong.

The rationale behind the use of price charts is that all the supply and demand
information has to be synthesized into a single piece of information - price. If
fundamental forces are at work moving the market, the price will show it.

Also affecting the price movement is activity by speculators who utilized

information other than fundamental supply/demand data, such as psychological
judgements. These market moving decisions are also reflected in the price.
Although some of the decisions may be based on hopes, fears, greed, they
repeat themselves often enough to allow the technician to make a future forecast
based upon past price patterns.

Only a handful of orthodox price patterns exist. They divide themselves into two
distinct groups: Reversal patterns or continuation patterns. Chapters 9 thru 17
detail these formations.

8-1
Hierarchy of Bar Charting Analysis

1. Identify the direction (if any) of the major price trend

2. Check the internal health of the trend by using volume and open interest.

3. Locate the closest underlying support and overhead resistance price levels.

4. Look for the possibility that an orthodox price pattern may be present - or in
the process of forming.

5. If an active price pattern is present, determine the measuring objective and

the stop-out price level.

6. Create a trading plan based on the bar charting conditions that exist; often
this is best accomplished when the market is not open.

7. Do not try to force a conclusion via assuming that a reasonable risk / reward
trade must be present; sometimes no trade is the best trade.

8. A chartist need not get married to a single market; diversify the analysis.

8-2
 Chapter Nine

Head & Shoulders Reversal Pattern

The Head & Shoulders (H&S) price pattern is the most reliable reversal pattern.
It can occur as either a Head & Shoulders Top or a Head & Shoulders Bottom.
This classical bar charting price pattern physically resembles a head & shoulders
when found at a market top.

For clarity, when referring to a bottoming of a price trend, technicians should use
the phrase H&S Bottom rather than ‘inverse Head & Shoulders’. By using this
terminology, there will much less confusion whether the trader is talking about
price topping or price bottoming.

Before discussing the pattern in detail, there are several overall observations:

1. Before a chartist should be looking for any reversal pattern, there must be a
definite price trend in place - so there is something to reverse.

2. Five obvious reversals of the minor price trend must be present on the chart to
form the pattern. Figure 9-1 clearly shows these five reversals.

Figure 9-1

9-1
Development of a Head & Shoulders Top:

In a price uptrend, a corrective selloff (from reversal point one) and the
subsequent rally to a new price high (off the low of reversal point two) implies
that the uptrend is firmly entrenched. These first two reversals of the minor price
trend will create what will be the ‘left shoulder’ of the completed pattern.

The highest price in the pattern will be called the ‘head’. It is quite possible that
extremely high (blowoff) volume might be present on the price move up to form
the head. And if the item being charted is a futures contract, total open interest
might decline on this final rally. If so, the proverbial ‘early warning signal’ will
have been flashed.

At this juncture in the development of an H&S Top (at a price level near the high
of the head), there is no overhead resistance present - only underlying support.
The ability of a technician to properly locate where support should be present is
of paramount importance. A price decline that violates support is the first
important sign that a reversal pattern could be developing. This is a sign of
weakness in the major price uptrend. W hen this occurs, a classical bar chartist
begins to study the graph in earnest.

It cannot be emphasized enough, that the selloff from the high of the head must
violate the support level. If not, the definition of a bull market will continue to be
present. This price decline below the support (which is in the process of
becoming the high of the left shoulder) can halt anywhere in the vicinity of
reversal point two. The price decline does not have to stop exactly at the same
price level of reversal point two.

Next, a price rally (off reversal point four) must occur. Ideally this rally will occur
on lower or declining volume. It is especially bearish if the volume on this
developing right shoulder takes place on volume readings lower than what
occurred on the rally for the left shoulder. If so, the chart is really telling the
technician that the market is in the process of trying to reverse the major price
uptrend.

This last price rally in the developing H&S Top must stop below the high of the
head. Symmetry (more on this later) would suggest that the high of the right
shoulder will be in the general price area of the top of the left shoulder.

At this time, any long positions should be closed out. Note that because the
pattern has not been officially activated (with a close below the neckline), a
conservative chartist should not be attempting to pick the high of the right
shoulder and trade it via a new short sale.

9-2
Activating a Head & Shoulders Top:

The neckline is the 2-point minor trendline drawn tangent to the lows of the price
bars at reversal points two and four. These two price lows are the reactions on
either side of the head. The style convention for drawing this boundary line of the
pattern is a dashed line.

Any classical bar charting price pattern is not officially activated until a close
outside the pattern is posted. In the case of a Head & Shoulders Top this must
be a close below the neckline. Once this occurs, the chartist has strong
technical evidence that the longer term trend has changed direction.

How Far Will Prices Move:

Elementary textbooks about trading often refer to establishing a risk to reward

ratio. Sounds good, but do these books ever detail how to determine this ratio?
Answer: No.
One of the helpful aspects of classical bar charting price patterns is that a
definite measuring objective can be derived. And, most importantly, the technical
trader will know where the pattern breaks down i.e., fails. This enables a chartist
to gain insight as to what the risk is, given the expected price move.

W ith any active H&S formation (Top or Bottom), price is expected to move a
minimum of the vertical distance from the extreme of the head to the neckline as
measured from where price penetrated the neckline. Note that this distance is
not measured from the close beyond the neckline; it is the price at which the
neckline was broken.

What to do at the Measuring Objective:

A close below the minimum downside measuring objective in an H&S Top is not
required. As long as price trades below the objective any time during the
international 24-hour trading day - the pattern has worked as expected.

Traders should try to follow the old adage of “letting profits run and cutting
losses”. W ith this in mind a trader should take partial profits when the minimum
measuring objective is reached and look for additional technical signs that a
greater percentage of the open trade (short) should be closed.

9-3

For sure, a trader should have been following the market down with trailing
protective buy-stop orders. Ideally, the protective stop would be located well into
or above the top of the closest overhead resistance.

Additional signs that more open trades should be removed would be blowoff
volume, a decline in open interest (futures only), or the posting of one of the four
minor trend change indicators (see Chapter 17).

The chartist will also be looking for one of the classical continuation patterns,
Triangle or W edge, to form. This would indicate the existing price trend has
further to travel - and additional directional positions can be established.

Symmetry:

It is amazing how often the Head & Shoulders formation exhibits symmetry with
respect to the magnitude of the two shoulders (figure 9-2), time spent during the
development of the two shoulders, and the number of shoulders.
More than one shoulder on each side of the head produces a complex H&S
formation (more on this later).

Pay attention to what happens on the ‘left’ side on the chart -

because it could be duplicated on the ‘right’ side of the chart.

Figure 9-2

If a selloff penetrates this former

² price high, it is the first price signal
that H&S Top may be forming.

Then note the magnitude and

duration of the previous reaction.
If a right shoulder develops, it
may prove to be very similar.

9-4

When to Enter a Position:

After monitoring a market closely and seeing a potential Head & Shoulders
pattern developing, it is very tempting to ‘lead off’ and establish a position prior to
the official breaking (close below) the neckline. This is a dangerous practice.
An aggressive short term trader, monitoring intra-bar price activity, might see
price trading beyond a neckline within the price bar. A sharp reaction then takes
price back within the pattern. Since the close of the price bar was not beyond the
neckline, the formation was not set off. Read, whipsaw.

It takes discipline to wait for the penetration of the neckline and the confirmation
of high volume if it was an upside breakout from an H&S Bottom.

W hether or not a corrective price pullback (rally in the case of the H&S Top) will
take place is the age old question that causes difficulty for a chartist. Volume can
be a useful aid. The higher the volume associated with any breakout, the less
likely a pullback will occur.

A look at the various possible shapes of a Head & Shoulders formation (Figure 9-
4) shows that it is impossible to make a blanket statement concerning when and
where to establish an initial position. The following is a reasonable strategy.

1. Establish a 50% position on any closing penetration of a valid neckline.

2. Establish the remaining 50% position on:

A. A pullback to just below (2 minimum price tics) the neckline, or
B. A pullback to the closest overhead resistance.

Where to Place the Protective Stop:

Any H&S formation is not destroyed until the extreme of the head is taken out.
But, placing a protective stop higher than the high of the head in a Head &
Shoulders Top results in a risk to reward that is, by definition, slightly worse than
one to one. Even though the H&S formation is a highly reliable pattern (86 - 88%
reliable), this is not an attractive risk to reward ratio.

The suggestion is to construct a ‘failsafe trendline’ to gauge where a chartist

should become worried about the pattern not working.
This line is drawn tangent to the extreme of the head and the right shoulder.
The protective buy-stop would be placed just above this downsloping line on a
Head & Shoulders Top pattern. Note that the protection slides down (is
tightened) with time. This is as it should be with any protective stop order.
9-5

Additional considerations:

It takes time to reverse a price trend. The five reversals of the minor price trend
in a Head & Shoulders formation represent the battle between the forces of
supply and demand. It pictures the influx of new information or moods that are
necessary to change the direction of the trend.

