1- 3

1-5
 BEHAVIOR OF PRICES

" ON WALL STREET

~I'C ARTHUR A.MERRILL
MARKET INCLINATIONS
lHELP PREDICTION
LpRODUCE PROFITS

Published by
The Analysis Press. Chappaqua. New York

1-7
All rights reserved including the right of reproduction
in whole or in part in any form.
Copyright © 1965 by Arthur A. Merrill.
Printed in the United States of America.
Published by Analysis Press, Chappaqua, New York.

Library of Congress Catalog Card No. 64-25611.

 DEDICATION:

To ELSIE

Always helpful; always inspiring.

1-9
THE BEHAVIOR OF PRICES ON WALL STREET -

CONTENTS-

PART I - INTRODUCTORY -
Chapter 1. How can you use this book for profit?
Z. Can you forecast the market weather?

PART II - BEHAVIOR AT CERTAIN TIMES -

Chapter 3. When are the best times in the year to buy?
To sell? (Including the Market Almanac
for every day in the year.)
4. When are the best times in the month to buy?
To sell?
5. When are the best times in the week to buy?
To sell?
6. When are the best times in the day to buy?
To sell?
7. Are there profitable seasonal differentials?
8. How does the market behave near holidays?
9. How does the market behave near income
tax time?
10. How does the market behave after bad news?
11. How does the market behave after margin
changes?

PART III - BEHAVIOR IN CYCLICAL SWINGS:

Chapter 12.. Does the market have trends?
13. Are there patterns of behavior in the weekly
swings?
14. What is the Dow Theory? Is it profitable?
15. What is the Elliott Wave Theory?
16. How long should you expect a Bull Market to
last? A Bear Market? A Primary Swing?
A Secondary Reaction? (Including Life
Expectancy tables. )
17. Is the market influenced by earnings?
18. Is trend following profitable?
19. Are some prices preferable to others for
purchase? For Sale?

PART IV - CONCLUSIONS -
Chapter ZOo Conclusions.
 PART V - APPENDIXES FOR STUDENTS:

ANALYSES FOR STUDENTS:

Appendix AI: Are the D-J Industrials representative?
AZ: Is the market random?
A3: How should turning points be measured?
A4: In Elliott's Theory. what are the variations?
A5: Can the Elliott Theory be made specific?
A6: How should swings be classified?
A7: How do swings vary with price?

DATA FOR STUDENTS:

Appendix A8: The D-J Averages - weekly bar chart
1928-1964
A9: Dow Theory dates and data.
AIO: Data and charts - Bull and Bear Markets,
primary swings and secondary reactions.
1897-1963
All: Bibliography

TOOLS FOR STUDENTS:

Appendix A12: How to make the Chi Squared test
A13:Elapsed Time Calculator
A14: P!E Calculator
A15: 300 Year Calendar

1-11
 IN APPRECIATION -

The suggestions made by the following friends have

been especially helpful: A. Hamilton Bolton, Robert W.
Breiling, Dr. Edward R. Dewey, Jack A. Dorland, Dernell
Every, Devin A. Garrity, Floyd L. Hogan, Christopher
R. Landmann, S. Jay Levy, William D. Merrill, Francis
N. Millett, Richard Russell, George Seager, William P.
Short, George E. Tener.
 "I often say that when you can measure what you are
speaking about, and express it in nurnbe r s, you know
something about it; but when you cannot express it in
numbers, your knowledge is of a meagre and unsatis-
factory kind; it may be the beginning of knowledge, but
you have scarcely in your thoughts advanced to the stage
of Science, whatever the matter may be. "

Sir William Thomson, Lord Kelvin

1-13
 (p.l-l)

Chapter i.
HOW CAN YOU USE THIS BOOK FOR PROFIT?

Should stocks be bought today? Should purchase be

delayed for a lower price?

Should stocks be sold now? Should sale be postponed

for a higher price?

This book will concentrate on these four questions.

There are two problems in investment; selection and

timing. The latter problem, which is dealt with in this
book, is considered by many to be the more important.
Poor stocks will rise in a rising market; good stocks
will fall in a declining market. There is profit in timing.

This book is addressed to both the long term investor

and to the short term speculator. All of the conclusions
are in the twenty chapters. which are written for the
non-mathematical investor. For serious students. the
fifteen appendixes will supply analyses. data for re-
search and some tools for further work.

A body of folklore has been built up ar-ound the mar-

ket through the years. Too often this folklore has not
been put to the test. This book will try to separate fact
from folklore, by the simple but laborious method of
consulting the record.

It isn It surprising, except to the scientific man in

another field, that the material hasn't been thor.oughly
sifted, for the size of the undertaking is frightening.
But modern methods and the computer are now available.

1-15
 (p.1-2)

Chapter s 3. through 11. will deal with the behavior

biases of the market at certain specific times. You will
find many unsuspected tendencies in the market, and
corresponding suggestions for profit. The market will
be found to behave with the characteristics of a warped
roulette wheel. Sometimes it has a definite bullish or
upward bias - at these times you should move with ce-
lerity in purchase, and should delay sale. At other
times it has a definite bias downward; at these times
sell promptly and delay purchase for profit.

Chapters 12. through 19. deal with the behavior of the

market in swings and trends. There are some sugges-
tions for profit from this knowledge.

This book doesn't deal with individual stocks, but

with the averages. However, most stocks tend to move
with the spirit of the market. The Dow-Jones Industrial
Average is used as an index. (An analysis of this selec-
tion is in Appendix AI)

This is a picture book. To paraphrase the old Chi-

nese saying: One chart tells more than a thousand sta-
tistics! To make it easier for you, we have inserted a
few blank pages, to bring the charts next to the corres-
ponding text.

May your profit increase.

 (p. Z-l)

Chapter 2:
CAN YOU FORECAST THE MARKET WEATHER?

The conclusion of this book is that there are certain

inclinations or leanings in the behavior of the market.
A knowledge of these inclinations is very useful in the
improvement of timing of purchase or sale.

We will be dealing with probabilities, not with cer-

tainties. But these probabilities, if known, can be used
for profit. A bent roulette wheel may be biased toward
the even numbers; despite this bias it will produce many
odd numbers. But the bias is profitable to the acute
customer. The casinos at Monte Carlo employ statis-
ticians to check the accuracy of the wheels from the
record. They would like to replace a wheel before a
customer discovers the bias!

The knowledge of bias in the stock market is rarely

sufficient, in itself, to justify a short term speculation.
However, this knowledge can help the speculator and the
investor by improving timing and profit.

It is important to consider significance. All of the

biases noted in this book could have occurred by pure
chance. If you throw a handful of a hundred coins on
a table, it is pos si ble to find them all "heads ". How-
ever, if this score occurs, you would have some reason
to suspect some double-headed coins! To measure the
significance of a score, statisticians compare the actual
score with the expected score, and ask themselves "How
many times would I have to repeat this experiment to
expect such a difference by pure chance?" There are
mathematical means of obtaining the answer; one of the
simpler methods is outlined in Appendix AIZ. In our
text, which is addressed to non-mathematicians, we will
use the following words:

"Probably Significant": The bias would be expected

by chance only once in twenty repetitions of the experi-
ment.
"Significant"; The bias would be expected by chance
only once in a hundred times.
"Highly Significant": The bias would be expected by
chance only once in a thousand repetitions of the entire
experiment.

1-16
 (p.3-1)

IS EASONAL --1896 -1963 I

80 (Pere ent ot years in which 0 -J Industrials
Increased in the month.)

70

60

48 4S"l.
Feb. Sep.
Mar. Apr. May June July Aug. Oct. Noy. Dec~
51% 5Z% 55% 50% 68% 10% 55"1. 61"1. 75"1.1

Fig.3A

-- SEASONAL--

%
\6 PRESIDENTIAL ELECTION YEARS I
~ ~
80

70
I--

60
........... .... .~~~~~.~~:~.8.~
-
t. ..... ..... ..... .... ..... .....
l-

50
1e"l.
Jon
44"1.
Feb.
11"1.
Apr. May5°~1 31"1.
Sep.
Mar. June July Aug. Oct. Nay. Dec
81% Ci6% 63'70 81% 75% 6S% 63%
-
40
'--

- ..........
Fig.3B
 (p.3-2)

Chapter 3.
WHEN ARE THE BEST TIMES IN THE YEAR TO BUY?
TO SELL?

Figure 3A summarizes the seasonal tendency of the

market for the 67 years from 1896 through 1963. The
most favorable month is December. In this month the
market increased in 75% of the 67 years.

The summer rally months, August and July win the

second and third prizes. August rose 70% of the time;
July increased in 68% of the years.

For maximum profit, of course, one should try to

buy before the strong months and sell before the weaker
months.

Figure 3B presents the record for the 16 presidential

ebecti on years from 1900 through 1960. There are some
interesting differences between these years and the total
in the preceding c ha r t; January, April and September
have been significantly weak in election years; March,
August and October are notably strong.

1-14
 (p.3-3)

Figure 3C applies the microscope to the summer

rally. The behavior by days is charted.

Two periods are included, to verify the conclusions,

and to note changes in characteristics. The principal
difference between the two curves is in the middle of
July. This period did not appear weak in the 1922-1942
period; it is notably weak in the 1943-1963 period.

The agreement between the two curves is more sig-

nificant than the differences. The strength in the early
part of July and in the middle of August is present in
both curves.

'I(,

100

THE
90 TWO 21-YEAR

(5 Ooy centered average)

80
1943-1963

10

60

40

30 JULY AUGUST

Fig.3C
 (p.3-4)

The next two charts (Figs. 3D, 3E) extend the micro-
scopic view to the entire year. Here is a Market Alma-
nac for every day in the year, based on the performance
on each of 19.253 trading days in the last 67 years.

On any day in the year. you can consult the chart and
find the average market weather for that time of the
year. For example, note January 31. The bar extends
to 580/0. which means that the market at that date rose
580/0 of the time and declined 42% of the time.

There are many interesting and significant points on

these charts. We will highlight some of them in the
chapters which follow. Note the summer rally in July
and August; note the year-end rally in December.

Footnotes for students:

(1) The data have been smoothed by a five day cen-
tered average.
(2) The bars are measured from the 500/0 level. but
should also be compared with the average for the entire
period (530/0), which is indicated by a horizontal line.
(3) The significance levels are marked on the right
hand scale (See Appendix AI2).

1-12
 (p. 3- 5)

70

65
·0.1%

1%
6
5%
JANUARY
FEBRUARY

-0.1%

70

65
0.1%

1%
60 5%

...
55

50~~~W@lllJ~1llJlD'"~"'"..... !~
I 20
25
30 ;
~
<II
~
>

5%

APRIL 1%
MARCH -0.1"'"

70

65
O.l"l0
1%
6
5%

...
o
55 z
~~

5°f.. . -'t1F[rrIJ1r~~.lfWw-eVJw~11~UJ~TT".J1lJ1wm .....

... - >
z ...
5 iii
5%
45 JUNE -''I.
MAY
-0.1%

Fig. 3D
 (p.3-6)

(Percent 01 yeors 1897-1963 in which the D-J Industriols posted on

increose lor the doy. - - 5 doy centered averoge.)

COPVRIGHT 1'84- AATHUft A MERRILL

-O.lor.
- lor.
GO - ,or.

50 ' 'eu " eo es 30 I

JULY
-5".
AUGUST
45 1 - I'll.
-o.rs,

%
70-

65-
-0.1%

- lor.
60-
- ... ,

SEPTEM BER OCTOBER

%
70

65
-0.''''

60 -I'"
- s...

55 ..
<>
2

....
~~
->

"j
2 ..

~~
30
1-sor.
45 NOVEMBER DECEMBER -1'110

1-10 Fig.3E
 (p.4-1)

whr ~lIrkrt iUmlinar --

wh, Silt r k , t SI (I nth

(Percent of mcnr hs Jon. 1697 - June 1964 in w~ic~ t~e O-J Industrials posted

on rn e re o s s for the day.)

0/0
..
-'
70 - >.
.
65- _ LAST HALF OF MONTH - - FI RST HALF OF MONTH -
..
u
z

~
62.6
.
z
;;;
60-
50.3
•
-0.'%
51.6 -.%
56.'3 -5%
562. 55.9
1j5.6
54

49.9 -5%
49.1
-1%
41.0 -0.1%

APR,JUNE,
FEB.- { 16171819202122232425262728
1 16 17181920212223242526272829 I 2 3 4 5 6 7 8 9 10 II '2 13 14 15
SEPT.,NOV.r-'6 17 18 19 20 21 2223 24 2526272829 3
OTHERS.-IB 17 18 19202122232425262728293031

-- - DAY OF THE MONTH - - -

Fig.4A
 (p.4-2)

Chapter 4.
WHEN ARE THE BEST TIMES IN THE MONTH TO BUY?
TO SELL?

Does the market have a bias based on the day of the

month?

The record has been examined - every day from 1897

through June. 1964. The result is presented in Fig. 4A.

The chart has been split in the middle. Because of

the important beginning-of-the-month behavior. the first
of the month has been placed in the center of the chart.

The conclusion is clear. There is a definite bias in

the market near the beginning of the month. Note the
strong market on the first two days. Here is a bullish
bias which is highly significant.

If behavior is typical, then, one should try to buy for

profit before the last three days of a month, and sell for
profit after the first three days of a month.

1-8
(p. 5-1)

Fig.5A
 (p.5-2)

Chapter 5.
WHEN ARE THE BEST TIMES IN THE WEEK TO BUY?
TO SELL?

The market prior to 1952 was open on Saturdays.

The span since that time, with a five day week, has been
reported in Figure 5A.

The result is interesting. Monday is truly a blue

day. The market rose only 43.3% of the time, which is
a highly significant amount below the average for all
days in the period.

Friday, on the other hand, is highly significant in a

bullish direction. This is contrary to the lore of the
market, since Friday is supposed to be a day when tra-
ders "unload to free their minds for the weekend. "

These tendencies through the week will be examined

hour- by-hour in the next chapter.