Most of the charts in this manual are daily, high-low-close bar charts. This time
frame is most suited for the interpretation of volume and open interest on a
futures chart. The H&S formation ‘works’ on any time frame chart - from minutes
to months.

For the extreme time frames of minutes or months, volume is not usually a
consideration. For the day trader, it is the speed of execution that is important,
not the dutiful analysis of volume and open interest (the latter being impossible on
an intra-day chart). And for the long term position trader, the new fundamental
inputs are what drive prices.

Figure 9-3
H&S Top

9-6

Every
Head &
Shoulders
has a Different Shape
Figure 9-4

9-7
Head &
Shoulders Top on a Weekly Chart
Figure 9-5
Zinc - LME

It is interesting to note that Zinc was in a major bear market during the same time
frame that Gold and Silver were setting new all-time highs. Believing the pattern
on the Zinc chart produced the correct directional view for this metal.

9-8

Head & Shoulders Bottom

A valid Head & Shoulders Bottom formation contains two important volume
considerations:

1. Ideally volume on the price selloff to form the right shoulder will be ‘low’ or
declining.

2. The upside breakout, a close above the neckline, should occur on a noticeable
increase in volume during that trading session (or price bar).

A classical bar chartist does not trust an upside breakout of any price pattern or
2-point trendline that occurs on less than high volume. An apparent upside
breakout that has been engineered solely by technical traders will not usually
generate enough volume to be valid. Needed is the ‘smart money’, the
sophisticated fundamental traders doing enough buying such that, in combination
with any technically based traders, volume will escalate to a distinctively higher
level.

W ith specific regard to a daily futures chart, total open interest would ideally
increase on the upside breakout. Declining open interest would mean that the
price rally was due to net short covering. Short covering on an upside breakout
does not automatically invalidate the breakout; it does imply a high likelihood of a
price pullback (decline) toward the neckline.

As with the H&S Top, the minimum upside measuring objective of a bottom is
determine in exactly the same fashion. The vertical distance from the head up to
the neckline is graphically added to the neckline when a closing price is posted
above the neckline. The Swiss Franc future in figure 9-6 is an example of a Head
& Shoulders Bottom on a daily chart and figure 9-7 is an H&S Bottom on an
hourly (Gold) chart.

9-9

Figure 9-6 Head & Shoulders Bottom

 9-10

Figure 9-7 Head & Shoulders Bottom on an Hourly Chart

Not
e
the symmetry of three shoulders on either side of the low of the head.

9-11
Figure 9-8 Head & Shoulders Bottom on a Monthly Chart

This multi-year Head & Shoulders Bottom occurred in the years just after the T-
Bond future began trading. Volume is not normally analyzed on a weekly or
monthly chart. The ‘growth curve’ upward in volume can be seen as this future
gained in popularity as a trading and hedging vehicle.

Paul Volker was the chairman of the US Federal Reserve Bank at the time when
inflation was running rampant. The left shoulder selloff began with the “October
Massacre” in 1979. The Fed stopped pegging interest rates and tightened the
money supply. The prime rate reached a high of 21 1/2% - and the back of
inflation was broken. The large H&S Bottom depicts this process of increasing
long term interest rates and then a bull run in price that continued to set new price
highs as late at September 2008 (124-24) on this notional 6% coupon bond.

9-12
Complex Head & Shoulders Formations
The term complex refers to the fact that ‘two or more of something’ is found in the
pattern. This could be two sets of shoulders on either side of the head. Or it
could be Double Top for the head.

A chartist should always be thinking - symmetry. Most complex H&S formations

tend to exhibit symmetry in the number of price gyrations seen on either side of
the head.

Multiple measuring objectives are possible if several necklines exist. There is no

technical reason that the largest obtainable objective should not be considered
viable. A chartist does have to be aware that any H&S formation cannot be
expected to retrace more than the price move that preceded it. This was the
problem with measuring objective on the Copper chart at the end of this Chapter.

Figure 9-9 Possible Complex Head & Shoulders Top

9-
13
Figure 9-10 Complex Head & Shoulders Top on an Hourly Chart

9-14
 Historic Charts

Figure 9-11 H&S Top ‘failure to form’

“The most widely advertised (possible) Head & Shoulders Top

since Edwards and Magee wrote their book”

Quote from a member of the Chicago Board of Trade on the trading floor in 1982

9-15
Can Technical Analysis be used in a manipulated market?

9-16
 Manipulation ended badly

Bre-X common stock - in Canadian Dollars

Suspended May 8, 1977

Quote from the Wall Street Journal, Wednesday May 7, 1997:

“Bre-X Minerals, Ltd. shares staged a spectacular collapse to near-worthless

levels in frantic trading as investors reacted to the news that the company’s
supposedly huge gold discovery is really a highly engineered fraud.”

9-17
Example of a Head & Shoulders Failure

9-18
 Historical Chart

The severity of the price pullbacks are what make stock index futures trading so
difficult. Often price moves much farther into classical support or resistance levels
than any other futures contract. In contrast, individual equities chart very well.

9-19

Multi-year Head & Shoulders Top

 on a Monthly Chart of a U.S. Equity Index

The Dow Jones Industrial Average and the S&P 500 Index also met a Head &

Shoulders Top objective on their monthly charts during the week ending October
10, 2008.

9-20

Failure of a Head & Shoulders Bottom to Reach

 Its Measuring Objective - and Why

Any H&S formation, Top or Bottom, cannot be expected to re-trace more than the
price move that preceded it. This price extreme is the limiting factor if application
of the standard height objective results in a price beyond where the previous price
move began. This was the case in the Copper chart below.

Adding the vertical height of the pattern, 5.04, to the neckline break, 100.00,
resulted in a conservative (arithmetic) upside price target of 105.04. This was
higher than where the previous price downmove began - at 104.10.

The May Copper did ‘attack’ 104.10 with the 103.62 daily high. But, the traditional
objective at 105.04 was never reached.

Copper - May 1992

9-
2
1

Notes
9-22
 Chapter Ten

Broadening Formations

As a chartist gains experience, the pitfalls associated with each particular price
pattern are encountered. There is a reversal formation known as a Broadening
Formation that contains five reversals of the minor price trend - just like the Head
& Shoulders pattern. A theoretical example of a Broadening Top can be seen in
figure 10-1 and an actual example in figure 10-2.

Because the fifth reversal of the minor price trend takes place beyond the
extreme of a possible ‘head’, it keeps the technician ‘honest’ in not jumping to the
premature conclusion that an H&S is forming every time four reversals of the
minor trend are in place.

There is no way to know beforehand that a Broadening Formation might be

developing. Although, a downward sloping neckline when a possible H&S Top
might be forming and an upward sloping neckline when a possible H&S Bottom
might be forming is a big clue.

Patience and knowing that such a pattern does exist (although not frequently) is
the first step. A well thought out game plan for an entry into a new position is a
must.

According to the old Edwards & Magee textbook, there is no specific measuring
objective. Ken Shaleen, a.k.a. CHARTW ATCH will suggest using a standard
‘double’ measuring objective using the last two price extremes. This technique
worked out to the exact point in the Japanese Yen example in Figure 10-3.

One further note. The price volatility associated with any Broadening formation is
more likely at a top in any of the physical commodity futures. This is why it is
unusual to find this formation at a price bottom on this type of chart. Price
bottoms on the physical commodity future tend to be more ‘rounding’.

Carrying on the notion of where price volatility would be expected, in an interest

rate future, it would be at high rates - i.e., low prices. This can be seen on the T-
Bill chart in figure 10-4.

On the currency futures or spot foreign exchange charts the volatility could be at
either extreme. This is what happened on the Yen future in figure 10-3.

10-1
Figure 10-1 Theoretical
Broadening Top

Figure 10-2 Broadening Top

10-2
Figure 10-3 Broadening Bottom

10-3
Figure 10-4 Broadening Bottom and Complex H&S Top

10-4
 Chapter Eleven

Double Top and Double Bottom

Double Top

The definition is simple, yet the misunderstanding involved with the Double Top is
extraordinary. ‘Two price highs at approximately the same price level’ is only
one-half of the definition. Prices must decline below and close below the low
point established in-between the two price highs before a Double Top pattern has
been officially activated.

Figure 11-1
Double Top

Many traders see a price move stop in the same general area of a previous price
reversal. But what trading usefulness does this have? Should a short (or long in
the case of a possible Double Bottom) be taken? The answer is no. A new
position should not be initiated until the pattern has been activated.

Before proceeding with examples, Ken Shaleen a.k.a. CHARTW ATCH must
share some historical observations:

1. A ‘double’ formation, Top or Bottom, is not a common pattern on a future or

forward chart. The pattern is more prevalent on a cash or spot market chart -
where the basis convergence in a futures contract is not continually pulling the
future toward the cash price.