1-6
 (p.6-1)

- wq.r Ailarkr. ~lmanac

• ill h r Slarkr. lay anb • rrk
...-'
~
(Percent

which
01

the
trod ing

D-J
days

Industrials
Jon. 1962

posted
th,oug h

an ine ree se
Jun. 1964

from
in

Ihe
..
-'

~
;;
preceding hou r . ) ;;:
% z

7
.
"
t
0.1%
63.2
61.• 1%
81.1

59.1 5%
60
MO N DAY
55.2
&<.1
....... •••••• ~~~~.~~E:.-••~~.I.~ .1.. . ..... ......
50
Open II 12 2 Close Open II 12 2 Close
47.3
44.'
TUESDAY 5%

40
,...
38.2 0.1%
31.8
36.1
34.4"

30

70
61.2
O.I~

,%
60
60.5
61.3
60.5
'9.7 ....
54.8 54.5 54,8

.... .. ..... ...... ......

50
Open II 2 Close Open II 12 2 Close

W EON ES DAY - ....

40 4L5
43.1
,%
40.3
THURSDAY 0.1"1.

30

70

-I
Ol}
60 '9.0
57.9 0.1
I
.
~

;;
5

.
-'
-'

50
Open Close 5
...
-'

46.8
I
0.1
e

40
FRIDAY TOTAL ALL DAYS
40.9
5
I 1;
(625 DAYS) 0.1 J::

Fig.6A
30
 (p.6-2)

Chapter 6.
WHEN ARE THE BEST TIMES IN THE DAY TO BUY?
TO SELL?

In this chapter we will note that the market has a

bullish bias in certain of the hours in the week; it has a
bearish bias at other times.

Figure 6A summarizes the action of the market on

625 trading days (January 1962 through June 1964).

There are many interesting and significant hours in

the week:

(I) Note the strong opening on Monday, and the rela-

tively strong openings on Wednesday and Friday. The
average for the opening (The chart in the Southeast cor-
ner) is significantly bullish.

(2) Note the strong first hour to llAM on Thursday

and also on Tuesday and Wednesday. The average for
this hour is significantly bullish.

(3) Note the weak close, which is evident on every

day except Friday. The average close for all days of
the week is highly significant in the bearish direction.

Are these differences consistent through the entire

period? If we divide the record into three parts (1962,
1963, 6 months 1964) and record the bullish hours with
a "plus" and bearish hours with a "minus", here is the
record:
open 11 12 1 2 close
1962 -+ +
1963 + + +
1964-6m. + + +
Total + + l'

Note the unanimity in the bullish hour to llAM, and

the bearish hours to IPM and to the close. The overall
record is quite consistent.

1-4
 (p.7-1)

I SEASONAL 01 FFER ENTI A LS I

Percent of times D-J Industrials increased

in speci f ied period - -

0/0
80-

70-

60-

50
47 %
4.0-

30-
ROSH HASHANA JULY FOURTH THANKSGIVING
to to to
YOM KI PPUR LABOR DAY NEW YEARS

(1934 -1963) (1897-1963) (1897-1963)

Fig.7A
 (p.7-2)

Chapter 7.
ARE THERE PROFITABLE SEASONAL DIFFERENTIALS?

Some market sayings are founded on the seasonal

tendency. Chart 7A reports on three of these Seasonal
Differentials.

The first is "Sell at Rosh Hashana; buy back at Yom

Kippur." This is based, perhaps, on the lack of buying
by orthodox Jews in this period. It is, also, in the
midst of the September bearishness.

This "saying" has been checked back through thirty

years. The re sult is at the left of Figure 7 A. The mar-
ket produced a decline in 53% of the cases, and gained in
47"0. This checks the "saying"; however, the difference
from expected is not significant.

Another"saying": "Buy on July Fourth and sell on

Labor Day and Pay Your Expenses for the Year." This
recommendation has been checked for every year back
to 1897. The score is good. The market rose in this
period, which is the summer rally season, 68% of the
time. However, the difference from the average for all
months in the period (58%) does not rate significance.

We have tried our hand at creating a "saying": "Buy

at Thanksgiving and Sell at New Years and Pay Your
Christmas Bills." This is an obvious way to collect on
the usual December rise. The score is excellent - 750/0
success for the period 1897 through 1963. This is suf-
ficiently different from average to justify the rating:
"probably significant".

1-2
 (p.8-1)

• HCLIDAY BEHAVIOR
- JAN. 1897 - June 1964

(Percent of occasions on which the D-J Industrials

posted an increase for the day.)

DAY BEFORE HOLIDAY: DAY AFTER HOLIDAY:

Hoi i days

71.2

Long

Holidays

67.7

All
Holidays

Fig. SA
 (p.8-2)

Chapter 8.
HOW DOES THE MARKET BEHAVE NEAR HOLIDAYS?

Chart 8A is an analysis of the performance on the

day before and the day after all holidays in the last 67
years. It presents some very significant results.

The usual market commentary preceding a holiday

predicts a sell-off. "since traders like to be free of
worries on the holiday." This is not checked by the
facts.

The actual count reports a tendency in the opposite

direction. The day preceding a holiday is usually a good
day. The market has risen two-thirds of the time. In
the case of a long holiday. the market has risen 710/0 of
the time on the preceding day. These ratios are highly
significant; they would be exceeded by chance only once
in several thousand repetitions of the 67-year history.

The score for the day following is not significant. in

the case of a short holiday. After a long holiday. the
day is significantly inclined toward bearishness.

Z-l
(p.8-3)
-HOLIDAY BEHAVIOR-
1897· JUNE 1964

(PERCENT OF YEARS IN WH ICH THE D-J INDUSTRIALS

POST E 0 AN INCREASE FOR THE DAY --)

DAY DAY DAY DAY

BEFORE HOll DAY AFTER BEFORE HOLIDAY AFTER
HOLIDAY HOLIDAY HOLIDAY HOLIDAY

WashinGton'.
a:~.11
Birthday

0::. "'0 ao Election I I no

••
····0
••
Good
c=J~~.'
Day

Friday

79.1

aoU
'0
ThanksGivinG D'" '0

13.1

Memorial
~o'--'----------....,..-.,.o
Day 48.3

BO.6~
aoL--L-----------.:== 49.3
'0.

r - - 54." 74.

Independence
ao ••
Day

80.1 55.0
New Yean Day
.0 '0.

••
La bor
Day
!4.!

aD
"'0
00
Total
Holidays
All
Fig.8B

'7'"'~~D
 (p.8-4)

Figure 8B presents the score for each of the current

holidays. The day before the holiday is especially
bullish on the days before New Years, Memorial Day,
Independence Day, Labor Day, and Christmas.

The day after each of the holidays has a rather poor

record. The day after Washington's Birthday is es-
pecially bearish. There are two exceptions: The day
after Independence day, which is at the beginning of the
summer rally season, is bullish. The trading day
following Thanksgiving, which is usually a Friday, is
also bullish. This may be a reflection of the tendency to
rise on Friday. (See Chapter 5)

If you are considering purchase near a market holi-

day, therefore, make your purchase at least two days
prior to the holiday. If you are considering a sale,
postpone it until 2 PM on the day preceding the holiday -
and your chances for profit will be improved.

(Footnote for students: The data in Figure 8B differ

from the Market Almanac (Figures 3D, 3E) because the
latter has been smoothed by a fi ve- day average. )

2-3
 (p.8-5)

WEEK PRECEDING ELECTION

Percent of Days in which D-J Industrials INCREASED--

80

70

60

50
I 56.2% I 75.0% 68.8'0 58.4% SUN- 81.2% ELEC-
I 43.8% I I 43.8% DAY TION
DAY
40

30
M Tu W Th F 5 5 M Tu
20

Fig. Be
 (p.8-6)

Figure 8e reports the record for the week before

election for the sixteen presidential election years from
1900 through 1960.

The last part of these election weeks seems bullish.

Thursday has a fine record. This day rose twelve times
and declined four.

The Monday preceding presidential elections is the

most significant day. The market rose on thirteen of
the Mondays and declined on only three. This ratio
would be expected to occur by chance only once in 500
repetitions of the experiment. so that this day appears
to have a highly significant bias in the bullish direction.

{Footnol e: Since the preparation of this chart. we have

passed the 1964 election week. It was quite typical:
Monday: Down
Tuesday: Down
Wednesday: Down
Thursday: Up
Friday: Up
Monday: Up
Election Day.

Z-5
 (p.9-1)

INCOME TAX TIME a THE MARKET

10. Trading Days Before Dote
(March 15,1933 - 1954; April 15,1955-1963)
Percent of Days in which D.-J. Ind. closed HIGHER:

°/0
70-

60- AVERAG E
~ALL DAYS
50 63.3 63.3 56:7 55.2 62.1 62.1 59.1 1933-63
53.3"1. 53.3 40.0 41.4

40

30
-10 -9 -B -7 -6 -5 -4 -3 -2 -I Due
Dote
--DAYS- -

Fig.9A
 (p.9-2)

Chapter 9.
HOW DOES THE MARKET BEHAVE NEAR INCOME TAX
TIME?

Selling near the income tax due date is supposed.

according to market lore, to gi ve the market a bearish
tinge.

Unfortunately, the evidence in Fig. 9A does not sup-

port (or deny) this hypothesis. The evidence votes sim-
ply: "no demonstrated significant difference from aver-
age.

The statistics cover the span from 1933 to 1963. Be-

fore this thirty year period, the income tax was not a
major factor.

2-7
 (p. 10-1)
·MARKET ACTION AT TIMES (U' BAD NEWS
-OJI CLOSING PRICES; TWO TRADING DAYS PRIOR TO NEWS a FIVE OAYS FOLLOWING ~

100'4 ~ CLOSING PRIOR TO NEWS-·

0/.
105 .....---~----------~-----------p

Me KINLEY

100~---~

95~----+-----------I>-\--.J---------l

92 I I
I I
9/4/01 9/5 9/6 9/7 91'J 9/109/11 9112 9113 9116 9/17 9/1 B 9/20

105 -r-'- - - - , - - - - - - - - - -......... 105..------,----------.....,

HARDING ROOSEVELT
DEATH DEATH

100~--=-"::::...---------_I
......~----____l
IOO~-......,,-~--

BE FO R E

NEWS
1-----4<-- AFTER NEWS

95L-.l--.L..-.l--.l--.l--L-~
9 5 L_.l...._.L.._.L...._.L...._-L-_.l-~
4/10/45 04/11 4112 4113 04/16 4/17 4/18 1II/19
71311238/1 8/2 8/4 8/6 817 8/8 8/9
105..----.....,----------.....,
KENNEDY
DEATH
1001--==-4-----------1
EISENHOWER
HEART ATTACK

95~---1__-+--I------__l
BEFORE
BEFORE
NEWS
NEWS 1o-------+l.~AFTER NEWS -----.I
Fig.
I I

9/21155 9/22 9/23 9126 9/27 9/28 9129 9/30 11121 11/22 11/26 11127 11129 1212 lOA
 (p.lO-2)

Chapter 10:
HOW DOES THE MARKET BEHAVE AFTER BAD NEWS?

Figure lOA charts the market reaction to five cases

of tragic news.

The market has some very bad moments immediately

following the news. Pric e s are dri ven down by selling
to a surprising degree.

However, when a day has passed, the market re-

covers from its panic, and sometimes works upward
to a higher level

Z-9
 (p.11-l)

Chapter 11.
HOW DOES THE MARKET BEHAVE AFTER MARGIN
CHANGES?

Figures llA and lIB chart the market behavior after

the last six margin reductions and the last six margin
increases. In each case, the lower right hand part of
the chart summarizes the behavior.

In the case of margin reductions, the market tends

to work higher in the four following weeks. The behavior
in the fifth and sixth weeks is quite mixed.

The average behavior in the case of margin increases

is surprising - the market, on the average, tends to
move upward in the nearby weeks. However, an exam-
ination of the individual charts reveals this to be a con-
tinuation of the preceding trend. In nearby weeks, then,
the trend seems to be more important than the margin
increase.
N
I

-
• MARGIN R EDUCT ION S

Pili OE C!lANGES IN NIARBY WEEKS 100% = O,J.1. Close on Frldo~ precedlnQ

the reduclio,..
105, I IO~ ~------~~----'---~1105, I I
2/1/<47 I 3/~0/49 2120/e3
, 04 r04 - 1041-
'OO'l. TO 7e'l. 7~'l. TO ~O'l. 7e'l. TO eO'l.
'D' '0' 'D'
102'- '02 102'-

101'- '0' , 0'

1001 ( \ 10,,1 \ ! . . . . . . 7/ I
•• 99

••
97 "0

9$1 ! I J T'" ~I"'" 'T ."," "T'O' ..(..... -C' ' OJ'' -,'" ·T·... "'I n J6
•• $ 10 I/!7 1/1!.4 Ill! 2/7 ttl" 2~li!1 2111 '" ~14 51(' l21 I
9
12/27 1/3 lifO 1117 1/24 1/31 211 2/1<4 2121 2/28 3/7 N
'6... I i I I
--
ro s I i IO:i
1/16/58
1041- 1104
70'l. TO eo'l. eO'l.
1031- 1103,-

'02
"I \
'02

101 '01

'DO

•• ..
9. to

97"- 97 9T ,
r,om :H'C'~'''91
.0 •• to flVI c"'or'l
.., I
'IS.',. 11/1' '1110 'III' II] 1110 .", 1/14 VII ,I? 1'14 I~~$I~I" 1114 7/1 711 1/11 "" 7'11 " I Ifli 'III 111:1 "'I ':,L.',-.,.:...-.......-\tc'-tj--'-...L--.......--------l
rill 111101 " . 7 " » "'0 " , ,

Fig.llA
 lIll AlGIN XNCIU,ASI&S ,. PInCE CHANGES IN MEARJaY WIE"." 100"'OJ, Clo.. on F,idoy
% precedinQ the reduction
'05
7/5/45
10+ 50~ to 75%
103

102

'01

100

99

98

97

96

95 r8efore
:;;-
lec r ee s e After Il'\creose
94 ....
....
93 1 I , I I I ,

~12.6 612 6/9 6/113 S/Z36/30 7/6 7/13712.07/27 8/3 8/10 12I'~ 12.2 /29 J/~ 112 I/S /2.6 2/2 /9 116 /2.1 3/2 1219 116 /22 /30 1/6 113 12.0 /27 2/3 flO 117 124
....,I
% --
105. i
8/5158 10/16/58 11/5/G3
'04
50~ to 70~ ~O% to 70"4 i
103 r
MediOn\/
I, r
'02
.........'
101
.:"
.... -s., _w-wo".,
..
1001 %
39

98

97
x··
96

MAl

lie 1215 el! Ie fl~ /22 129 9/, /12 9~ /12 119 /26 10/3 flO 111 /24 131 11/7 114 If! 9/27 /4 III 118 lie 1111

Fig. llB
 [p, 12-1)

Chapter 12.
DOES THE MARKET HAVE TRENDS?