2. A ‘double’ formation is not all that reliable price pattern. Often the pattern is
activated and some price progress toward the measuring objective is made - but
the minimum objective is not reached. This negative does not automatically
mean that a technician trading this pattern will suffer a loss. Proper money
management and the movement of a protective stop order can mitigate the poor
performance of the pattern.

11-1
The measuring objective of a Double formation is straight-forward. In a Double
Top the vertical height of the pattern is subtracted from the price low in-between
the two highs or lows.

To determine the height of the pattern, a straight line (a 2-point trendline) is

drawn tangent to the two price highs. The height of the pattern is from the low
point in-between the twin highs up to the line. This is a method of ‘graphically
averaging’ the two highs if they are not exactly the same price. This is
schematically shown in Figure 11-1.

After the breakout there may be a price pullback, rally in the case of a Double
Top. If so, it is easy for a chartist to determine the location of overhead
resistance. It is the now former price low in-between the twin highs. A
conservative trader might wish to wait before initiating a new short until a pullback
to test the overhead resistance is posted. Note that there was no pullback on the
weekly Euro-fx chart in Figure 11-2.

Figure 11-2

11-2
Double Bottom

In a Double Bottom, price must close above the high point in-between the two
(approximately equal) price lows before a trading decision can be made.
Because this is an upside breakout, it must occur on a noticeable increase in
volume. A chartist should not trust any upside breakout that takes place on sub-
standard volume.

Once an official upside breakout is posted, the upside measuring objective is

clear. Price should rally to as far above the high point as the vertical distance in-
between the twin price lows. This is schematically shown in Figure 11-3.

As usual, there may or may not be a price pullback following the upside breakout.
It is always difficult to know whether a pullback is going to be posted. Thus, a
trader has to weight the opportunity loss of not initiating a new long on the
breakout. Yes, the risk / reward of a new long placed only a pullback is better -
but, the pullback might not happen.

W ith either the Double Top or Bottom, a protective stop should be placed far
enough beyond where the classical resistance or support begins - such that a
pullback would have to do an inordinate amount of damage to the expected
resistance or support level.

Figure 11-3
Double Bottom

11-3
 Double Bottom and H&S Bottom
on ‘non-storable’ commodity futures charts

11-4
This is a classic chart. Note the typical open interest action in the transition from
a bull to bear market. After the downside breakout from the Double Top, look at
where the pullback stopped - at the overhead resistance.

11-5
 Should Price Pattern Recognition be Applied
to Options Charts?

Price plot of
June 70 Call
option

Price plot of
underlying
instrument -
June 1985
T-Bond future

The answer to the

question posed above is - No

Because an option is a ‘wasting asset’ the bar charts do not lend themselves to
classical price pattern analysis. A trader should do all the technical work on the
chart of the underlying instrument - and then pick an options strategy suitable for
trading that pattern.

Time erosion caused the price of the 70 strike T-Bond Call option to decline much
farther than the pullback after the price decline into the Double Bottom occurred
on the chart of the underlying instrument.
11-6
 Chapter Twelve

Continuation Patterns

The Symmetrical Triangle Formation

Figure 12-1, Valid Symmetrical Triangles

In the ‘art’ of classical bar charting, the Symmetrical Triangle formation is the premier
continuation pattern. It is created by a net sideways price movement on the chart,
after which, a significant directional price move ensues. Note that this new price
move can be in either direction - but there is an overwhelming tendency for a valid
triangle to act as a continuation pattern. This means that the prevailing price trend
that was in progress prior to the triangle forming, continues.

10-1
Most of the time markets are trending. The major price trend is not continually
reversing from bullish to bearish and then reversing again. If the Symmetrical
Triangle is best classified as a continuation pattern, it is not surprising that this
pattern is the most common of all bar charting formations. It ‘works’ on all time
frame charts - i.e., 5 minutes to monthly.

Directional Considerations: W hen this formation occurs on a high-low-close bar

chart, it leads to a continuation of the prevailing price trend approximately 75+% of
the time. Note that this pattern can be a reversal pattern - although this occurs in
less than 25% of the cases.

It is the lower probability case where the triangle acts as a reversal pattern that
keeps a chartist ‘honest’ by not always taking a directional position within the pattern
prior to the breakout.

Shape: It only takes two points to define a straight line. Two converging line
segments form a proper Symmetrical Triangle. Four reversals of the ‘minor’ price
trend are necessary to create the two converging boundary lines. Each line must
slope in a different direction. The lower boundary line must slope up; the upper
boundary line must slope down. The two lines intersect at the apex of the triangle.

Although the definition of a Symmetrical Triangle is not so strict as to require that the
triangle be equal lateral (line segments of the same length) or equal angular (same
angle for each converging line), the word symmetry is important. One side of a valid
Symmetrical Triangle should not even approach twice the length of the other side.

Figure 12-1, Valid Symmetrical Triangles shows the ideal bearish and bullish
formations. Any triangle is not an active pattern until a closing tic mark is posted
beyond one of the boundary lines. (More on this later.)

Where to Locate Reversal Point One: The first reversal point (1) in any valid
triangle is always at a relative price high in a bull market or at a relative price low in
a bear market. Remembering this simple rule will keep a trader from incorrectly
assuming that a valid triangle is present. Figure 12-2, An Incorrect Interpretation
shows how a bad mistake can be made. The more likely outcome in Figure 12-2 is
a Head & Shoulders Top rather than a Symmetrical Triangle as a bullish continuation
pattern.

12-2
Figure 12-2, An Incorrect Interpretation

Volume Within the Triangle: A general statement can be made with respect to
volume within any triangle: It usually moves irregularly lower. This is common
sense. Because the price swing near reversal points 1 and 2 are greater than the
price swings near reversal points 3 and 4 it is not surprising that volume would
typically be greater at the beginning of the net side ways price movement.

A chartist should not use a microscope, looking for nuances in volume within the
consolidation pattern. The downsloping line overlaying the volume in the chart of
Exelon common stock (Figure 12-3) is a reasonable example of how volume
contracts within a symmetrical triangle.

12-3
Figure 12-3, Exelon Common Stock

10-4
Breakout Volume: One of the first things that any novice bar chartist learns is that
a valid upside breakout from any price pattern should occur on a noticeable increase
in volume. This is due to the idiosyncrasy of the trading public to prefer to be on the
buy side rather than on the sell side. A low volume upside breakout is not to be
trusted.

Figure 12-3, Exelon Common Stock shows volume of 2,807,400 shares on the
‘second breakout’. Although this was not a major jump in volume, it was higher than
the ‘first breakout’ from the inner Symmetrical Triangle. The section on Re-drawing
the boundary lines later in this chapter explains what happened with the two triangles
that formed.

Downside chart breakouts (a close beyond the lower boundary line of the triangle)
do not have to occur on increased volume. Often volume increases a trading
session or two after a downside breakout. This is because the market is making it
painfully aware to the bulls that they are ‘long and wrong’. The undercapitalized
longs are forced to liquidate their losings positions; volume increases during this
panic exit.

Location of the Breakout: Another vital consideration of any triangle breakout

is its location with respect to time The breakout cannot occur too far into the pattern.
Too far is defined as more than 3/4 of the horizontal distance from reversal point one
to the apex. If a breakout is posted beyond this 3/4 distance, the triangle loses its’
effectiveness. Anything can happen. Price could move up, down, or sideways.

Measuring Implication: The majority of classical bar charting price patterns

utilize the vertical height of the pattern to gauge the future extent of the price move.
This convention is also applied to the triangle formation.

The vertical height of any triangle is always measured at reversal point 2 to the
opposite boundary line. This is true whether the triangle is found in a bull or bear
market. This distance is added to or subtracted from the price of the boundary line
on the price bar which posted a close outside the triangle. This is minimum distance
price is expected to move.

Figure 12-3, Exelon Common Stock shows that the vertical height of the largest
Symmetrical Triangle measured straight up from the 59.75 low was 1.87. Adding
this distance to the 60.76 breakout price yielded an upside measuring objective of
62.63. This arithmetic objective was met.

10-5
Old time chartists often use a graphic method of obtaining an objective. This takes
the vertical height of the pattern and moves it over to the breakout. Note that this
method of obtaining an objective will result in the same objective if the price scale
used is arithmetic. Use of a logarithmic scale for price will result in higher absolute
upside objectives and not as aggressive downside objectives compared to an
arithmetic scale chart.

Futures traders use arithmetic price scales and equity chartists tend to use
logarithmic scales.

Pullback?: The question of whether or not a price pullback (after a valid breakout)
will occur is always difficult for a chartist to answer. Volume may provide some
insight. For sure, a technical trader must keep the size of any position small enough
to withstand the adversity if a substantial price pullback occurs.

After a valid breakout, underlying support (for an upside breakout) is created at the
price level of the breakout. Additional support should be present at the boundary
line just penetrated (Note that this line is moving farther away during a pullback).

Support should also be present at the price level of the apex. There is nothing
special about the location of the apex in time, after a breakout.

The last bastion of support (again, in the case of an active bullish pattern) is the
opposite (lower) boundary line of the triangle. A violation of the opposite boundary
line or its extension beyond the apex, even intra-price bar, officially destroys the
triangle.

This last support (line) gives the technical trader a worst case scenario for the
triangle - with it still ‘working’ as expected. Thus, in theory, the protective stop order
should be placed just beyond the opposite boundary line. The good news is that this
stop order is continually ‘tightened’ as time passes.

Reliability: An educated guess as to the reliability of an active Symmetrical

Triangle meeting it’s minimum measuring objective is 76 - 78%. This probability is
about 10 percentage points below the reliability of a valid Head & Shoulders
formation. The H&S pattern (top or bottom) is the most reliable of all classical bar
charting price patterns.

10-6
Redrawing the Boundary Lines: A ‘fact of life’ concerning the triangle formation
is that sometimes reversal points 3 and 4 have to be relocated. This usually
happens when the two boundary lines of a possible triangle converge too quickly.
This is shown in Figure 12-4, Re-drawing the Boundary Lines.

Figure 12-4, Re-drawing the Boundary Lines

The more difficult case (of having to relocate the boundary lines) is contained in
Figure 12-3, Exelon Common Stock. Exelon did experience what looked like a valid
price breakout to the upside on October 10. Admittedly, volume at only 2,268,300
shares was not impressive. The price move up after the breakout did not exceed the
first point in the triangle. Then, a severe price pullback(decline) occurred; the
pullback entered the triangle and destroyed the formation by violating the
opposite(lower) boundary line. However, the extent of the pullback did not take out
reversal point two (59.75). An ensuing price move in the direction of the major price
trend (up) created the conditions where a second set of reversal point 3 and 4 were
present.

10-7
Re-drawing the two boundary lines on the Exelon chart, continuing to use the original
points 1 and 2, resulted in a larger triangle. It became a viable Symmetrical
Triangle.

It is this sometime problem of re-locating reversal points 3 and 4 that drops the
reliability of the triangle formation below that of a Head & Shoulders reversal pattern.

Another Method of Obtaining a Measuring Objective: A chartist can also

construct a ‘parallel objective line’ to construct a reasonable triangle measuring
objective. This line is derived from a point and an angle. The line begins at reversal
point one and increases/decreases at the same angle as the opposite boundary line
of the triangle. This is shown in Figure 12-5, Another method of obtaining a Triangle
measuring objective.

The assumption is that price will touch this parallel objective line. It is a secondary
objective. To resolve the dual techniques of obtaining a triangle measuring
objective, a chartist could plot both targets. W hen the closest target is reached,
existing positions can be partially liquidated. The market would then be followed by
a trailing protective stop order.

Figure 12-5, Another method of obtaining a Triangle measuring objective

10-8
What to do when the Measuring Objective is Reached: As the economists
say “all other things being equal”, 3/4 of the long or short position should be closed
out when the height measuring objective is met. Note that a close beyond the
objective is not required. The market must simply trade at or beyond the objective
any time during the international 24-hour trading day.

The reason that 1/4 of the trade should be kept open is because of the word
‘minimum’ used in conjunction with the measuring objective. The price move could
go much farther. A trader is trying to follow the old adage of letting some profits run.

Additional technical factors such as blowoff volume, declining open interest (for a
futures contract) or a Key Reversal price posting (a minor trend change indicator)
would prompt a trader to remove 100% of all directional positions.

Other Continuation Patterns: If either boundary line is horizontal, the pattern

that forms is a relatively rare variation of the triangle; a 90 degree angle is present.
This results in an Ascending (bullish) or Descending (bearish) continuation pattern.
The market ‘wants’ to penetrate the horizontal line.

The Right Angle (90E) form of triangle is found more frequently on cash (spot)
market charts than on futures charts. This is due to basis convergence that is
constantly occurring on any futures chart. This is the narrowing of the difference
between the cash price and the futures price as time passes.

If both lines slope in the same direction, yet still converge, the resulting pattern is a
Rising or Falling W edge. Sometimes two entities that tend to move in the same
direction with one another will exhibit a Symmetrical Triangle on one chart and a
W edge pattern on the other. They are both continuation patterns.
Examples of right angle formations and wedges are located in chapters 13 and 14.

Conclusion: The Symmetrical Triangle is a common bar chart pattern. W hen two
reversals of the minor price trend are present, the technical trader receives a ‘get
ready’ signal. There should be a profitable trade ahead - but patience is required.
If the next two reversal points create the proper conditions for a Symmetrical
Triangle - the risk / reward parameters for a directional trade appear.

And, food for thought: A trader equipped with a knowledge of option strategies has
a much larger arsenal of positioning techniques within a Symmetrical Triangle prior
to the breakout. A ‘plain vanilla’, all or nothing, long or short only trader must wait
for the breakout before entering a new position.

10-9
Example of a Symmetrical Triangle

Crude Oil - Dec 2008

Light - NYMEX

10-10
 Chapter Thirteen

Right Angle Triangles

A Right Triangle is so named because one of the boundary lines is horizontal,

and a 90-degree angle is present - a right angle.

As a variation of a Symmetrical Triangle, the same number of reversals of the

minor trend (four) are necessary before the two converging boundary lines of a
Right Triangle can be constructed. The expected direction of the breakout is thru
the horizontal boundary line of the pattern. The two forms of Right Triangles are
shown in Figures 13-1A&B.

Figure 13-1A
Ascending Right Triangle . . . is a bullish pattern (price will ascend out of it)

Figure 13-1B
Descending Right Triangle . . is a bearish pattern (price will descend out of it)

13-1
The risk and reward parameters in a Right Triangle are distinctly defined.
Assessing the theoretical risk in any new trade is of paramount importance. The
risk parameter in a trade based on any active Triangle pattern is a price move
beyond the opposite boundary line. In a Right Triangle, the initial protective stop
loss order would ideally be placed slightly (say 2 minimum price tics) beyond the
sloping boundary line.

The reward parameter of any Triangle pattern is the same. Price is expected to
move beyond the boundary line penetrated, by the vertical height of the pattern.
This height is always measured from reversal point two - to the other side of the
Triangle.

Because the expectation is that price will breakout thru the horizontal boundary
line, a chartist / trader can ‘lead-off’ by placing a directional position within the
Right Triangle prior to the breakout. Note that this is in contrast to waiting for the
breakout from a Symmetrical Triangle - even though the high probability (75-80%)
is for the Symmetrical Triangle to act as a continuation pattern.

As an experiment, figure the approximate risk to reward of taking a directional

position within a Right Triangle prior to the breakout. It will be a risk of 1 to a
reward of 3. This is such a favorable ratio, that a trader will want to place this
position every time a Right Triangle is present.

Be careful, Right Triangle are relatively rare patterns on forward or futures charts.
As with the Double Top or Bottom formation, Right Triangles are likely to be more
prevalent on a cash or spot chart - where constant basis convergence between
the futures and cash market is not present.

And, as usual, markets tend to fall faster than they move up in price. This
idiosyncracy is based on the outside public tending to favor trades on the long
side of any market. Thus, it is important to place short positions within a possible
Descending Right Triangle prior to the breakout. If not, a gap down, thru the lower
boundary line, could then lead to the measuring objective being met - on the
same price bar as the breakout!

13-2
 cending Right Triangle
Eurodollar Time Deposit Futures
June 1989

(Objective was met)

13-3
 Ascending Right Triangle
US Treasury Bond future
Dec 1982

The time period during which the Symmetrical Triangle formed is examined on
Point & Figure chart in Chapter 18.

13-4
 Descending Right Triangle

The parallel objective line on the Cotton chart was not reached - but the traditional
height measuring objective was met.
A chartist can place both objective on the chart - and begin removing positions
when the first objective is reached.

13-5
Notes

13-6
 Chapter Fourteen

Wedge Formations

A W edge formation is a continuation pattern. Price is expected to continue in the

same direction it was going when the W edge began to form.

As with the premiere continuation pattern, the Symmetrical Triangle, a W edge

requires four reversals of the minor trend for the two converging boundary lines to
be constructed.

A W edge begins its development in similar fashion to the Triangle; however, the
fourth minor reversal of the trend is such that both boundary lines slope either up
or down.

Rising Wedge

A Rising W edge is a bearish pattern.

Both converging boundary lines slope up. A close below the lower boundary line
of the Rising W edge activates the pattern. The objective is to take out the lowest
point in the formation (reversal point one). A schematic diagram of a Rising
W edge is shown in Figure 14-1.

Figure 14-1
Rising Wedge

14-1
Figure 14-2

Rising Wedge - Gold, Dec 2008 NY

14-2
Falling Wedge

A Falling W edge is a bullish pattern.