IT the market has been moving up, should we expect

a continuation - or a reversal?

IT the market has been declining for two days; should

we expect a decline on the third day?

Some academic analysts have concluded that you can

match a coin and get the answer. They admit that it
should be a slightly warped coin, because the market
spends more time rising than falling.

But we have evidence in this book (See Appendix A2)

that should interest the ac adernrc i ans , We have found
there are trends in the short term (hourly, daily) move-
ments; the evidence for the longer periods isn't conclu-
sive.

Here are four conclusions from the data to improve

your profits:

(1) IT prices have moved up in the last hour, or two

hours in a row, or three, the odds favor a rise in the
next hour. (The statistics are highly significant. )

(2) IT prices declined in the last hour, or two hours

in a row, or three, or four, or five, the odds favor a
decline in the next hour. (The statistics are highly sig-
nificant. )

(3) IT the market moved up on a certain day, the odds

favor a rise on the next day. (highly significant)

(4) IT the market declined on a certain day, or two

days in a row, the odds favor a decline on the next day.
(Highly significant. )

The weekly data have revealed no significant trend-

inclined bias. However, in Chapter 18, we will show
that it can be profitable to follow a long-term trend in
the market.

2-13
 (p.13-1)

TWO WIII 'ATTII.NS

1924 - JUNE 1964

BE HAVIOR OF THE D-J IN DUS TR IALS

IN HI E THIRD WEEK

TWO WEEK PATTERN MOVED MOVED TOTAL

PATTERN; NUMBER UP DOWN OCCASIONS

Il1IiI 420 287 707

rnw 2 278 221 499

~ 3 289 211 500

ImJ 4 220 177 397

Totals 1207 896 2103
57.4% 42.6% 100.0 %
"
Fig.13A

TBI.I .. W.... I .ATT ... lI.

1924-JUNE 1964

BEHAVIOR O~ THE D-J INDUSTRIALS

IN THE FOURTH WE EK

THREE WEEK PATTERN MOVED MOVED TOTAL

PATTERN: NUMBER UP DOWN OCCASIONS

~ 240 180 420

~ 2 163 124 287

Y 3 166 112 278

~ 4 115 106 221

8 ULLISH I2fIiiI 5 180 109 289

~ 6 114 97 21 I

J 7 123 97 220

Imm 8 107 70 177

Totals 1208 895 2103
% 574% 42.6% 100.0t.

Fig.13B
 (p.13-2)

Chapter 13
ARE THERE PATTERNS OF BEHAVIOR IN THE WEEKLY
SWINGS OF THE MARKET?

Are there habitual patterns of behavior in the weekly

movements of the market? In the preceding chapter, we
noted no significant tendency toward trends in the weekly
movements. In this chapter we will examine the weekly
behavior, in periods of two, three, four and five weeks,
considering the usual mixture of upward and downward
movements.

First, consider a two-week pattern. There are four

possibilities: UP-UP (up two weeks in a row) UP-
DOWN, DOWN-UP, DOWN-DOWN. How many times
have these four situations occurred in the last forty
years? What happened in the third week - the week
following the pattern? Is there a bullish or a bearish
bias in the week ahead?

The answers are tabulated in Figure 13A. In the

case of two rises in a row, for example, there were
707 cases in the forty year period. The market rose
420 times in the third week and declined 287 times. Un-
fortunately, the results for the two week pattern are not
significantly different from the expected results. No
bullish or bearish tags can be applied.

Now - consider the three week period. There are

eight possible patterns. These are in Fig. 13B. Here
we find a bullish pattern in number 5: DOWN-UP-UP.
If, in the last three weeks, the market has declined one
week and risen two weeks, the odds favor a rise in the
week ahead. In the forty year period, in a similar sit-
uation, the market rose 180 times and declined 109
times.

Pattern number 4 in Fig. 13B is interesting. For

an UP-DOWN-DOWN pattern, the record for the fol-
lowing week is UP: ll5 times; DOWN: 106 times. This
appears bullish. Actually, it has a bearish tinge, be-
cause the expectation based on all patterns in all weeks,
is UP: 127 and DOWN: 94. The market rose less than
expected and declined more than expected.

2-14
 (p.13-3)

JOUR. Will. PATTIRN B

1924 - JUNE 1964

BEHAVIOR OF THE D-J INDUSTRIALS

IN THE FIFTH WEEK

FOUR WEEK PATTERN MOVED MOVED TOTAL

PATTERN NUMBER UP DOWN OCCASIONS

luiululul 134 106 240

wmIUfm 2 103 71 180

~ 3 100 63 163

BEARISH~ 4 62 62 124

BULLISH~ 5 108 58 166

YhJ 6 66 46 112

Y 7 67 48 115

wm 8 G5 41 106

~ 9 106 74 180

A 10 60 49 109

N II 65 49 114

~ 12 51 4G 97

~DO
13 73 50 123

BEARISH ~ 14 48 49 97

J 15 57 50 107

16 42 28 70
~
Totols 1207 896 2 103
% 57.4% 42.G% 100.0%

Fig. isc
 (p. 13-4)

The four week patterns are charted in Fig. 13e.

There are sixteen possibilities.

Three patterns are especially interesting. Numbers

4. and 14. appear bearish, and Number 5. is bullish.
Number 5. has a very impressive record. In the week
following this pattern (UP-DOWN- UP- UP) the market
rose 108 times and declined only 58 times.

2-12
(p. 13- 5) 'IVE Will: PATTIRNS
1924 - JUNE 1964

BEHAVIOR OF THE o -J INDUSTRIALS

IN THE SIXTH WEEK

FIVE WEEK PATTERN MOVED MOVED TOTAL

PATTERN NUMBER UP DOWN OCCASIONS

~ 77 57 134

~ 2 68 38 106

my 3 66 37 103

~ 00
4 40 37 77

BULLISH
~ 5 67 33 100

8lJLL'SH~
o 0 6 42 21 63

~ 00
7 3B 24 62

~
o 0 0 8 37 25 62

~ 9 55 53 lOB

BEARISH~ 10 27 3 I 5B

~ II 40 26 66

~~ 12 20 26 46

~ 00 13 39 2B 67

~
o 0 0 14 23 25 48

~ 15 37 2B 65

~ 16 24 17 41

~ 17 5B 48 106

BEARI9H~ 18 35 39 74

~
o 0
19 35 25 60

~ 20 24 25 49

~ 21 40 25 65

BEARISH~ 22 23 26 49

~ 23 27 24 51

~ 24 29 17 46

~~ 25 52 2\ 73

~00
26 32 18 50

~ 27 25 23 4B

~ 2B 29 20 49

~ 29 34 23 57

~ 30 26 24 50

~ 31 20 22 42

32 IB 10 28
~
Totals 1207 896 2103 Fig.13D
100.0%
" 57.4" 42.6°.4
 (p.13-6)

The five week patterns are in Fig. l3D. The inter-

esting configurations are marked "Bullish" and "Bear-
ish" in the left margin. There are two especially sig-
nificant patterns. Number 25 (DOWN-DOWN- UP- UP- UP)
is especially bullish. This pattern appeared 73 times
in the last 40 years. In the week following the pattern
the market rose 52 times and declined only 21 times.

Parte rn number 12 (UP-DOWN- UP-DaWN-DOWN) is

especially bearish. This pattern appeared 46 times in
the 40 year period. The expectation for the following
week is UP: 26. DOWN: 20. The historical record is
UP: 20. DOWN: 26. This is a significant difference
from expectation.

2-10
 (p.14-l)

Chapter 14:
WHAT IS THE DOW THEORY? IS IT PROFIT ABLE?

The Dow Theory began with Charles Dow (1850-1902)

founder of the Dow Jones News Service. and founder and
first editor of the Wall Street Journal. In his capacity,
he wrote many editorials on the subject of the market.
These editorials are the foundations of the Dow Theory.
His work has endured. Today the interpretation of his
ideas is providing a livelihood for many men.

S. A. Nelson, a writer and publisher. tried to per-

suade Dow to write a book summarizing his ideas. but
he didn't succeed. Nelson had determination, however,
and collected all of the important editorials from the
files of the Wall Street Journal. He published these in a
small book "The ABC of Stock Speculation". In this
book he labelled the Dow chapters "Dow's Theory" --
and the name was established. Nelson laid the corner-
stone.

A few years after Dow's death, William Peter Ham-

ilton became editor of the Wall Street Journal. He pro-
ceeded to develop Dow's ideas from the status of general
statements into a workable method. In 1922 he wrote a
book "The Stock Market Barometer", which erected a
structure on Dow's foundation. A short time before his
death he supported his thesis by a famous editorial "The
Turn of the Tide". This editorial. in the October 23,
1929 Wall Street Journal (one week before Black Tues-
day) definitely called the end of the long bull market and
the beginning of the great bear market.

Robert Rhea (1887-11/6/39), because of an airplane

accident in the first world war, was bedridden from 1918
to his death. In Colorado Springs. he began a study of
the Dow ideas - first as a hobby. and later as a profi-
table vocation. His own wealth accelerated from the
practice of his ideas. He was long of stocks through
most of the twenties; he had no stocks at the time of the
1929 crash; he thereafter sold short for two years.
Perhaps I am prejudiced, but I believe that he contribu-
 (p.14-2)

ted more to the development of the Dow Theory than

Dow or Hamilton. You will understand my prejudice
when you learn that Rhea liked to count and to measure.
(This is the underlying thesis of the book you are read-
ing.) Rhea's writings are a goldmine of statistics for
the serious student.

Some very capable students of the Dow Theory are

active today. One of the best is Richard Russell of San
Diego. Russell has written a book "The Dow Theory
Today." He supplements his analyses of the market
with other market barometers.

Another capable interpreter is E. George Schaefer,

author of "How I Helped More Than 10, 000 Investors to
Profit in Stocks." In this book, he develops his ideas
of a "New Dow Theory. "

Another acute interpreter is Perry Greiner, who

was associated with Robert Rhea in the last years of
Rhea I s work.

Many books have been written on the Theory (See

Appendix All),but the fundamentals are not complicated.
Hamilton wrote "The essence of Dowis Theory can be
summed up in three sentences. In an editorial Decem-
ber 19,1900, he says 'The market is always to be con-
sidered as having three movements, all going on at the
same time. The first is the narrow movement from day
to day. The second is the short swing running from two
weeks to a month or more; the third is the main move-
ment, covering at least four years in its duration. I "

Hamilton's quotation was an oversimplification, of

course. Actually, it is an example of the beginnings of
the "Theory". The work of Hamilton and Rhea, and
others, has expanded the beginnings into a fair-sized
book shelf.

The outline which follows is a condensation of these

writings.

2-8
 (p.14-3)

DOW THIOILY BASIC I

B r e Ollit h r OU9h - [stoblishes(Al end

of Bull Market and beginning

of Bear Market at point (9)

01 preceding

Bear Market a beginning of Bull Market

Of point (I).

(A) - Both Industrials and Roils must breokthrough

(I) Li.e. confirm) for (l valid siQnal.

PRIMARY

These Primol'V Markets lost trom one yeor to several years.

PRIMARY SWINGS: (1)-(2). (3)-15), (61-(7),18)-(9), (9)-110),1111"(131,(141115)

IAlso collod "logs'.)

SECONDARY REACTIONS: (2)-(3), (5)-(6), (7)-(81, (101-(11),113-'4)

These lost from three 'Meeks to three months and retrace u.e. "correct")

1/3 to 2/3 of the p receding Primary Swing.

Fig.14A
 (p. 14-4)

THE DOW THEORY -- A SEVEN POINT SUMMARY:

Point (1) The fluctuations of the Dow Jones rail and in-
dustrial averages are a composite index of all the hopes,
fears, and knowledge of everyone who knows anything of
financial matters. They therefore can serve as a baro-
meter of the future. Hamilton wrote: "Consciously or
unconsciously the movements of prices reflect not the
past but the future. When coming events cast their sha-
dows before, the shadows fall on the New York Stock
Exchange. "
Rhea believed that students should concentrate their
studies on the averages - other statistics could be ig-
nored!

Point (2) The market consists of three movements, which

are going on at the same time:

(2A) The Primary Markets. These are the great

bull and bear markets. They last from one year to sev-
eral years. These great movements are fundamental in
nature, and are beyond the reach of manipulation.

(2B) Secondary Movements: (See Figure 14A). These

are called Primary Swings (or legs) when in the direc-
tion of the Primary Market. When in the other direc-
tion (rallies in bear markets; declines in bull markets)
they are called Secondary Reactions. The Secondary
Reactions are distinguished from minor changes by their
magnitude (usually sufficient to retrace one-third to two-
thirds of the preceding primary swing) and by their du-
ration (three weeks to three months. )

(2C) The minor or day-to-day movements. If the

Primary Movement is considered a tide, the secondary
is similar to the waves that please the surfboard en-
thusiasts, and the minor movements are the small rip-
ples. The minor movements are not considered impor-
tant in the Dow Theory.

2-6
 (p.14-5)

Point (3) The great bull and bear markets each consist
of three phases. These are not related to the secondary
movements described above in (2), and should be con-
sidered separately:

(3A) In a bull market, the First Phase represents

the improvement in public confidence and a correction
of the undervaluation developed in the last phase of the
preceding bear market.
The Second Phase is a response to improvement in
corporate earnings.
The Third Phase is one of rampant speculation, when
stocks are advanced on "hopes and expectations. "

(3B) In a bear market, the First Phase is a correc-

tion of the speculative excesses of the Third Phase of
the preceding bull market.
The Second Phase is a deterioration of prices in gear
with a slide in the earnings of shares.
The Third Phase is a final depression of prices
caused in part by distress selling.

Point (4) The trend can be determined by the action of

the secondary swings. Refer to Figure 14A. The rise
from point (1) to (2), at the time, might be considered
a secondary reaction in the preceding bear market. But
when prices turned upward at point (3), which is higher
than (1), the picture begins to look bullish. Then, when
prices broke through the level of (2) at point (4), the rise
is known to be part of a bull market which began at point
(1) •
In a downward swing, the same type of analysis
would be valid. If the drop from (9) to (10), the classi-
fication would be "s econdary reaction" until the drop at
point (11), when the classification becomes dubious.
Then, at the breakthrough at point (12), the classifica-
tion is definitely changed to "Bear Market".
This method of determining trends by "breakthrough"
has been aptly compared to the determination of tides.
by the extent of waves on a sandy beach. If the farthest
sweep of a wave is marked on the beach with a stick,
and the next wave carries past this point (a breakthrough),
the tide is probably rising.
 (p. 14-6)

Point (5) In determining the trend in (4), both Industrial

and Rail averages must be considered.----changes and
break-throughs by one average are not considered sig-
nals until the moves are confirmed by the actions of
the companion average. The confirming action, how-
ever, need not be on the same day.

Point (6) The averages sometimes move horizontally,

within a 50/0 band, for three weeks or longer. When both
averages do this, it is called a "Line". When both aver-
ages break out of this "line", in the same direction, an
important move in the same direction is probable. The
breakout is usually made with an increase in volume.
Hamilton noted that "lines" seldom occur at the be-
ginning or ending of secondary swings, but usually in
the center.
Rhea's analysis recommends the consideration of
"lines" of less than 5%. If both averages, for example.
have been moving in a 2% band, a breakout from this
would be significant.
Rhea found that in a bull market, downside penetra-
tions of "lines" are very speculative, while upside pene-
trations are more reliable.

Point (7) Volume: A market which is "overbought" be-

comes dull on rallies and develops activity on declines.
Conversely, in "oversold" markets, the tendency is to
become dull on declines and active on rallies.

Robert Rhea supplemented Hamilton's tabulations

with a detailed listing of all of the bull and bear markets
and secondary movements in the period from 1896 to
the time of his death in 1939. These tabulations have
been charted in Figures 14B and 14C. These charts
trace the Primary Markets with heavy lines; the light
lines are the Secondary Movements. In addition, the
points of confirmation of the primary markets are
marked with the letter "C". Profits and losses obtained
by buying and selling at these points of confirmation are
indicated by arrows. Since the confirmation always fol-
lows the top or the bottom of a primary market, the pro-
fits are always less than the total move. The data for
these charts are in Appendix A9)

You can profit from this theory. It doe sntt always

gi ve a correct forecast, and the forecast isn It always
clear. But the profits over the years, when cumulated,
are the equivalent of compound interest at the rate of
11. 7% per year.
2-4
 (p. 14-7)

DOW THIORY BULL • BIAR IIAI.IIT8 ...

1896 = 1919
:• 1>
1 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - ---- " .
~-----,----------M-II

00'11 Jones lP'\dustriols - Chart bCl8eo on data by Ro bar r Rhea --

----- 100
@ c ~ GONFIRNATION POINT

, s
I

. ~
c
V.ell
0

'"
~ V
7

~..
......
---
... 6

~------------l50

_c

"....
..
Itf-----;~------,;:--------.,_-----------------
..
;;
::
-

....
a
~

z ~
e
~ ~
E

II'
- -
.
" -
"AI i
•
-
251-------------------------------
-
Fig.14B
 (p. 14-8)

400----------

DOW TBIOIY lULL

I • •• AI MAIKETS

I 1919 =1939
-
-
Dow Jones Indusl ria Is --
-

--- - - - - - - -------,. IJ • 0 c..... --

.00 Hamilton's editorial 10/25/29
Chari based on dolo by Robert Rhea. 'J
-The Tur n of Tide,"

1~1
the

• C = CONFIRMATION POINT

250 - - - - - - - - - - - - - ----- --- ~--- -------

-

-
~

200 - - - - -- - ----_.,- ,,--- . __ ._-~

..
0

., N
~

- 0
a
..
----------- ---- ----
.
,

.~ ~---
C ;::

'50
-
- - --
A

~ fj
-
- ~

~
;e
- " t
11
0<
~
-
e

~t--
----- -----

'{
.
-~
!:
.
~

~
~
!! 0
a
1

~ ~c1-
'00 ~

_ 0

90
t
~~
. !V
\~
~

-
c : ~
;;;
c
00 ... --
\
70 --

60 -_._. ---- --_.

--
-

50- .... - _.. ---

-
-
-
~

~
40-------------------------=~-------- MA'

z-z FIGURE 14-C

 (p.15-l)

Chapter 15.
WHAT IS THE ELLIOTT WAVE THEORY?

Some refreshing ideas about the behavior of Wall

Street were developed by Ralph Nelson Elliott (187l-Jan.
1948). We will outline these ideas in this chapter. Ad-
ditional details are in App. 4, and an analysis is in App.
5)

After many pleasant hours of study and analysis, my

own conclusion is this: Elliott's ideas resist reduction
to specific rules for profit. They require interpreta-
tion; i. e. two Elliott enthusiasts in the same situation
will quite probably have two differing conclusions. If
you are reading this book for specific suggestions. you
can skip this chapter.

Ralph Elliott spent many years in Latin America as

an accountant. In 1927 he retired to Los Angeles where
he developed his wave theories. These were published
in a monograph "The Wave Principle" in 1938, in a ser-
ies of articles in the Financial World in 1939, and in a
book "Nature's Law" published in 1946. His ideas proved
so interesting to investors that he came to New York
and spent the last years of his life writing a financial
report.

Basically, he believed that prices tend to fluctuate

in natural ways. The most important tendencies, he
believed, were founded on the Fibonacci series of num-
bers: 1-1-2-3-5-8-13-21-34- ... Each number in this
series is equal to the sum of the two preceding numbers.
and the ratios are found many places in nature. The
count of seeds in the whorls of sunflowers and pineapples
are examples. Elliott found the numbers existing in the
timing of waves and in the ratios of stock market prices
at various turning points.

3-1
 (p. 15- 2.)

I BAS He I LLJGT T
BULL MARKET BEA R MA RKET
...
y

WAVES OF
MAJOR DEGREE ~

Jl:
5

INTERMEDIATE
DEG REE """
r
5

II JZ:
5

MINOR DEGREE """

 (p.15-3)

Elliott's most basic conclusions are illustrated by

Figure 15A, a chart which is derived from one that he
published in "Nature's Law". The tendency that he il-
lustrates is simply that moves in the direction of trend
tend to be in five waves; those in the contrary direction
tend to be in three waves.

The top chart illustrates the "major" or primary

waves in a bull and a bear market. The bull market has
five waves; the bear market has three.

The center chart illustrates the tendency toward the

breakdown of the primary waves into waves of the next
smaller or "Intermediate" degree. Note that each ma-
jor wave in the direction of trend (up in a bull market;
down in a bear market) is made up of five waves of in-
termediate degree. These are marked I, 2, 3, 4, 5. Each
wave counter to trend is made up of three waves. These
are marked a, b. c.

The lower chart is a continuation of the same basic

tendency. The longer of the "Intermediate" waves is
made up of five waves of minor degree; the shorter of
the "intermediate" waves breaks down into three waves
of minor degree. An exception is the wave marked
"flat". (This is a variation which will be described in
Appendix A6. )

3-3
 (p. 15-4)

Unfo r tunat e ly, the simple pattern of Elliott's "basics"

(Fig.15A) does not line up very well with the record.
You can check this visually by a review of the bull and
bear market charts in Appendix AlO. Elliott took care
of this by designing a series of variations. (These are
in Appendix A4)

For example, consider Elliott's basic pattern for the

direction of trend: (five waves) The score, when applied
to the record:
Less than 5 waves: 58. 60/0
Elliott's basic five: 21. 4
More than five: 20.0
100.0

Elliott's basic pattern for corrections is three waves.

The score:
Less than 3 waves: 65.4%
Elliott's basic three: 26.0
More than three: 8.6
100.0

In other words, the exceptions are rno r e prevalent

than the basics.

(Footnote for students: See variations in Appendix A4;

a support for the conclusions above in Appendix A5)
3-5
 (p.16-1)

- DE'INITIONS

BULL MARKET

To qualify I must be
PRI MARY SWINGS
,,
at least 40% of (B),
•
or more than one OR LEGS •,
year in duration.

SECON DARY
REACTIONS
---'-----(8)

To qualify, must be
at least 5% of (Al
and one week in duration.

BEAR MARKET

To qualify, must be ,
at least 40°' of (81, ,, , SWINGS

or more than one

, LEGS
yaar in dura t Ion.

, ,
,
,,
SECONDARY
REACTIONS
IB)-
To qualify, must be at
least 5% of (Al and one
week 10 duration.

FIG. 16-A
 (p.16-2)

Chapter 16.
HOW LONG SHOULD YOU EXPECT A BULL MARKET
TO LAST? A BEAR MARKET? A PRIMARY SWING?
A SECONDARY REACTION?

Suppose that we are in a bull market - or a bear

market. How long should we expect to continue? What
is the life expectancy?

Suppose that we are in a primary swing - or in a

secondary reaction. How long?

In this chapter, you will find some life expectancy

tables, which are based on the experience of two-thirds
of a century.

But - what is a bear market? A secondary? The

Dow definitions (Chapter 14) are too fuzzy for our pur-
poses. Some "authorities" called the 1962 slide a Bear
Market; others. using the same Dow definitions, called
it a Secondary Reaction. Some say that we are in a Bull
Market which began in 1949; others say 1937, or 1962.
We must define our terms with more precision before
we can proceed with analysis.

There is no ultimate authority. To permit analysis,

we have made our own definitions. (If you disagree, you
can modify these definitions, and with the help of a com-
puter, set up modified tables.) We believe that the defi-
nitions in Figure 16A are defensible, for they are de-
signed to approximate the swings tabulated by Hamilton
and Rhea. (The analysis is in Appendix A6)

These definitions might be considered a sharpening

of the Dow Theory, except that we have not included
"Percent Retracement" or "Confirmation by the Rails"
in the definitions. Instead, our definitions use amplitude
and duration only. They provide a means for the definite
classification of all waves into four classes. which ap-
proximate those delineated by Dow Theorists.

3-7
 (p.16-3)

With the solid foundation of specific definitions, we

can erect a very i.nteresting and useful structure. We
have summarized this in the "Life Expectancy" tables
which follow. We are using the life insurance technique,
but we will think positively, and shun the name "mortal-
ity table". These tables are based on all swings of the
D-J Industrials from 1897 through 1963.

Note that these life expectancy tables never say

"move ended". Instead, as with human life expectancy,
there is always an expectation for continuation. If we con-
sider all men who are 95 years old, we know that they
will all continue to live, but for varying periods. The
average time is the additional life expectancy.

These tables consider the life expectancy of a move

in terms of amplitude of swing in %. duration of swing in
days and in number of legs. Here is an example. Sup-
pose that we are in a bull market. We have advanced
67.6%; the duration so far is 880 calendar days; we are
in the fourth leg. What is the expectancy for continua-
tion? This was the situation November 1. 1964.

First, consider the size of the swing in Fig. 16B. The

life expectancy in a bull market after a swing of 600/0, is
47 percentage points additional; after a swing of 70% it
is 42%. Our swing is 67.6%; interpolating we estimate
43 additional percentage points. This means that. in the
past, the average bull market carried on for an addi-
tional43 percentage points, or to a complete swing of
lll%. (68% plus 43%)
 (p.16-4)

DEFINITIONS - Used in the following Life Expectancy

Tables (See Fig. 16A and Appendix A6):

(1) The great primary Bull and Bear Markets include

all swings of the D-J Industrial Average with an ampli-
tude of more than 40% 2!: a duration of one year or more.

(Z) The Primary Swings or Legs are smaller swings

in the direction of the primary market, which are
terminated by Secondary Reactions.

(3) Secondary Reactions are contrary swings within

the bull and bear markets which exceed 5"/0 in
amplitude and one week in duration.

(4) Minor Swings are all swings smaller than (Z) or (3).

LIFE EXPECTANCY - PERCENT SWINGS

Percent Life expectancy, additional percentage points

Achieved: Bull Markets: Bear Markets:
30 67 4l.
40 58 40
50 5Z 35
60 -f1.- Z8
70 4Z ZO
80 38 13
90 37 6
100 54
lZO 88
140 107
160 110
180 III
ZOO 114
Median total % swing:
-- 96% 66%

Fig.16B

3-9
 (p.16-5)

Now consider duration. In Fig. 16C. the expectancy

after 800 days is 590 more days; after 900 days it is
580 more days. If we interpolate our duration of 880
days. we estimate the duration to be 582 days. This
means that the average bull market which, in the past.
lasted for 880 days. carried on for an additional 582.
more days.

Finally. consider legs. In Figure 160. we note that

after four legs. the expectancy. on the average. is for
three more legs.

You can use the same method for the expectancy in

primary swings and in secondary reactions. The tables
are in Figures 16E. 16F. 16G. and 16H.

These tables should help you to make more profit by

adding perspective to your decisions.

Footnotes for students;

(1) In order to make downward swings comparable to

upward swings. we have made 1000/0 the lower point in
all cases. A swing from 80 to 100 and back to 80 is
called a 250/0 swing upward and a 250/0 swing downward.
Our 400/0 swing requirement for a bear market is the
equi valent of a 28.60/0 reduction; the 50/0 requirement for
a secondary is equivalent to a 4.760/0 reduction.

(2) The tables are based on smoothed data. The actual

turning points are tabulated and charted in Appendix AIO.

(3) The data for secondaries (0/0 swing and duration) in-
cludes the first primary swing of the following primary
market. since the latter is not distinguishable from the
previous secondaries until the primary market is def-
initely established.

(4) The tables exhibit some strange changes. At times

the figures decrease. then increase. then decrease.
(See Fig. 160) The reason is not clear. Perhaps the
samples are too small for regularity.

(5) Duration is measured in calendar days. not trading

days. See Appendix Al3.
 (p.16-6)

LIFE EXPECTANCY; DURATION IN DAYS:

Days Life Expectancy - Addional Days:

Achieved: Bull markets: Bear Marl<ets:
100 640 540
200 630 460
300 630 380
400 620 300
~ ""610- 230
600 600 200
700 600 190
800 590 170
900 580 160
1000 580
1200 560
1400 550
1600 540
Median total days duration:
-- 794 629.