Both converging boundary lines slope down. A high volume close above the
upper boundary line is necessary to activate the pattern. Note that high volume is
necessary to validate the upside breakout from the Falling W edge but is not
necessary (although ideal) on a downside breakout from a Rising W edge.

The minimum upside measuring objective of a Falling W edge is to take out the
highest point in the pattern (reversal point one). A close above reversal point one
is not required - only an intra price bar move above it. Figure 14-3 is a schematic
diagram of a Falling W edge.

Figure 14-3
Falling Wedge

Figure 14-4
Falling Wedge
Plywood futures

14-3
Falling Wedge on the Gold / Silver Ratio

Monthly Chart

14-4
 Rising Wedge

Soybean Oil
March 1981

Note the quote from the Edwards and Magee book. The Rising W edge is a very
powerful sign that a bear market is in progress.

14-5
 What if a boundary line is very close to being horizontal?

W hen a boundary line is so close to horizontal that there is a question of what

pattern is developing, additional charts of markets that are thought to move in
concert must be analyzed.

14-6
 Chapter Fifteen

Flags & Pennants

‘The Flag Flies at Half Mast’

A Flag is a dynamic and potentially very profitable price pattern. Flags occur
within the course of a ‘rapid, straight-line’ price move. They mark a half-way,
breath-catching, resting place before the prevailing trend resumes.

Flags normally slope against the trend. In an uptrend, the ‘body’ of the Flag
slopes downward; several low volume small range down days produce the body
of the Flag. In a price downtrend, the Flag is composed of several narrow range
price up days. Figures 15-1A&B show a theoretical Bull and Bear Flag.

Figure 15-1A
Flag in a Bull Market Figure 15-1B Flag in a Bear Market

15-1
On a daily futures chart, the body of the Flag seldom lasts more than five trading
sessions before a resumption of the trend occurs. W hen an apparent breakout is
occurring, a quick trading decision is necessary. This because a limit move that
trading session is frequently the result. On an individual equity chart, the body of
the Flag can extend up to weeks in duration. Everything moves faster when the
item is listed as a futures contract.

The beauty of a Flag is that the risk is small compared to the expected reward.
In addition, the trader will know quickly if the newly placed position is a winner or
loser, and is not faced with an agonizing wait for confirmation of the wisdom of
the trade. There should never be a ‘pullback’ (reaction) back into a Flag pattern
after a valid breakout.

One hundred percent of all existing positions predicated on the Flag pattern
should be removed once it has reached its minimum measuring objective of
duplicating the rapid straight-line price move that preceded it. Historical
observation has shown that once a Flag objective has been reached, a violent
price move in the opposite direction often occurs. Any trailing protective stop
order would more than likely be executed. Therefore, a trader should take profits
at the objective and try not to give back the significant gains that were made in so
short a time period.

Three caveats:

1. Because of the dynamic nature of a Flag pattern, a chartist will ‘want’ to find
this pattern as often as possible. Be careful, it is not a common pattern. A Flag
should really only ‘fly’ in new life-of–contract high or low ground. A proper Flag
will not be found within a trading range. Look to the left on the chart; ideally there
will no price activity in the price range of the Flag for quite some time.

2. Flags work best on daily time frame charts. The suggestion is not to look for
this formation on intra-day charts. The reason is that a shock variable (from an
economic report) will cause a financial future to move swiftly on say, a 10 minute
chart. This will make the 10 minute chart ‘look’ like a possible flag pole is
present. After moving net sideways for several price bars is there another shock
variable to sent the market an equal distance away? Usually not.

3. No market can go below zero to the downside. Therefore, a conservative

method of obtaining a downside measuring objective is to measure down from
the high in the body of the Flag instead of from the Flag breakout. Notice this
nuance in figure15-1B and the actual application on the Silver chart in fig. 15-3.

15-2
 Bear Flag

Silver - Dec 2008

New York
Figure 15-3

Note that the start of the flag pole on the Silver chart in figure 15-3 began at the
breaking of a 2-pont trend (the two points used to construct the line are off the
chart to the left). The measurement did not start at the relative price high of
1809.50. The measurement always starts at a previous breakout.

Also note that because this commodity is bounded by zero on the downside, a
conservative measuring objective subtracts the 265.80 point flag pole distance
from the 1523 high in the body of the Flag on this arithmetic scale chart.

15-3
 Example of a Flag Failure

Four trading
sessions
later

15-4
Bull Flag

15-5
 Comparison of Hourly and Daily Charts

U.S. Treasury Bond futures

June 1985

Hourly Daily

15-6
Chapter Sixteen

Gap Theory

16-1
Four Types of Gaps

16-2
16-3
Gap Theory - Graphical Summary

16-4
Suspension Gap

16-5
Ex- Dividend Gap

16-6
Island Formation

16-7
 Island Top

Chesapeake Energy Corp (CHK)

16-8
 Gap Theory Exercise

Gilts Dec 1995

Classify this gap

See the next page for the answer and ‘how it came out’.

16-9
 How it Came Out

Gilts Dec 1995

(Two days later)

Because the close was within the trading range, the gap was a Pattern Gap.

It was closed the next trading session.

16-10
 Suspension Gaps

Suspension Gaps, within a 24-hour bar will only appear on a hand constructed
chart - until the chart package programmers add a considerable amount of
sophistication to their algorithms.

See page 2-7 for a mechanically reproduced Corn chart.

16-11
 Gap Analysis

Soybeans - Nov 1999

Because the Triangle objective had been met, the gap between 435 and 440
initially looked like it could be an Exhaustion Gap. W hen it wasn’t quickly filled - it
had to be re-classified as a Measuring Gap.
Also see the Corn chart on the previous page and the mechanical version on page
2-7.
16-12
 Chapter Seventeen

Minor Trend Change Indicators

Market participants with short term trading horizons, e.g. day traders, often ask:
How can bar charts be utilized for their type of dealing? One of the answers to
their query lies in the short term forecasting significance of minor trend change
indicators. There are four of them:

1. Key Reversal
2. Inside Range
3. Outside Range
4. Mid-Range Close

It must be emphasized that these configurations are helpful in gauging only what
might happen the next trading session. In fact, the next trading session could
even be the next major world time zone! As such, they change the minor trend in
the opposite direction for as little as one price bar. The end of a major price
move could be signaled by one of the minor trend change indicators - but this
assumption must not be given too much weight.

The first step in applying any of these indicators is the identification of the
direction of the ‘minor’ price trend. A technician will examine the previous three
price bars. If there are three higher highs, the minor trend is considered to be up.
If one of these one-bar indicators is then posted, the expectation is that the next
price bar will register a lower low than the indicator’s price bar low.
Note that nothing is being predicted for the location of the next price bar’s close.
If the minor price trend is down (three consecutive lower highs), and a minor
trend change indicator appears, a higher high is expected.

Figure 17-1
Key Reversal.

17-1
Key Reversal
The definition of a Key Reversal High, after a series of higher price activity, is a
new high and a lower close. It forecasts a lower price low.
A Key Reversal Low, after a series of lower price activity, implies that a higher
high should be set. Figure 17-1 depicts both of these conditions.

The Key Reversal on a daily chart is one of the most widely known yet least
understood reversal pattern. A high percentage of traders jump to the erroneous
conclusion that a Key Reversal marks the end of a major price move. It might,
but this is asking for too much. Regarding it as minor trend change indicator will
allow for the construction of a much safer trading plan.

In most cases, the minimum measuring objective inferred by any of the four minor
trend change indicators is fulfilled during the next price bar. It must be noted,
however, that a more rigorous definition of ‘worked as expected’ means that the
objective was reached before the one-bar pattern is destroyed. This means that
several price bars could be posted before the objective is achieved - as long as
the pattern is not destroyed.

As an example, figure 17-2 contains a Key Reversal (K-R) at a price high. The
expectation was that a lower low, than the K-R, should be posted. Note that the
next price bar, after the K-R, did not achieve this objective. But neither did this
next price bar cause the K-R to ‘fail’. A failure would mean that a higher high was
set. The example continues with the second price bar after the Key Reversal
setting the expected lower low. The pattern worked. The Inside Range that was
posted after the Key Reversal simply perpetuated the original indicator.

Figure 17-2
Key Reversal that ‘worked
as expected

These one price bar reversal indicators

tend to work best on daily bar charts.
But, a trader should not overlook their
significance on an hourly or weekly
chart.

17-2

Why should these price configurations work?

They all indicate that the forces of the bulls and bears have reached an
equilibrium. The forces previously in control of the market are losing momentum
but opposing forces have not gathered enough strength to turn the trend; the tide
will turn in the next price bar.

The analogy of a ‘see-saw’ is applicable. Both sides are in balance (temporarily),

but the rival forces are about to gain the heavier weight to move prices in the
opposite direction - for at least one price bar.