Fig.16C

LIFE EXPECTANCY - NUMBER OF LEGS -

Legs Life expectancy - Additional Legs:

Achieved: Bull Markets: Bear Markets:
1 6 4
2 5 3
3 4 2
_4_ 3 1
5 2 1
6 2 3
7 3 2
8 4 2
-9- 4 1
10 4 o
Median total legs: ""6."""7 ---s.o
Fig.I6D

3-11
 (p.16-7)

LIFE EXPECTANCY - PRIMARY SWINGS - PERCENT SWING

Percent Life Expectancy - additional percentage points:

Achieved In Bull Markets: In Bear Markets:
5 11 12. 3
6 10 11. 3
7 9 11. 1
_8_ 9.5 11.1
9 9.8 11.1
10 10 11.1
12 10.5 10.7
14 10.7 10.0
16 11 9.4
18 11 8.8
20 11 7.8
25 10.8 ..........;.9:...4~ _ _
30 10. 1 14
35 9.1 19
40 8 24
45 7 27
50 5.1 27
Median Total "!o 16.0 ~

Fig.16E

LIFE EXPECTANCY - PRIMARY SWINGS - DURATION IN DAYS -

Days Life Expectancy - Additional Days:

Achieved: In Bull Market s: In Bear Markets:
7 70 37
10 67 34
15 64.5 32.5
20 64 29
30 61 30
40 56 30
50 50 29
-2L- 44 27
70 45 24
80 46 22
90 48 18
100 48 14
150 50 14
200 47
300 33
Median total duration: 77 44

Fig.16F
 (p.16-8)

LIFE EXPECTANCY - SECONDARY REACTIONS - PERCENT SWING--

Percent Life Expectancy - Additional Percentage Points:

Achieved: In Bull Markets: In Bear Markets:
5 5.0 6.0
6 5.0 7.0
7 4.8 7.7
8 4.5 7.8
9 4.2 7.5
10 3.8 7.0
12 3.6 6.1
14 4.2 6.0
16 4.5 6.7
18 5.0 7.1
20 5.0 8.0
25 6.0 9.8
30 3.0 11.5
Median Total Swing: ~ 11.1

Fig.16G

LIFE EXPECTANCY - SECONDAR) REACTIONS - DURATION IN DAYS -

Days Life Expectation - Additional Days:

Achieved: In Bull Markets: In Bear Markets:
7 18 31
10 19 29
~ __
2 1__ 29
20 22 29
30 22 29
-1Q...... 21 ---.l.L-.
60 40 36
80 61 41
100 87 45
150 80 53
200 40 57
250 60
300 55
Median Total Duration ~ 3"8
Fig.16H

3-13
(100-;- %YIELO)

'l='
-~
.....
I
~

17A
'65
CAl DIVIDENO RATI
... 0 ... . . ... l!.~ ..
e ..
 (p.17-2)

Chapter 17
IS THE MARKET INFLUENCED BY EARNINGS?

The value of an investment at any time is the price

at which it can be sold. Earnings and dividends should
be an important factor in this value.

It is clear that, at any given moment, the di vidends

and earnings of individual stocks are definitely impor-
tant factors in the relative prices.

But when we consider different times in history, in-

vestors have evaluated these factors through a wide
range.

Note Figure 17 A. This chart plots the cost of buying

one dollar of dividends in the D-J averages.

The cost swings through a wide range. There were

times when you had to pay as much as $36.90 to get
enough shares to pay one dollar a year; at another time
you could have made the same purchase for $10.

The points in the chart have been divided into three

areas. The top one quarter of the points fall above $28.
This has been labelled the "expensive range." The low-
est one- quarter fall below $17. This has been labelled
the "bargain area". The median price for the entire
period is $22.

When prices are in the expensive area (They are in

this area at the time of writing), it appears that one
should not hold out hope for much higher prices; when
they are in the "bargain area" one should not expect to
buy at much lower levels.

There seem to be limits in the judgment of investors.

Investors aren't willing to pay much over $28; they
haven't been willingto sell for much less than $17.

3-14
 (p. 18-1

l!_B_E IFHE PEIRCENT PENETRATION MIT80D _

SELL on a
Downward Penetration

Ignore all Turning

wi th swings of less
than 5 % (100% is

Profit

Buy -)- - - - - - - - - - - - - - -

______ .; _"~~I~s_a~"_ _ _ _ _ _ _ Loss

Profit

on an Upward Pe ne rr et ion
at a Turning Poinl Fig. 16-A
 (p.18-2)

Chapter 18 -
IS TREND FOLLOWING PROFIT ABLE?

Is it practicable to follow the longer swings of the

market? When does a short turn become a trend?

The Dow Theory and the Elliott Wave Theories are

investigations in this direction. In both of these inves-
tigations, a considerable amount of "interpretation" is
required. The case is never specifically' defined.

In this chapter, a simple and definite trend following

method will be outlined and tested.

It is described in Fig. 18A. It uses the "break-

through" idea in the Dow Theory, but it rigidly defines
the qualifications. A hundred men using the Dow Theory
could have a good debate on the subject of a break-
through; a hundred men using the Elliott Wave ideas
would have a real donnybrook; a hundred men with Fig.
18A would come to identical conclusions. The results
of this method over any span of time can be measured
defini tel y •

The figure "5%" is not essential. The important

point is that it's specific. If you select a higher percent,
there will be fewer errors in the selection of trend;
there will be fewer losses. Balancing this advantage will
be a lower profit; you will buy later and higher in a
trend; you will sell at a lower figure.

The results have been tested over two periods of

time. The first is the 1897-1939 period for which Ro-
bert Rhea tabulated the Dow Theory signals. For this
period, the result of the Dow Theory is quite excellent.
(This was reported previously in Chapter 14) The pro-
fit, including short sales, averaged 12.9% per year. H
2% is deducted from each trade (one buy and one sell),
the rate is reduced to 11. 7% per year. If the short-
sale profits are excluded, the rate is reduced to 8.4%
per year.

These figures are a considerable improvement over

buying and holding for the entire 42 year period, which
is 3.5% per year.

3-12
 (p.18-3)

The score for the 1150/0 Penetration Method" is almost

as good. before the deduction for expenses: 12. 340/0 per
year. However. since there were 84 trades in the period
(The Dow Theory had 2Z). the overall profit after expen-
ses is reduced to 7.90/0 per year. If short sales are ex-
cluded. the rate for the period is 6.60;0 per year. Both
of these figures. however, are substantially better than
the buy-and-hold rate of 3.50/0 per year.

One disadvantage in the "Five Percent Penetration"

method is the number of whipsaw trades. In the 84.
trades. no less than 37 were made at a loss. The los-
ses, however. averaged much smaller than the profits.

The method has been checked back over the entire

period from January I, 1898 through November I, 1964.
In this period there were 123 trades. Of these 63 were
profitable and 60 were made at a loss. The median pro-
fit, however. was 10.30/0 while the median loss was only
6.30/0. The largest loss was 16.80/0. while there were
eight profits larger than 500/0 and one of 1040/0.

The average growth for the entire period. after de-

ducting 20/0 for each trade. amounted to 4.920/0 per year,
including short sale trades. If the short sales were ex-
cluded, the total rate of income was not as high. as ex-
pected. However. since there were half as many trades.
the net income after expenses actually rose to 5.50/0 per
year! Both of these figures are an improvement over
the buy-and-hold figure of 4.850/0.

The conclusion is that a definite policy of trend-fol-

lowing would have produced profits. However. the use
of short selling with this method is dubious.
3-10
 (p.19-2)

Chapter 19:
ARE SOME PRICES PREFERABLE TO OTHERS FOR
PURCHASE? FOR SALE?

Analysts have long noted "resistance points" and

"support levels". In the former, resistance to a rise is
created by the offering of stock at a certain level; in
the latter, support arrests a decline by supplying purch-
ase bids.

These resistance and support points have been found,

in individual stocks, to be located near previous turning
points. We have checked this tendency to see if it ap-
peared in the D-J average. The major turning points
from 1914 to 1964 have been charted in Figure 19A. In
this chart note any tendency toward clustering. This
would indicate a point of persistant resistance or sup-
port. There aren't very many clear supports for the
theory in this chart.

There are a few interesting spots. Note 71. 8 72.8

76.1 90.2 96.2 136-138 142-146 176 186-187 570630
690

3-8
 (p.19-3)

=DUTI.IBUTRON' WITHllN THE POINT=

-- All Common Slacks on N.Y.S.E --

I 1961 HI G H PO I N TS : I
1/8= 12.5%

II 96 2 LOW POI NT S: I
1/8 = 12.5 %

7/8

3/4

5/6

3/6

1/4

1/8

o 26.4%

FIG. 19-8
 (p.19-4)

There is a very definite tendency for support and

resistance based on the number itself. Retail stores
have found that a price of $1. 97 is more attractive than
a price of $2.01

Note this tendency in Fig. 19B. All of the common

stocks on the N. Y. Stock Exchange have been examined.
(Preferred stocks have been excluded.) In the upper
part of the chart the high points in 1961 have been tabu-
lated. All of the price has been ignored except the frac-
tion at the very end. For example: a stock with a high
of 195 would be in the upper bar; one with a high of 6 7/8
would be in the next bar; one with a high of 26 3/4 would
be in the third bar.

Note that the even number is a strong resistance

level. 19.6"/0 of the stocks topped out at the even price.
The next most popular resistance is at the 1/2 wi th
18.8"/0. The quarter points follow; the eighths are least
popular.

Stock seems to be offered for sale more often at

the even price; the half is almost as big a resistance
level.

If you plan to sell at the top, then, you should avoid

offering at the eighth, since the price will probably move
up to the next quarter.

The reverse, of course, is true on a stop-loss sale.

These should be placed on the eighth points, since prices
will probably not stop at the eighth. but will move higher.

3-6
 (p.19-5)

Consider the lower half of Fig. 19B. The low points

of all common stocks on the NYSE were tabulated for the
1962 decline.

The even price, again, is the most important sup-

port. 28.4% of all stocks stopped their decline at an
even price. A lot of purchase orders seem to be placed
at this price. The next most important support point
is the one-half, where 17. 5% of the stocks stopped their
declines. The 114 price is close behind.

For profit on purchase, then, one should not place

purchase points at the 1 I 8 level, since prices, when they
hit that level, will probably continue to the next even
quarter before reversing. For stop loss, the reverse
should be true. This order should be placed on the 1/8
points, since a price will probably continue to the next
1 14. The 718 point (just below the major support) should
be a good place for a stop-loss.
 (p.20-l)

Chapter 20:
CONCLUSIONS:

The major conclusion of this book is this: The mar-

ket has definite upward tendencies at certain predictable
times; it has leanings in the downward direction at other
predictable times.

Some observers have noticed a feed-back effect in

market forecasts. When a barometer or a method is
published, it tends to affect the behavior of readers;
this behavior in turn tends to affect the market. It will
be interesting in future years to note if the findings in
this book become spread far enough to produce a feed-
back. If so, they may tend to advance the timing of
changes. If the market has, in the past, shown a ten-
dency to rise on a certain day, buying in anticipation of
the rise on the preceding day may produce an upward
bias on that day.

The study of these inclinations should continue, there-

fore, and not be considered a static collection of knowl-
edge. Readers should note new developments, and add
these to the background reported in this book.

The possibility of change, however, should not dis-

courage the investor. The inclinations of the market
reported in this book have a long and substantial found-
ation. The market is widespread; it has inertia.

The inclinations and biases reported in this book

should be useful both to speculators and to investors.
If purchase is considered, and the market has an inclin-
ation downward, a prudent man should delay purchase.
If a sale is contemplated, and the market has a down-
ward inclination, a prudent man should sell immediately.

The recognition of these inclinations, which are re-

viewed in the preceding chapters, should reduce l os se s
and increase PROFITS.

3-4
 (App. AI-p. I)

Appendix A I:
ARE THE DOW JONES INDUSTRIALS REPRESENTATIVE?

In 1896, Charles Dow began to report the averages

of a few selected industrial stocks. After 67 years of
use, it is still the most widely-quoted measure, although
a number of very sophisticated indexes have entered the
field.

Twelve industrial stocks were used until 1916, when

the list was increased to 20, and the new average was
worked back to December 1914. Only the closing aver-
ages were calculated. The method was simple. The
stock prices were listed and divided by the number of
stocks to get an average. If a stock split, a multiple
was applied before adding with the others; if the split
were two-for-one, for example, the stock price after
the split was always multiplied by two before combining
with the other prices.

On October 1,1928, a number of changes were made.

The number of industrial stocks was increased to 30 (the
current number); the closing prices were supplemented
by calculation of the highest and lowest averages for
each day, and the method of coping with splits was al-
tered. The multiples were eliminated, and the adjust-
ment was made in the final divisor. The divisor was
also adjusted when a new stock was substituted for
another in the list, in order to gi ve continuity.

To calculate the current average, list the stocks

(you will find them occasionally in Barron's and the Wall
Street Journal). Set down the prices; add them up;
divide by the divisor (See Barron's or the Wall Street
JournaL); the result is the index.
 (App. Al-p. 2)

There are four principle limitations:

(1) A minor irritation is the high price. $800 is a

number that many find misleading. An eight-point
change sounds big - but it is actually 10/0. If this irri-
tation disturbs you, simply divide the index and its
changes by ten before interpreting.

(2) A more serious objection is levelled at the high-

low calculation. These are not highs and lows for the
index; instead they are averages calculated from the
individual stocks in the index. Srnc e these highs and
lows occur at different times in the day, the average
at any given moment is always lower than the daily
"high" and higher than the "low for the day. "

(3) Representation: The D-J reports the price change

of 30 stocks. Are they representative of the entire mar-
ket?
The stocks have been selected because of their im-
portance. If you total the market value of the 30 stocks
(multiply price by number of shares listed), you will find
that this small group of companies is more than one-
third the value of all stocks listed on the N. Y.Stock Ex-
change. This is a very important list!
But this selection represents only the "blue-chips".
It is far from the random selection advocated for some
purposes by statisticians. Two-thirds of the market is
not represented. If the blue chips rise, the D-J rises;
the remainder of the market may be declining.