Key Reversal on a Daily Chart

Several additional criteria are important to the understanding of the Key Reversal
phenomenon. W hen using a daily high-low-close bar chart, trading off a presumed
Key Reversal early in the trading day can be extremely hazardous. Seldom does a
Key Reversal make itself evident early in the dealing day. More often, the price
move to a higher or lower close occurs in the last 20 minutes of trading. Thus,
evidence of a Key Reversal early in the day is more likely to result in prices
making another violent move to a new low or high - before the late move in the
opposite direction.

Note the “anatomy of a Key Reversal” on the daily & hourly charts of the
NASDAQ-100 future on October 10, 2008 in Figure 17-3. Price was down, then
up, then down and then finally up in the last hour of trading.

Volume Considerations

By generating the wicked price gyrations, often expanding the range for the day
several times in both directions, high volume will result. Thus, a technician does
not trust a Key Reversal that occurs on ‘average’ or ‘low’‘ volume. In the old open
outcry environment in the futures markets this required access to a knowledgeable
source down on the trading floor for a guesstimate of volume activity during the
trading session. The move to electronic trading, with on-line volume, has made
the interpretation of this valuable internal variable much easier.

17-3

Anatomy of a Key Reversal

 NASDAQ-100 eMini Dec 2008 future
October 10, 2008

Figure 17-3A Daily Chart Figure 17-3B Hourly Chart

17-4

Open Interest Considerations

W hy does a Key Reversal take place? Because the losers are finally giving up!
They are forced to wait (sweat) almost all day with their losing positions, and
finally surrender - at any price, just to get out.
he old adage on an open outcry trading floor was that:

“When the last weak short is finished buying, the silence is deafening”

There are no more buyers - so price plummets. This situation results in open
interest declining. The participants with the correct market judgement going into
that trading session are taking profits; the loser is getting to the sidelines. Both
sides liquidate.

Thus, a technician does not trust a Key Reversal if open interest increases. The
change in open interest will not be known during the trading session.
Confirmation of the drop in total open interest will not be known until the next day.

Inside Range

W hen an entire price bar’s activity is contained within the previous bar’s price
range, it generates an Inside Range. The prior minor tend direction is expected to
be reversed - usually the next price bar. For example, after three or more higher
highs, an Inside Range appears; the next bar’s low is expected to be lower than
the price bar that formed the Inside Range.

The rationale is as explained previously. In the example above, the bulls have
been in control of the market but cannot force price to a higher high during the
Inside Range. By the same token, the bears have not gathered enough strength
to reverse price to the downside by generating a lower low. The forces are in
balance. It is likely that the momentum will shift to the bears for as short a
duration as one price bar. This will take prices down the next price bar, resulting
in a lower low than the Inside Range price bar. This is shown in the diagram on
the left in figure 17-4.

17-5

Figure 17-4
Inside Range
Outside R
ange

A widely swinging market resulting in a price range where the high is higher and
the low is lower than the previous price bar’s activity is known as an Outside
Range. It is expected to reverse prices - usually the following price bar.

Both supply and demand forces are evident, but it appears that the forces are
becoming more balanced and the prior trend will be interrupted, if only temporarily.
This is shown in figure 17-5.

Figure 17-5
Outside Range

17-6

Mid-Range Close
W hen prices close equidistant between the high and low of the price bar, a Mid-
Range Close is formed. Traders did not know where to close prices; the trend
previously in progress was not dominant going into the end of trading for this time
period. There is an excellent probability that an adverse move against the minor
trend is about to occur.

A technician should look for the opposing forces to gain the stronger hand - at
least for one price bar - via moving prices in the appropriate direction beyond the
range of the Mid-Range Close price bar.

W hat if the close is, say, 24 tics under the high of the price bar and 25 tics above
the low of the same price bar. And the market trades in one-tic minimum
increments. An exact Mid-Range Close is impossible. This is ok. The theory that
the market is trying to change its short term direction should still hold. Figure 17-6
contains examples of Mid-Range Closes.

Figure 17-6
Mid Range Close

Ther e
some markets in which it is ‘too easy’ to post a Mid-Range Closes - and therefore
the indicator should not be trusted as much as normally would be expected. An
example would be the Eurodollar Time Deposit futures. Often the daily range is
so small that, by default, the close is close to a mid-range.

The opposite type of market would be the stock indices. Seldom does the close of
the day land in the middle of the range. W hen it does happen, the indecision that
surfaced that trading session is likely to result in a reversal of the minor trend.

17-7

Locate the Minor Trend Change Indicators on this Chart

 and record whether or not they ‘worked’ as expected

Nickel,
LME 3-month, daily

1
7-
8

17-8
 Chapter Nineteen

Mathematical Models - Trend Following

Historical Overview

The mid-1970's saw the creation of ‘commodity funds’. Their proliferation was of
such magnitude that hundreds of millions of US Dollars was being traded by
these funds. Unlike the next influx of commodity funds just after 2000 that were
long only ‘index funds’ touting physical commodities as an asset class, the first
crop of funds were ‘directional’. They could be long or short.

Often the traditional commodity funds were operated by a very small number of
traders. How could a single trading advisor be knowledgeable about the
fundamentals affecting the 40+ actively traded futures contracts at that time?
The answer is that he could not. The trading advisors turned to mathematical
models for entry and exit signals for the futures in their portfolio.

There are two diametrically opposed mathematical approaches the directional

traders could use: Trend Following vs. Relative Strength. This latter category has
numerous additional subtitles such as Over Bought / Over Sold, Momentum,
Oscillators, and Stochastics.

A good estimate is that 90% of all managed money in the futures market that is
traded as a fund uses a Trend Following system. The rational is:

1. Try to be aboard every new emerging price trend.

2. Follow the old adage of allowing profits to run while cutting losses.
3. Utilize diversification to limit risk.

This chapter will discuss trend following techniques, the next chapter will discuss
the Oscillator concepts.

19-1
 Trend Following

Even though it’s possible to see sustained price trends just by looking at a chart,
technical trend indicators can help confirm that a market is in a trend. One of the
best and most commonly used trend indicators is the moving average. A moving
average distills market action into a single line on the chart, smoothing out the
sharp peaks and valleys that usually occur in futures (or equity) price movement,
and resulting in an easier-to-read picture of price direction

A Simple Moving Average

A simple moving average is merely the arithmetic average of price data during a
pre-determined period of trading. The more data included in the average
(generally meaning the longer the trading period), the slower the average moves
relative to the market itself (because each individual piece of data has less
impact). Although traders could use computers to calculate moving averages
based on every single price quote for a desired time period, most just use the
closing price.

For example, to calculate a simple 10-day moving average, add all the closes for
that period (10 data points) and divide by 10. Then, plot each day’s moving
average calculation (or whatever time period is being used) on a bar chart, and
connect each point with a line to get a smoothed version of the price action.

A moving average is recalculated every day. Price data from the oldest day (or
period) in the average is dropped (i.e., not used in the new calculation) and the
new day’s data is used instead. Thus, for a simple 10-day moving average, only
the most recent ten days’ closing prices are used, the eleventh day is dropped,
and the average "moves" or fluctuates with the market (but at a slower pace).
Analysts determine the time period, or the amount of data, used to calculate a
moving average based on whether they’re seeking a long-, intermediate- or short-
term view of the market. The most popular futures moving averages generally fall
in the 10- to 40-day range, although they can be calculated for very short time
periods, such as 5- or 10-minutes. The time frame also depends on the type of
markets being studied. For example, a 200-day moving average has traditionally
been popular in analyzing stock market trends, because stock indexes have, until
recently, moved rather slowly and stayed in long-term trends. But 200 days are an
eternity for the futures markets, and that time frame would be of little value if the
goal is identifying current trends.

19-2
The plot of a simple 10-day moving average using the daily closing price has been
placed on the daily chart of the Dec 2007 Euro-fx future.

Identifying Trends

Analysts have come up with all sorts of rules and conditions to determine whether
a market is in a trend — but, basically, a market can be considered in an uptrend if
prices are above the moving average line and the moving average itself is trending
higher. It’s the opposite for a downtrend. An uptrend is considered broken if prices
close below the moving average; a downtrend is broken if prices close above the
moving average. This provides the basis for a ‘trading system’.

19-3
In practice, technical analysts often experiment with more than one moving
average (computers make this easy) to see which one "fits" the price action best,
a practice that is called "optimizing." For example, in an uptrending market, prices
might consistently bounce off of the 40-day average but consistently violate the
20-day average; this tells the analysts that in this market a 40-day moving average
is a better trend indicator. This type of "optimization" is not without pitfalls,
however. Traders must keep in mind that markets often change character and
volatility. W hen that happens a moving average that had been giving fairly
accurate trend signals might suddenly become "obsolete" while another moving
average based on a different time period could exhibit a better fit. Note what
happened to the simple 10-day moving average on the Dec 2007 Euro-f`x futures
chart when price moved net sideways in a triangular consolidation.

In nicely trending markets such as the June 2007 Canadian Dollar future, a 40-day
moving average only had two whipsaws. They occurred when the long term price
downtrend was in the process of changing to a long term price uptrend.

19-4
Using Multiple Averages

Since a single moving average doesn't always give a clear indication of the trend,
technical analysts often use a combination of two or three averages that move at
different speeds (calculated based on different time periods). Analysts commonly
work simultaneously with a 9-, 18- and 40-day average to track short-,
intermediate- and long-term trends. If all three averages are moving higher, it’s a
good bet that the market is in an uptrend. If there is divergence among the
moving averages, for example two are moving higher but one is headed lower,
then some judgment is required, and a trend follower might trade less
aggressively or stay out of the market until all the averages come into
agreement. An example of a chart with three moving averages plotted on it is the
June 2007 Swiss Franc future.

19-5
So many traders follow some of the moving averages, such as the ones
mentioned above, that the averages themselves can become the source of
significant support and resistance. For example, the 40-day moving average is
so popular as a trend indicator, and often a component of trend-following
systems used by futures fund managers, that prices often rebound or fall from
that average due to the sheer number of traders entering positions against that
level. Conversely, a violation of the 40-day moving average can trigger significant
liquidation of trend-following positions, even if this is not warranted by the
fundamentals, and this can create a false trend reversal signal.
Always keep in mind that a moving average is a lagging indicator which gives a
signal after the underlying market has embarked on a new trend.

Deciding which two averages to use is up to the individual trader and also poses
a dilemma. Using two shorter term averages to generate cross-over signals (say
a 4- and 8-day moving average) has the advantage of signaling quick changes in
market direction and enabling a trader to take a position early in a trend.
However, this approach does involve a fair number of incorrect signals and whip-
saw losses. Using longer term averages (say a 20- and a 40 day) generates
fewer false signals and fewer whip-saws, but suffers the disadvantage of
signaling positions to be taken well after a trend has begun, and this choice
poses a larger risk per trade than the shorter-term approach. Traders,
accordingly, must be very clear on whether they’re looking to capture short-,
intermediate- or long-term price moves and how much risk they’re willing to take,
before deciding which moving averages to track.

Weighted Moving Average

Since any combination of moving averages will result in erroneous signals to

varying degrees, technical analysts have tried to diminish the error potential by
weighting the averages. W ith this approach, they attach more importance to the
most recent price data, believing that "what's happening now" is more significant
than what happened previously. Other techniques for limiting error include
exponential "smoothing," offsetting the averages (shifting the moving average
line to the right on the chart), and calculating an average of averages. All of
these methods attempt to shift the position of the moving average relative to the
actual price movement so as to get a better trading signal, one that indicates a
trend change sooner but with fewer false cross-overs.

19-6
The dashed line on the Dec 2007 Japanese Yen future is a weighted 10-day
moving average. Notice how it reacts slightly faster to price changes than the
simple 10 day average.

Already discussed is the fact that analysts/traders often "optimize" moving

averages, a fairly simple process with computers and one that can help limit the
number of false trading signals over a selected period of price action. But
because markets are volatile (exhibit large or small price swings), an optimized
average may not work well beyond the "test" period. In practice, simple moving
averages can work just as well as jazzed-up versions.
19-7
Moving Averages & Trendlines

Analysts also use moving averages with other trend indicators such as trend lines
or oscillators to improve the reliability of the trading signals they get from the
averages. If prices violate a trend line and the moving averages also give a cross-
over signal in the same direction, it’s considered a double indication that the trend
is changing.

Note the chart example of the Sept ‘07 New Zealand Dollar future. It contains an
important 3-point upsloping trendline. This line was broken at the same time as
the downward crossover of the price and moving averages.

19-8
Trend Following Funds

Managed money has become a major factor in commodity markets since the late
1970's. Large sums that are now invested with fund managers. These managers
typically use technically-based trend following systems (unless they are a ‘long
only’ commodity index fund). The impact of the collective actions of the trend
following funds can be particularly pronounced when a market is changing trend.
Even though fund managers know this and try to design trading systems using
different trend indicators (including some pretty esoteric stuff), they still often
enter and exit markets at about the same time. Thus, whenever a trend changes,
everyone gets a signal regardless of the technical method they’re using and the
rush of buying or selling that occurs often creates a big move in the market.
Although most managed funds use sophisticated technical systems, traders can
get a feel for where concentrated fund activity might take place by watching the
40-day moving average. It provides a pretty good proxy for fund-trend following
buy and sell activity.

Moving Average Divergence

Another way to use moving averages is to keep track of the difference between
two averages. This difference, or "divergence", is an indicator of a market’s
momentum. In other words, if a market is in an uptrend and a short-term moving
average is rising at a faster pace than a longer term average, the price
momentum is increasing to the upside, i.e., the difference between the short-term
and long-term average is widening, implying that the trend is healthy and may
have further to go. More importantly, since momentum usually shifts before the
trend does a change in momentum can help anticipate a change in trend.

Moving average divergence is sometimes tracked on a type of chart called a

histogram (a type of bar chart) but is more often plotted against a zero line. For
example, if the value of the fast moving average is above the slow moving
average, the difference between them (subtract the value of the slow moving
average from the fast moving average) is a positive number and is plotted above
the zero line. If the fast moving average is below the slow one, the divergence is
a negative number and is plotted below the zero line. These types of charts are
easy to construct by hand, but the math-challenged need not worry since moving
average divergence indicators are included with most charting software
packages.

19-9
The Sept 2007 British Pound futures chart contains a simple 10 day and 21 day
moving average.
At the bottom of the chart the difference between the two averages is plotted as a
histogram.

19-10
 Chapter Twenty

Mathematical Models - Oscillators

Among the myriad of mathematical models, a category exists that tries to locate
price levels where the existing price trend could be in the process of exhausting
itself. The trader is looking for a reversal of the price trend. In general, this type
of model can be referred to as an oscillator. Some of the specific names for
these models are Relative Strength Index, Stochastic, Momentum, Overbought /
Oversold. Note that this type of model is diametrically opposed to the trend
following models.

The Relative Strength (RSI) and Stochastic models are two of the most popular.
They strive to locate ‘overbought’ and oversold’ levels on a price chart. The
theory is that if a trader can identify a market that is in an overbought condition,
the trader will want to take an opposite (i.e., short) position.

The calculation of these indicators is simple arithmetic. Personal computers and

plotting packages have made their use widespread. The indicator can be
calculated over any number of discrete periods. An active futures trader rarely
uses more than 13 periods - e.g. 13 hours or 13 days to calculate the resulting
number.

This output from the RSI or Stochastic model will be a number that can vary
between 0 and 100 but it will never reach either extreme.
The ‘trick’ is to set the parameters that constitute overbought and oversold.
Pioneering work in the field of mathematical models was the book New Concepts
in Technical Trading Systems, by J. W elles W ilder. In it, W ilder suggested a 14
day period and parameters of 70 for overbought and 30 oversold in the equity
market examples. There is nothing magic in these threshold levels. A technician
has to back test the model to determine suitable parameters. And even these are
unlikely to be optimal - as the markets change their characteristics.

Using threshold levels of 80 and 30 on the 9-day Relative Strength Index for the
Dec 2007 British Pound future produced a successful short sale (when the Index
when above 80) and a buy (when the Index went below 30).
Note that the RSI remained in neutral territory while the price of the British Pound
moved net sideways within a triangular consolidation.

20-1
20-2
A Caveat

Simply shorting a market when an indicator breeches a supposedly overbought

level is a path to ruin. A general observation can be made: 17 consecutive
profitable trades could be wiped out in one huge loss - where the value of the
indicator continues to move farther into extreme overbought territory as a price
move of unprecedented proportion takes place. And, it is difficult to determine
an effective stop loss point when a market is in a runaway mode.

In an attempt to create a safer entry into a position, technicians have worked with
not pulling the trigger on a new position unless a ‘failure swing’ occurs. This
means that a trader does not automatically place a new position when the
indicator enters ‘over’ territory. Rather, the trader watches for a pullback in the
indicator back into neutral territory. The next push by the indicator into ‘over’
territory is watch very closely. If the indicator fails to go to a new extreme and
then moves back into neutral territory again - this is the time to enter.

The entry signal just detailed was present on the hourly chart of the Dec 2007
Euro-fx future after the price run up. The 9-bar RSI moved above the overbought
threshold of 84, dropped back below 84, moved up above 84, and then moved
down below 84. This last move down was the short sale entry signal.

W hen this signal occurs it is not unusual to have a divergence between what
price is doing and what the indicator is doing. Note that the Dec Euro set a new
price high - but it was not capable of pulling the 9-hour RSI above the level
marked ‘A’ on the chart.