(4) Weighting: Statisticians wince when they learn

the D-J weighting method. In a representative average,
each stock should be weighted by some rational method,
so that its importance in the index is properly reflected
by the index. In the D-J, the weighting is the price.
Owens Illinois Glass is more important in the index than
Standard Oil of New Jersey, because its price is higher.
(The total value of the stock of Jersey is more than 25
times that of Owens Illinois. )

3-2
 (App. AI-p. 3)

DOW-JONES INDUSTRIAL.S
700
USE LEF HAND CALE.-
70

600 60

50
50

40

30

300

MAl

1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964

Fig. AlA
 (App. Al-p. 4)

The lack of logic in the weighting method is evident

at the time of a stock split. If a D-J stock is split four-
for-one, the importance of the stock in the index (if this
influence is measured by the effect of a 10% change in
price) is reduced to one-quarter of its importance be-
fore the split. Has the split changed the true import-
ance of the company?

(It should be not e d that lower-priced stocks are

usually more volatile - see Appendix A7. This charac-
teristic helps the "Dow". The change of one dollar in
the price of each stockin the index carries equal weight.
It is easier for a hundred dollar stock to move one dol-
lar than for a $20. However, it's easier for a $20. stock
to change 5% than for a $100. stock. This characteristic
off-balances some of the illogical weighting after a stock
split. )

After we consider all of the limitations, the D-J still

seem to represent the market quite well. We have
compared it with the Standard and Poors 425 Industrials
in Figure AlA. The latter index represents some 87.3%
of the total value of all stocks on the NYSE, and is
weighted in a rational manner. If the scales are adjusted
it is found that both indexes move together in a remark-
able fashion.

The conclusion, after a study of Figure AlA, is that

the Dow Jones Industrials represent the market quite
well.

4-1
 (App. A2-p. 1)

TIINDS : CONTINUATION or IUJNS

(Ok of times that runs continued in the same direction, e. did not reversel

RU NS UPWARD I I RUNS DOWNINARD"l

0/0 6,10
60 60'----r---.-------.-
<, ....,

.. ....... .. .. .. .. ... ...

50 50
Jan. 1962- ....... .... ... 0 •••••• .. .....
June 1964

40 40 ~
I 2 3 4 5 1 2 3 4 5
- RUN - - RU N -

60,.----.-----,----,------,-- 60

g DAYS
50
Jan.1697-
June 1964

401-----L-----1------L------J- 40
I 2 3 4 5 I 2 3 4 5
- RUN - - RUN -

6 60

50
/ r--.... ~ )..:, .....
....,

50
l
j
Jan.1697-
June 1964

40 40
1 2 3 4 5 1 2 3 4 5
- RU N- -RUN-

FIG. A2-A
 (App. AZ-p. Z)

Appendix AZ:
IS THE MARKET RANDOM?

This appendix deals with market~.

A "run" is a continuation of moves in the same di-

rection. If the market goes down. then up. then down:
the run is one in the up direction. If the market goes
up. then down four times, then up: the run is four in the
down direction.

Hourly runs have been examined from January 1962

through June 1964. Daily runs have been noted through
every day from January 1897 through June 1964. Week-
ly runs have been studied for the same period from Jan-
uary 1897 through June 1964.

Figure AZA charts the percent of the times a run con-

tinued. You will note the tendency toward continuation
in the case of the shorter runs (hours. days). The runs
in the weekly series don't have the same inertia. and
make a good case for a "random walk". (Pure chance)

4-3
 (App. AZ-p. 3)
RUNS IN BOURLY DATA
-UPWARD- -DOWNWARD-
50 0 500
40 0 400
;~
30 300

20 o
K,. 200
:'\
~
15 0'
"'-
'\. 150
- '\
'0 0
80
'I.
100
80
....
"\\. '\.
6 0,

40
".
\"
60

40
,
'\ \
30
30
\.-
". ~ACTUAL
0
~
20 ,..'.:,,-
\,:,,~
: (:\rACTUAL
10
';:'\
10 ..... :"
e 8
,,:'.'':-'':\
':.\.
6 6
'.' .:~ \ ...: ....
4 4
"::::'1 ".::.::,\
3 3 ..,
2
EXPECT E D-"'\ A 2
EXPECTED-"" J
'V::,\ ,
\:;
'\
I
2 3 4 5 6 7 8 10 1\ 13 14 15
I
134567891011 .. t
I
"
') 16r
- Length of Run - Hours -Lenqth of Run-
FIG. A2B
I Ptriod - 3691. Hour. Jan. 1962 - June 1964; CoJ Indu.trials)

RUNS IN DAR LY DAT A

-UPWARD- -OOWNWARO-
3000 3000

2000 2000

1000
800
~ ~
1000
800
f\,
-
\. ~
600 600
... 400
"

\. 300
T
~ 200
\
'\ '00
\ ~
"'\. 80
80
'\.
,'\.
,.\
&0
40
30
,
'\

~'t-ACTUAL
20
~ACTUAL

\ "'- EXPECTEO~\
10
8
,
A 6
\
EXPECTEOJ;: ~.~
4
3
-V
\,
\1,
,
" 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 8 10 II~ '2 13 ~ I
- Lenqth of Run - Cays - - Lengfh or Run"
FIG. A2e
(Period - 19044 Cays Jan.IS97- June 1964; CoJ Industrial.)
 (App. A2-p. 4)

• RUNS IN WilILY DATA =

UPWARD-- DOWNWARD--
50 0 500
---
30
0f\----=---------~~---- 300
\.

20 0

10

0
-. -
-
20 0

10 0

0
\.
\
,
-lI.. 50 \,
~ \.
0 30
'~ACTUAL \.
2 80

10
\~.... 10
EXPECTED~ACTUAL
7 ._~

'\:~ \~
51----_
3
~ [?
2
EXPECTED --<\ \ :\

1
1 2 3
___ L--l.--I
4 ~ 6
I".
7
.i.,
6 9
,\
10 1\ 12 13
1
2 3 4 5 6
C%.
7 8
I
- Longlh of Run - Weeks - - Lenglh of Run-
FIG. A2-0
(Period - 3484 Weeks Jon. 1897 - June 1964)-; O-J Induslriole)

The detailed data are charted in Figures AlB, A2e,

and A2D. In these charts the number of runs is charted
against the length of the r un ,

Note the large difference between expected and actual

in the hourly chart. (Fig. A2B)

In the daily and weekly charts, the actual is not far

from expected, except that the difference in the case of
a one-day r un is significant.

4-5
(App. A3-p. 1)

(e)

(e)

Fig. A3A
 (App. A3-p. 2)

Appendix A3:
HOW SHOULD TURNING POINTS BE MEASURED?

"Nat'ralists observe a Flea

"Hath smaller Fleas that on him prey
"And these have smaller Fleas to bite 'em,
"And so proceed ad infinitum. "
(Dean Swift - 1733)

In this appendix a method of c Ia s sification and meas-

urement is proposed which permits the definite meas-
urement and tagging of every market turning point. This
method has been found useful m the analysis of the Dow
Theory and the Elliott Wave Theory. It has helped in
establislunent and testing of the "50/0 Penetration Method"
described in Chapter 18.

The method is quite simple. Turning points are

first grouped into definite pairs; the vertical distance
between the two points in the pair is a measure of the
importance of the two turning points in the pair. The
following is the proposed method:

(1) Pairing must follow these requirements:

(See Figure A3A)
(a) Betwen points (c) and (d), in either chart,
the curve has reached the level of (A) twice and twice
only, at points (A) and (d).
(b) Between points (c) and (d) the curve has
dropped to the level of (B) twice and twice only, at points
(B) and (c).
(2) The importance of turning points (A) and (B) is
called "degree". This is equal to the vertical distance
between the turning points (A) and (B) in percent of the
lower turning point (B):

Degree of (A) =Degree of (B): (e)xlOO%

(B)

4-7
 (App. A3-p. 3)

390 ~·------,-----l- --- 1-" -_._-- .,-------

_PEAK ___ 386.10 ... 9EPT. 3
....__. --- _ .. - TURN ING POINTS

. 5% DEGREE 6 HIGHER-
- - _.. _--- ---
S -DO W JONES INDUSTRIALS
~t--'-
. __ ..
-0 oi Iy -
'--1
350 350 \--~I-+~'------ - - - - - - ------
I-~u
~I 0 I
f---+-+----+-I-----------------i
-\ff----arll- ~
--
A
c I
. [

-----.----l.
II-

.. .,.
II-
300
IL '"
co .
=
_~_.,,_.

~I
.,;
- .... ._--------j
,. I
H
H co
iii
i
N
E

J1~
E
#. ,. .
11-
....
~

L ,m·~'.h
"!
01 ..
0

~
""
250 250
J J
lIP!
''''0
>- 'h.: !I
~
.,
... ·W 0 ~.~
:J II M
1
.."
0 ~ Q

;n 1

-. G
11- 0

--~
..'" '"
co ...

20 0 200~--

K K
--_ ..
~

I OCT.
L--_ _ 'l---l NOV.
---_._--
OEC_ OCT_

1919

Fig. A3B
 (App. A3-p. 4)

The paumg is not difficult. You will find that a

turning point can be paired with one and only one other
turning point. The 1929 peak, for example, is definitely
paired with the 1932 low point; the November 1961 peak
is definitely paired with the June 25, 1962 low point.

In the case of the most recent turning points, the

classification may have to be tentative until the two re-
quirements (above) have been fulfilled.

When turning points are paired and tagged, the

swings or waves between them can be measured. It is
suggested that the following nomenclature be used:
(1) A swing connecting two paired turning points
is called "a complete swing." A swing between two non-
paired turning points is called a "partial swing".
(2) The amplitude of a swing is considered to be
the vertical distance in percent of the lower turning
point. This will make upward swings comparable to
downward swings.
(3) The duration of a swing is the time interval
between the two turning points.

Figure A3B is an illustration of this proposed method

of classification.

4-9
 (App. A4-p. 1)

Appendix A4:
IN ELLIOTT1S THEORY, WHAT ARE THE VARIATIONS?

All of Elliott I s published writings have been re-

viewed. The notes are summarized in this appendix. An
analysis and an application to the movements of the
market are included in the next Appendix AS.

In Figure A4A and A4B we have summarized and

combined all of Elliott's various examples for the period
1929 through 1945. This period spans the time begin-
ning with the first Dow Jones high-low statistics and
ending with the last analysis published in "Natur e ts Law,
a few months before his death. The nomenclature has
been standardized, in order to make it easier for a stu-
dent to trace through Elliott's interpretation of the wave
patterns.

In addition to his basic conclusions listed in Chapter

IS, he tabulated many variations. which are charted
completely on the following pages.

He listed some subsidiary conclusions:

(l) No confirmation is required by a companion

average. (The Dow Theory requires confirmation of
the Industrials by the Rails.)

(2) Actual high and low figures are used, rather than
the closing prices. Elliott said "in fact it was only with
the establishment of the daily range in 1928 and the
hourly averages in 1932 that sufficient reliable data be-
came available to establish the rhythmic recurrence of
the phenomenon called the "Wave Principle".

4-11
 (App. A4-p. 2)

(3) News has little effect on the course of a wave

series. It may affect the amplitude and timing.

(4) Prices tend to move in channels. These can be

useful in the interpretation of waves.

(5) Elliott listed his names for the degree of waves

in the following order:
Subminuette (only in the hourly data)
Minuette
Minute
Minor
Intermediate
Primary
Cycle
Super Cycle
Grand Super Cycle

(6) There are many exceptions to the basic pattern.

These are described in Figures A4C through A4L.
 (App. A4-p. 3)

IElliott's Waves

DATE T.P. DEG. WAVE DATE T.P. DES. WAVE

11/28/28 17.69 4/18/38 14.19 1

3/26/29 15.27 A 5/27/38 14.19 2
5/6/29 14. 13 1 7/25/38 14.44 3
5/31/29 14. 13 2 9/28/38 14.44 4
9/3/29 851. 92 B 3 11/10/38 63.04 A 5
11/13/29 52. 16 1 1/26/39 12.20 1
4/16/30 52. 16 2 3/10/39 12.20 2
5/5/30 10.97 a 4/11/39 31.43 B 3
5/14/30 10.97 b 6/9/39 9.13 1
6/25/30 19.00 c 6/30/39 9.13 2
9/10/30 19.00 d 8/3/39 14.30 3
12/17/30 27.52 3 e 9/1/39 14.30 4
2/24/31 27.52 4 9/13/39 31. 43 IV C 5
10/5/31 39.34 a 1/15/40 6.31 1
11/9/31 39.34 b 4/8/40 6.31 2
1/5/32 28.62 c 6/10/40 25.69 A 3
2/19/32 28.62 d 8/12/40 5.50 1
7/8/32 851.92 I C 5 e 8/16/40 5.50 2
9/8/32 63.83 A 11/8/40 25.69 B 3
12/3/32 18.60 1 5/1/41 14.22 1
1/11/33 18.60 2 7/22/41 14.22 2
2/27/33 63.83 B 3 12/24/41 8.95 3
7/18/33 34.46 C 1/6/42 8.95 4
10/21/33 34.46 1 4/28/42 Ill. 02 V C 5
2/5/34 32.34 2 7/15/42 4.78 I
7/26/34 32.34 D 3 8/5/42 4.78 II
12/6/34 5.36 1 11/9/42 4.16 A
3/18/35 12.86 2 11/24/42 4.16 B
11/20/35 8.02 3 4/6/43 5.90 C
4/30/36 15.22 4 4/13/43 5.90 D
3/10/37 Ill. 02 II E 5 7/15/43 13.55 III E
6/17/37 16.58 A 8/2/43 6.45 A
8/14/37 16.58 B 9/20/43 6.45 B
8/27/37 2.29 1 11/30/43 13.55 IV C
8/31/37 2.29 2 1/11/44 3.57 1
9/13/37 6.73 3 4/25/44 4.96 2
9/30/37 7.45 4 7/10/44 5.86 A 3
10/19/37 21. 91 C 5 9/7/44 5.86 B
10/21/37 10.65 a 10/6/44 2.78 1
10/25/37 10.65 b 11/16/44 2.78 2
10/29/37 21. 91 1 c 3/6/45 6.91 3
11/23/37 19.91 2 3/26/45 6.91 4
1/15/38 19.91 3 6/26/45 6.00 C 5
2/4/38 13.43 4 7/27/45 6.00 D
2/23/38 13.43 D 5 12/10/45 4.84 V E
3/12/38 4.66 1
3/15/38 4.66 2
3/23/38 2.83 3
3/24/38 2.83 4
3/31/38 63.04 IIIE 5 Fig. A4-A
 (App.A4-p.4)

11 ELLIOT;r'S

Derived fro
I
---
hi'
WAY S
-------+-----+----/-t--+----+---t----t
charts nd lobul tions.

(See Bill iogrClph~)

3001-+-4--1-+0----+-- ----f--+-+-+f-- -- ---- - --+------+c-c:---:-Ic-------

hos been s te ndardized:
-- ---+----+--+---1
Deoree -- - - I ][ m III J[
est ----- A B CDE
----- 12345
v. ------abcdll
-- _. - -- ._-+----+----/-t--+-----I---+---I

200 f----fl----+"-

f-----+---Ifl-+---+-- - Y\+ ---+- ----- ------- -- ----",---- --+-+---j--+-----jf----:Hl

a C
- ---lltt-+--+--+-t-tt--.--n-- -------.----~- ---j- -+---+----f---tH-+I

f--+-.....,+-+---+--+- ~ • •
150 , ~I .,
.
II,ncludes
• c

AI
'
-
0
A
_N'
Ile.tensio~ a 'V D

r - -- j
! ---+---+---tttr • V'l
! ,
l
---+--tt+-'+--+t--+~-i-tt-~I+t------'~---r--+---1
I :J3[

a •
r - f - - - - + H I - - f - - t - i t -- II - e• • \1 N. I 1

~
I ' ;
I
, I I a
100 f------;--+ttt---+-ft-tt-I-U-ii- --- trion le

I • •
E

~
1Ir •
e
:l[

, "Five lesser
woves
I D

I
An "'n .rted Ziq 09"

70 --

,.
e A Flo'

• A8 DE d 9r••
a'. "Intern ,diot. I/ov ••
60

50 ,
•
, ••
A CD d 9 her. 0"

hQ! • Primar)' Mo ement·

40
1929 '30 '31
.•
c
'33 '34'3! '37 '38 '39 110 '41 '42 '43 '44
Mal

115
I x '38
FIG. A4-8
 (App. A4-p. 5)

ELLI OTT VARIATIONS - IN LINE OF TRENO-

Elliott found that waves in the

line of trend (1,3,5) sometimes
elongated into fi ve waves of the
same degree - rather than breaking
down into fi ve waves of a lower
Extension in Wave I, degree. This elongation could
upward trend. Occur in waves I, 3, or 5. but not
in waves Z or 4. It us wlly occurred
in the last or fifth wave. He called
these elongations extensions.

Extension in Wave I,
downward trend.

!I

;/! 2
Extension in Wave 1,
upward trend.

~
. Extension in Wave 3,
I 4 downward trend

3
5

;J!
Extension in Wave 5,
upward trend.

Extension in Wave 5,
downward trend. FIG. A4-C

4-14
 (App. A4-p. 6)

I ELLIOTT VARIATIONS - IN LINE OF TREND

9
7
8
3 6
I 4
/
2

2
4
3 6
6 8
7
(y)
9
y
FIG A4-0 !5
' II
e
3 a d
e
4 _ Ell ten.ion of Mlnuette Wove

- - - - - . Eaten.ion of Minute Wove

~..- - - - - - - - - _ . ElIten,ion at Minor WOVl!

,+------------------+~ Intl!rmediate Wave

FIG. A4-E

.. I, 1 '~
t.Extention ~seeond Retrocflllint
[First RI tr Deem.nt

Fig. A4-F

4-13
 (App. A4-p. 7)

Since extensions are comparable with the other

waves in a series, they may be equal to them and the
entire movement may seem to be nine waves of about
equal size. (Figure A4D)

At the end of a major move, there may be a series

of extensions of extensions in the fifth wave. The move
comes to an end finally on the last minuette wave of the
last minute wave of the last minor wave of the last in-
termediate wave. (Figure A4E)

Elliott found that extensions tend to be "double- re-

traced" i , e. by a downward and then by an upward move.
(Figure A4F) This is interesting only in an extension of
the fifth wave, since an extension in the first wave is re-
traced by the second and third waves and an extension in
the third wave is retraced by the fourth and fifth waves.

4-12
 (App. A4-p. 8)

I ELLIOTT VARIATIONS- OORREOTIONS

I ZIGZAGS I
Elliott recognized corrections of
five types: zigzag. flat, irregular,
b
MINOR triangular, and complex. Thev will
'y\c be illustrated by Figures A4-G
through A4- L
c

N
Three sizes of "Zigzags" are
MI NOR-
illustrated by Figure A4-G In the
b INVERTED minor size, each wave is single;
in the intermediate size the distri-
bution is the usual 5-3-5 found in
the "Basic Elliott" chart Figure 15- A
In the major size the two longer waves
INTE R M EDI AT E are doubled.

Note that the "inverted" versions

are included. These are corrections
of a downward trend.
INTERMEDIATE
INVERTED

MAJOR- A DOUBLE
ZIGZAG
c
C

c
c

MAJOR -

INVERTED

FIG. A4-G
 (App. A4:-p. 9)

I ELLIOTT VARIATIONS-CORRECTIONS

I FLATS I
B

~
MINOR In a zigzag the pattern is
5-3-5; in a flat it is 3-3-5.
C This gives the "flat" correction
a level appearance - and its
name.
MINOR-
INVERTED Examples of this correction
are in this Figure A4-H
B An example is marked in
Figure 15- A with a "zigzag"
also marked for comparison.

INTERMEDIATE -

c
A c
~
INTERMEDIATE-
INVERTED

MAJOR

5
C

C
5

MAJOR -
INVERTED

B
FIG. A4-H
4-10
 (App. A4-p. 10)

ELLIOTT VA R I A T ION 5 - COR R ECTI 0 N S I

IRREGULAR]

B "Irregular corrections" are distinguished

by the height of the "8" wave, which
)\I\T. IRREGULAR
advances to another top higher than the
orthodox top ("0. T~').

TOPS Since the correction makes a new top,

," A ~\ it is also called an "irregular top". An
I C example is the 1929 peak.
I

Normally "C" terminates below

B the bottom of "A". Liquidation
O.T.

~
c in the third or "C" wave is usually
5 a e more intensive than in the first
wave.
b I 4

3
6
C

Y\ c
MINOR

B
c

~
a 2
b I 4
13 3 INTERMEDIATE
A 5
C
B
c

MAJOR
(In the larger and mo r e important
corrections the "c" wave may
5 consist of three smaller five-wave
C sets. )

FI G. A'4:I
 (App. A4-p. 11)

[ ELLIOTT VARIATIONS - CORRECTIONS

TRIANGLES I
Triangular corrections have
five waves or legs each of which
has no more than three lesser
waves. In the small triangles
the legs can be single waves.

Triangles are found in the

fourth wave of a five wave
movement - exclusively. They
form the base for the fifth wave
or "thrust".

One of the boundaries of a

triangle may be horizontal. The
fifth leg may terminate within or
outside of the bounda rie s .

Triangles have been as short as

seven hours (a short example is
"'-
(I)I in (Fig A4-K) and as long as 13
--. ~I
:;'1
years. A. Hamilton Bolton
~I has outlined a 21. year triangle
• 1
in his analyses.
1

a
•
\
\
\
\
\
---~ - -- - -d - ---':;4-
\S
\~
\~

FI GURE A4-J

4-8
 (App. A4-p. 12)

- Dow Jones Industrials - Hourly Data-

40

"A small triangle a peared in th hourly record

in October 1937 (it signalled a immediate
acceleration and ext nsion of the downward
movement; the dyn mic Octobe 18-19 'panic'
followed." R.N. lliott 6/28/39

35 -
THURSDAY FRIDAY SAT.
10/14/37 10/15/37 10/16/37

FI GURE A4-K

30
 (App. A4-p. 13)

I ELLIOTT VARIATIONS - CORRECTIONS

I COM PLEX I
/

,,
/
This group includes some
final complex variations.
2 4 6 " In the words of Elliott: "The
/'VVV\" DOUBLE THREE rhythm of c o r r e c ti ve movements
is the most difficult feature of
,/ I 3 5 7
the wave principle. "

/
/
2 4 6 8 10 '
,,'VV\;/\!\./\.,,, TR I PLE THREE
" I 3 5 7 9 II

" 3 2 4 6

\,' / "~
: .
~
- - 1

DOUBLE THREES

I
,/ 'I ~
I
3 5 7
MIXED -UPWARD a
I 3 2 4 6
DOWNWARD

, 0t!
\

" \
'.
I
2 x
I

2
3

\
\
,
\
\
FIGURE A4~L

4-6
 (App. AS-p. 1)

Appendix A5:
CAN THE ELLIOTT THEORY BE MADE SPECIFIC?

Elliott had some problems in the application of his

principle. Many of his turning points were debatable;
his wave patterns tend to proliferate into variations.

The s election of turning points is crucial. You can

identify almost any pattern IF you select the proper
turning points.

A method of measuring turning points has been de-

veloped; it is described in Appendix A3. When Elliott's
turning points are put to the test, some will be found
to be rather light. No doubt he had his reasons, and
could give them to us if he were alive. He skipped over
some very important turns, and included some that are
rather puzzling. (Compare Figs. ASA and A5B with A4B)

Another problem is the division of waves into sub-

waves. You can divide a wave into almost any number
of subwaves, if you make the subwaves small enough.

We have tried to read Elliott's mind, from his lim-

ited writings, in order to develop a basis for a meas-
ured analysis:

Waves and Subwaves:

(1) In the breakdown of waves into subwaves, the

first and most promising line of attack is a study of
Elliott's own examples. These have been charted from
his writings in Figures A4A and A4B in the preceding
Appendix A4. The turning points that he used were
tabulated and evaluated, and the average wave ampli-
tude was measured.

4-4
 (App. A5-p. 2)

(2) A second line of attack is a study of the "basic

Elliott" chart Figure 15A in Chapter 15.
This chart outlines a complete cycle. If you will
count the number of waves of major degree, interme-
diate degree, and minor degree, you will get the fol-
lowing;
Major Degree: 8 waves
Intermediate Degree: 34 waves (4.25 times as many)
Minor Degree: 142 waves (4. 18 times as
many as in the intermediate degree.)

Using this as a guide, we counted down the rank of

turning point degrees:

Degree: Number of turning points:

80% 4
400/0 10
200/0 44
100/0 170
50/0 540

The frequencies in the Basic Elliott chart of 8, 34,

and 142 are similar in proportion to the actual count in
the selected groups of 10,44, and 170.

(3) A third possible approach is based on Elliott's

statements that "a wave is divided into waves of the next
smaller degree." If waves (or turning points) are
ranked in any given period of time, from the highest de-
gree to the lowest, and the rank is scanned downward,
you will find at a certain point that the waves are parts
of preceding waves. This is evidence that you are moving
into a subwave classification.

For example, the ten most important turning points

in the period 1928-1962 are the following:
(1) 851. 920/0 9/3/29
(2) 851.920/0 7/8/32
(3) 111. 020/0 3/10/37
(4) 111. 02% 4/28/42
(5) 63.830/0 9/8/32
(6) 63.830/0 2/27/33
(7) 63.040/0 3/31/38
(8) 63.04% 11/10/38
(9) 52.160/0 11/13/29
(10) 52. 160/0 4/16/30
 (App. A5-p. 3)

You will find that numbers (7) and (8) are, in time,
between (3) and (4), and evidently form a subwave. No.
(9) and (10) are between (1) and (2) and evidently form a
subwave. A similar method was used to check the other
groupings.

The three approaches outlined above were combined

and cross-checked to yield the following proposed class-
ification:
Degree: Nomenclature:
Cycle 800/0 I, II, Ill, IV.

Primary 40% A, B, C, D .....

Intermediate 20% I, 2, 3, 4 .......

Minor 10% a, b, b, c, d ....

Minute 5%

Figures A5A and A5B chart all of the 10% turning

points from 1928 to date. The waves and subwaves are
marked with the nomenclature in the above table. The
same classification formed the foundation for the tables
in Chapter 15 on page 15-4.

4-2
 (App. A5-p. 4)

.001---r-----r---~----i----t~~
x
-TURNING POINTS== 10% DIGRItIt Ik
1---
3/10/37

--1--- - - --,--+--,·-+1+1
2

~ ---- A

V
-- --
7 9

50 r-------- f--.---- f-.---------t..-----4-1 I--

3
- --
B
Fig.A5A
 I I (App. AS -p.S I
90 0 I
I
TURNING pOllns 10% DIGal1 a: BIGBIR
I I
80 0

I
(10%
I
CLASS
,
WAVES)
, r
IIII A
9

Dow Jo nes IndustrIals· 3-10-37 10 date -

.--- 1
70 01-----

60 0
'\~ A

8 ~
500
A~ I

l
40 0 -- 6 ----

m
A B C A
I 2 I 234 3
5 12 34 5 6 1 B9 I
abcobcobcdefgtl iabcdeobcabc abcabcde c e e e e abc ef90bcdeabcabc obe
300

250 ~----- .-.