20-3
W hether this does provide a safer entry signal is open to conjecture. Traders are
always trying to refine their ability to construct risk and reward parameters. And
unlike classical bar charting where a measuring objective is present from an
orthodox price pattern, an RSI or Stochastic model does not provide a price
target.

20-4
Relative Strength Example
Hourly Chart

20-5
How it Came Out

20-6
A ‘Get Ready’ Signal on a 9-Week RSI

20-7
 ‘How it Came Out’

The failure swing and divergence led to a sizable price rally.

20-8
 Chapter Twenty-Two

Spreads & Spread Trading

The bailiwick of old time commodity futures traders were the grain and oilseed
spreads. The different seasonal factors plus the fact that distinct crop years
existed, lent itself to spreading - where the trader was long something as well as
short something. This form of trading also has a place in the modern markets.

Before listing some spreading considerations in the form of an outline, the

overwhelming premise for a chartist must be stated. That is:

All the hopes, fears, and moods are reflected in once single item - the price.

In the case of a spread, traders must be paying attention to the specific

differential between two prices. The price of one arbitrary item cannot be
subtracted from the price of another arbitrary item and expect that the plot of the
spread differential would/should act in classical fashion.

A trader has to do a minimal amount of fundamental analysis - and come to the

conclusion that enough traders are paying attention to the specific relationship
between the two items in question. If so, the spread should be charted.

The reader is encouraged to examine the chapters that follow, on classical bar
charting and then return to this chapter and then look at the various spread charts
- most of which contain a price pattern.

I. Overview

1. W hy trade spreads?

A. Lower risk? Not necessarily.

B. Attractive margins? - in futures, yes

C. ‘Staying power’? - Sometimes

22-1
 D. Aid in trading outright positions? - Definitely

i. Spread indicating tightness of supply (or the opposite)

ii . Picking the best month for an outright position

iii. Bull spread not working, may imply only a temporary

‘technical’ bull move is occurring.

2. Proper mentality toward profiting from a spread =

think the difference only

3. Definitions

A. W hat is a straddle? A spread

B W hat is a switch? A spread

C. W hat is pairs trading? A spread using individual equities

D. W hat is a cross? A division, one currency into another, neither

being the US Dollar

E. W hat is arbitrage? Exactly the same product, different location

4. Do not spread to protect a loss - ever!

II. Storable vs. Perishable Futures Contracts

1. Definition of a ‘storable’ futures contract

To be considered storable, the futures must be one which can be

accepted on delivery when the near-by (long) contract matures - and
be eligible without re-inspection for delivery, when the more distant
contract which was sold, matures.

22-2
2. Spreading mentality of old-time CBOT vs. CME members

A. W hich Chicago futures exchange traded predominantly storable

commodities and which traded perishable commodities (according to
the definition stated above)? The Chicago Board of Trade = storable.

B. W hy did Crude Oil futures flourish on the New York Mercantile

Exchange (and not the CBOT)?

i. W hat did the NY Merc trade in the ‘old days’? Potatoes

ii. Is Crude Oil (futures) a storable commodity? No

III. Charting a Spread

1. Can / should a spread be charted?

A. Some minimal fundamental analysis must be done to

determine the answer.

B. An example of a spread that should not be charted: the

Canadian Bacon spread (Canadian Dollar - Pork Bellies).

2. Use close-only prices

A. W hich futures close to use? ‘Pit’ close.

B. Is there a close in the cash foreign exchange markets? No

3. Does the plotting software support spread definitions?

A. W eighting the legs differently - e.g., 3 Cattle vs. 4 Hogs

B. Three items

i. Soybeans vs. Meal & Oil

ii. 3 Crude Oil vs. 2 Gasoline & 1 Heating Oil

22-3
4. Plot:

A. Near-by minus more distant month (inter-delivery spread)

B. Difference in Dollar value: Second named item is subtracted

from the first named item (inter-commodity spread)

5. Spreads tend to ‘trend’ vs. create classical price patterns

6. A ‘flat’ on spread chart = a minor trend change indicator.

See any of the various ‘current’ charts

IV. Definitions

1. ‘Normal’ market: Near-by trading below back month

2 . ‘Inverted’ market: Near-by trading above back month

3. British terms:

A. Contango = normal market

B. Backwardation = inverted market

V. Order Entry

1. Place order as a spread order:

A. Indicate a specific (limited) spread differential

B. Do not ‘leg’ into a spread.

C. Do not ‘leg’ out of a spread

22-4
2. State long leg first - if calling the CBOT open outcry floor
(‘Buy’ is on the left side of the order pad)

3. CBOT floor traders traditionally used positive numbers (‘or better’ is

always implied)
See Order Entry Example

4. Electronic spread entry places premium or discount on near-by contract.

5. Interest rate traders (at the introduction of T-Bond futures) used

negative numbers (if applicable)

6. Last trade prices vs. where the spread is really trading, e.g.,

A. Dec - July Corn spread is quoted as 16 ½ to 17 (premium July)

B. Market order to buy Dec - sell July arrives;

execution is at 16 ½.

C. Market order to buy July - sell Dec arrives;

execution is at 17.

D. Better strategy for either side;

enter limited order at 16 3/4 (split the difference)

7. Use of protective stop (loss) orders

A. ‘Mental’ stop - dangerous but necessary

B. Determined location of stop based on:

i. Money management

ii. Spread chart

C. Stop placed on the volatile leg vs. ‘dead leg’ of the spread

D. W hen to exit - when the trendline is broken or (partially) when

objective met
22-5
8. Limit price moves (future is locked limit up or down)

A. Can a spread to exit from locked limit position (if any of the
other futures contracts are freely traded).

B. Can use options to determine where futures are trading.

C. Do not buy last market to trade limit-up

D. Do not sell last market to trade limit-down

VI. Limited Risk Spreads

1. Available only in storable futures contracts where calculation of carrying

charges is possible.

2. Limited risk spread = long near-by vs. short deferred

A. Risk is limited to move to ‘full carry’
B. Reward is unlimited

3. Forward pricing

A. The ‘basis’ = cash price minus futures price

(Basis in grains was unusually unstable (wide) when strong inflow of
‘Index Fund’ money occurred in 2007/08)

B. How to compute ‘full carry’

i. Interest cost
Could be bankers acceptance rate, fed funds + 1 3/4 - 2%
or prime, or prime +1%

ii. Storage cost (0.15 cent/bu/day or 4 ½ cents/bu/month

for a 30 day month - as of April 2008)

iii. Insurance cost - negligible, “most often ignored”

iv. Commissions - very low for an exchange member

v. Inspection / handling - if applicable

22-6
 C. Risks in a limited risk spread

i. Tendency for discounts to widen (toward full carry) with

the passage of time - as maturity approaches

ii. Price rise increases total value of contract & thus

financing costs increases full carry

iii. Interest rate increase (full carry increases)

iv. Obtaining commodity that is not re-deliverable

(KC W heat, Lumber)

v. Buying cash commodity & unable to get it into position

‘regular’ for delivery

vi. Greater than limit move (down) by expiring future

vii. Government intervention - ceiling on near-by future

Unlikely - but it has happened

3. Question: How could (did) futures traders speculate on interest rate

changes prior to the introduction of short term interest rate futures?
(Answer: Storable commodity spreads trading near full carry)

4. Fallacy of using ‘newspaper’ cash prices - don’t do it.

VII. Some Specific Spreads

1 . Gold / Silver ratio

A. Long term ratio (approx. 100 years = 32.5

B. Ratio on March 12, 2008

i. 48.82 (cash) and moving downward on monthly chart

ii. 49.19 (Dec 2008 futures), on a ‘bounce’

22-7
 C. Dollar value of contracts should be approximately equal

i. CBOT one Kilo Gold vs. CBOT 1000 oz Silver future =

close enough to equal Dollar values.

ii. Could weight the legs and/or plot a Dollar difference chart

2. S&P 500 Index vs. Dow Jones Industrial Average

A. 95% correlation during majority of trading days

B. Both dominated by large capitalization, blue chip stocks

C. E-mini S&P 500 - E-mini Dow

i. One-to-One

ii. ‘Ratio’ the spread = “dollar neutral”

5 x 6 or 10 x 11 (using 2007 year end prices)

iii. Spread margin = approx. 10% of normal outright initial margin

3. Crack spread:

i. 2 Gasoline & 1 Heating Oil vs. 3 Crude Oil

ii. Spread = [{Gasoline price x 42} x 2 + Heating Oil price x 42

minus (Crude Oil price x 3)] divided by 3

iii. Spread result is in Dollars/barrel

22-8
 Soybean Spreads

New Crop - New Crop

Old Crop - New Crop

22-9
 A Cross Chart

The old Sterling / D-Mark Cross

W henever a horizontal line is present on a chart - that is the line that the
market ‘wants’ to go thru.

22-10
Falling Wedge on the Gold / Silver Ratio

Monthly Chart

22-11
Notes

22-12