200
------ 1\ -- . __ f -
.
~
--
V '~ A -- --- ,-
II -- -- _._--
.8. _ _ 2 _ --
2
'L - --

150 -- - - - '-- _. - - --- - - --

2
\ti 4
- -f--- - - _._- -- ----
A~A
1
\
h\r-:- 4 -
_. - ._- ._--- --
VV II
! I
-- -- -
I V
---- -_.
A- f----- f-f-- - - f - f---.- -- -_. ._--

5-1
100 --_.-

90 1----.
1937
---
--
3
A

--
'38
- --

'39
1IT
40'41'4243 '46
I I, I
m

'41 4849' 5053'55 '56 '57 '59 'CO '61 -62

II I
FIC A50
""
 --
=. DISTRIBUTION or SUE --
PRIMARY MARIlETS PRIMARY SWINGS SECONDARY REACTIONS
PERCENT
AMPLITUDE
ABOVE 180 Iii 340Y.
160 II
140
12
I 110% II
100 II II
90 ., I
60
-MEDIAN I
70 - Middle - -Middle
1111 50% eo% 111I 62-4
60 --
of cases of casee 'II I
50
45 ttI+ I I i
I 'tl
40 I 40%
11+1- I
35 r - - - 3 5 % >
0-
I III I
30
III 27 'II.
-
ntt _ Ttt+ - Middle - r-Middle
III
I
'tl
25 -Medion- -50% - - eo% 26%
_tttt I
of case, ot cou. III
20 I
16
1114-
ill
nit
11I1 17%
-...
16 tilt 1111
1tI+ III
_, 15% 11I+_ _Mlddl. _
14 Middle -
12 _M.dlon - 50% - - ·80%-
"rt+I-1II1 1M 11I1
10.6% • of co ... _ _ot ca •••
10 1111
II
9 111+-1 8.4 -14
8 1111
III
7
111I
6 6%
II
5
II
4.5
4.0
III
3.5

Fig. A6A
 (App. A6-p. 2)

Appendix A6:
HOW SHOULD SWINGS BE CLASSIFIED?

The Dow definitions in Chapter 14 are not specific

enough for our purposes. They require "interpretation".

We are fortunate in this analysis. for one of the

most authoritative of the Dow Theory experts, Robert
Rhea. has set down dates and classifications. These are
tabulated in Appendix A9. We have summarized his
turning points and swings in Figure A6A. A tick mark
method of scoring is used.

There is quite an overlap between the size of the

three types of moves. However. if you will concentrate
on the middle 500/0 of the cases, you will find very little
overlap.

Consider the primary markets. There are seven

swings with a magnitude below 500/0 which have been
classified as bull and bear markets. Some of these have
been qualified because of their duration. However, when
we consider duration alone, Robert Rhea clas sified fi ve
swings as "major" which lasted less than a year. If we
balance duration against swing. we have picked up all
but two of Rhea's major swings with the definition "All
swings greater than 40% or more than one year in dura-
tion. "

The great primary move in a bull market is con-

sidered to be broken up into "primary swings" in the
upward direction. which are interrupted by "secondary
reactions"in the downward direction. In a bear market.
the primary swings are downward and the secondary
reactions are upward. Still smaller moves are called
"minor movements". In the preceding paragraph we
have set up the qualifications for the largest classifica-
tion. What is the qualification for the next sized move?

5-3
 (App. A6-p. 3)

= SUMMARY= PERCENT SWING·

• Middle 50% of 011 coses ond Medions (M) -
(High -Low)
Percen1 :. .100%
(Low)

-15 Bull Markets

150
140

15 Bear

100 1
9 6 (M)
Markets
'--v---'
90

1 t
80
70
67(M)
60

I
60

50
P RI MARY SWINGS
4

35 In Bull
In Bear SECONDARY
Markets
3 ~
Markets REACTIONS
~
26
24
In Bull In Bear
20
1
1 6(M)
f
17(M)
Markets
'--v---'
Markets
~--'
17

t
15

t 14

t
I
9
II II
19(M}
II(M}

8
7

6 DATA-1897-1964 (JUNE)
! 1
G 7

Fig. A6B
 (App. A6-p. 4)

The Dow Theorists set up the general description for

the secondary reaction to be: Duration three weeks to
three months; Amplitude sufficient to retrace one-third
to two-thirds of the preceding Primary Swing.

However, this definition first requires a definition

of the Primary Swing.

We can kill two birds at once by specifying the lower

limit of a secondary reaction. The primary swings are
the remaining pieces in the bull and bear markets.

To this end, we have measured the "secondary reac-

tions" tabulated by Robert Rhea, and have set up the fol-
lowing definition for our analysis: "A secondary reaction
is any reaction exceeding 50/0 in amplitude and more than
one week in duration."

Robert Rhea included five secondaries which were

less than 5% in amplitude; however, he did not include
55 swings which exceeded 5%. Our definition, there-
fore, includes almost all of Robert Rhea's "secondaries";
it also includes a sizable number of swings that did not
qualify under hi s definition. (Some of these, no doubt,
were not qualified because of a lack of confirmation by
the Rails. )

With these specific definitions, we are able to pro-

ceed with the work of counting and measuring. All of the
movements are dated and charted in Appendix AI0. In
Figures A6B through A6E we have summarized the size
of the movements.

For an example of the use of these charts, consider

the 1962 slide. It was a swing of 40.03% and lasted 222
days. The amplitude qualifies it as a bear market under
our definition. In Figure A6B you will see that it is
within the middle 50% of the swings of the 15 bear mar-
kets. The swing is far above that of the typical second-
ary reaction.

5-5
 (App. Ab-p. 5)

= SUM MAllY = DUIlATION =

- .,. Mid d Ie 50°4 0 f a II cas e san d M ed ian s (M 1

- - D urafion = Calendar Days --

DAYS 15 Bull Markets

J

1500 1480
15 Bea r
Markets
1000
800
1 800
794(M)
•
j
GGO(M)
600
500
400
365 3G5
1
300
PRIMARY SWINGS

200 In Bull
Markets SECONDARY
150 ~ In Bear REACTIONS
123 Mar kets
100

80
t
77(M)
In Bull
Mar kets
In Bear
Markets
68
60
50

40
!
43
t
44(M)
~---

49
62

t
! 1
38(M)

30

20
26 25(M)

l t
18
15
DATA - JAN 1897-JUNE 1964

10

Fig. AbC
 (App. A6-p. 6)

In Figure A6C, the duration is found to be short for

a bear market and long for a typical secondary. On the
basis of duration, it is in a gray area. The amplitude
however, is sufficient to qualify it under our definition.

In Figure A6D, the range of "legs" is tabulated. It's

interesting to compare this table with Elliott's wave
theory. His basic bull market had three primary swings,
plus two secondaries. The average in Figure A6D is
seven primary swings, interrupted by six secondaries.
In the case of a bear market, Elliott specified a typical
two primary swings and one secondary. The record
shows a typical fi ve primary swings (interrupted by four
secondaries.) Elliott's count seems to have under-
stated the situation.

In Figure A6E. the "per cent retracement" by second-

ary reactions is charted. Hamilton set the size of a
secondary as "40 to 60% of the preceding primary".
Rhea set the size at "one-third to two-thirds of the pre-
ceding primary". Our chart suggests the r ang er "usually
one-third to seven-eighths of the preceding primary
swing. "

5-7
 (App. A6-p. 7)

2 SUMMARY NUIIBER 0' LEGS =

15 Bull 15 Bear
Markets: Mar kets:

9 9
Middle 50%
of all { ••d;o.- 7 5
cases ---
5 5

Fig. A6D

= SUM MARY = PI RC I NT IlITRACIIiINT =

(Secondary Reaction Retracement, in Percent Of preceding

Preceding Primary Swing)

In Bull In Bear
Markets Markets

80.67. 86.3%
Middle 50%
of all Median -+ 56.1 'Yo 59.0%
cases -- {
37.1 % 39.5%

Fig. A6E
5-9
 (App. A7-p. 1)

MONTHLY iCH NGED . DC W ON S I [DUST I ,1 S

+--
= 921· OC
f----- II" .
800
. ,
700 "
, ... d, •
e
,

.. · · ·. .
e , ,
600
· ·
0

500 · ·, ,
:
·
D,' 00 ~
· 0 0..
0

e s
400

d' °
~

300 ,
0 0 ,
°
r--'°
8 i

. 0
J'
· 0
0 -- ,0
00 0
. 00
0

.. ·.
0

0 ;p
.
0

: o. 0

. ° /
·
200 · ·
·, · , , , , ,
,
(
,
,
e
. ., , ,
· . . ·· · · ..
" · ...
- ... '. ..
e
., f ·
.
~-
~

. . . . · - .a °· ····. ., , ·
g

.
e

.
0

OD ~
~

· , , ..· ". -' '.., ·.

% 0
..:l 150 e "
.·, ·
e
<
.
lIlI ·
. ., · . · ·. .- -· . ..
"
'.
.
'J'
e

. . ;c - . - · ·
OD
0
I:>
~-
Cl
lI:
·
·· ·. , ·. ·- ·•. .'
100

.'. . · . . :.. . . ·I~

OD
Ilil
lI:
90 · . · · · · '"
"f

e
.. 80 · . . . ··
~
e
70 ·· .- '~
EDIAN
Cl
60
Leost squor !l. r qre &io ( 0
- 11.180 -0. 89 (La 001-0
(of '" AIRp' Iud I

or: ['t) 0
15.1 (0
· --
1° 71 08 .;. OJI)
.
50
Not.: To make
•••eere camp r I
- °
to incr as" 100 I t 0

. co. ., o
d erta •• i , , 0 ho
· 'ho ond
JI 01
- °
m ",

40

cI
\. 15 20 30
MAl

40 50
02 0.3 0.4 0.50.6 0.8 I 2 3 4 5 8 78 91D

PIICINT AMPLITUDS =

Fig.A7A
 (App. A7-p. 2)

Appendix A7:
HOW DO SWINGS VARY WITH PRICE?

Stocks with low prices seem to move up and down,

percentagewise, more rapidly than high priced stocks.
Some observers have concluded that the variations tend
to change with the square root of the price.

In Figure A7A, the monthly percentage changes of

the D-J Industrials, for each month in the last 43 years,
have been charted against the D-J average.

The percent changes tend to be higher at the lower

price levels, as expected. A median line (dashed)
traces this tendency. The relationship isn't close,
however, and the data are scattered widely.

A least- squares regression line (the solid line) casts

some doubt on the square root relationship, when applied
to a stock average. This calculation results in a 0.7108
root rather than the O. 5 root.

5-11
 (App. A8-p. 1)

Appendix A8:
THE DOW JONES AVERAGES - WEEKLY BAR CHART -
1928-1964

The charts which follow present some interesting

history. The swings can be compared with the text of
this book.

The scales are logarithmic, so that equal percentage

changes on all of the charts are comparable.

If you would like to compare the current situation

with preceding history. you can trace the last chart on
a thin piece of paper. This can be transferred to the
earlier charts for a definite visual comparison.

5-13
 (App. AS-p. 2.)

400 --- -

. . ~.IO

350
150

H, 140 I--

300
I
I'~ 130

120
- -

250 110

IMlljll~
I--- I-

100

r
Wi J
90

200
w;
~ I.
190

180
-(MIG
OAl ,.
• a 0W'9
CL 9£ 0
f

LV
M f-
80

170

MAl
160 'V 70
1921 1929 IU. 1931
L-
 (App. A8-p. 3)

100
110 u

95
105 f---- --
-

~~
90
100 --

85 --- ..
95
i

80
IftIIIiIP~
90 - -
'lIM' ~u

85-1-- - - - - - --'-----
75 1---- e-- ._-

70
80 f----+-Il--+ - -- -
-r-
-
-
-
75t------b1l--+ - - -- f- --~
65 - - - e-
I
70f--- -- - -- -
I
1--
60

6 5 fj---++--+--~-+ - - j ----j--- -

55

60 1-9t-ttt~--+--t---+--+--t--t--1
-

50 ---

5 5 HII-*--+--~-+--+--+-----

45 - _ . -f---
50 r-++- --+--t---+--+--t--+-

40.158
MAl FIlAr
.0 4 5 I-----'--..L.---J'---t--..L.-'---'------i
5-14 19 32 1933 1934
 (App. A8-p. 4)

..---.----,-----,-----,----,---,--- ---,--- ---'j'" --'_.l_-i----l

I---+----t--
-.. --- -J-+--t---i-----t----r- - -f---t-
I I
- -- i - - --
I

OOI--+--r---:-I --- i .-+- ---- r--- -- - i j~

-J-- l-r- -1---1-- --- t--- --- --.

90~_+_+_---,-I-l
i ! i
I '. . ,I _ jl ,
r--+-----------
_ -J-_ I~ _
I

BO'\---+----r-HT ,~ I _ r - _ _
:: -t-- .___
e--
+-_I- __~-+
H
1--+---;-.--j-' ~ . i - . I I

'oc--- ~t±t~11-I _ _ .... ~... . .--_-jj

I t- I! i ! I 1
100

I, Ii, ,-- ---

. ~+l-
1_ . - - ---I

,,=e~=4 -- -' I 1>0

"-I~ I~
-

.0 - - -

c---
i

i-I
"
+
.1 +-,- 1. - - - , - - - - -

j- ~e
-- . --~'1 I ~~I- b
~ I
4e
\

,O~ Jl-f +.. ~ - 1- I.~ I 1>°1

- 11- r i I I I~ I
wc--- ..- c---t Il .~ ~ -
- ']
f- -- - . - - -. - 1----1 ir-- --- -- - .~ i

10 l---t,iIII-+t- -- -
I
I--
;
1-- ---f-- -----
II
~
i

!.-.- - _. - _.- - -,. -- -_.'. --- I

I -:
)0 ~ I---t--t- - r- 1-- t---j-- 1--+----1 . - - -' --
I
\100'
!

---f---t-- +-- 1-- - t-- t--- - - - 1-- I

I
I

IOr--'--ll--t-- .-.-.- --- --- .--

1935 1936 1937 1938
I[

1939
;MAI
I 90
 (App. AS-p. 5)

lf~+t- ---r= ~~-=-- -- -~~~

::W-t iji-'_~=~c=-=-~:
t - -~
,1-, f l l -~
80
f
--t +.-\'--1------
\--~+--II ..
. .-. ··1- -"'-- ---...
--f-- -
-- - ----- .
- - - - - - - --
-f--- -----

_....
IOU

+--+--
170 -1--- -- t- - ---- 1--- .----- --r-- --r---+7&

, -I - - I -,-- -- - - -_. - --ti--_I-- - -!

160 ,---1-- --- I -/ i - --I-- ----------+--+- 'uu I
---t- +--+---+- -+---+---i

,: f+~r-f=~~- - JJ

I 1-- - I -----I. ~". .lli ----:!

'T
- I '
-~ I
-I
--IN - '1-+ --
i ; ( , i --
j- -- --
'I

,+-- --t L_~ -I ~wt: i ~

l+ I-J- - ~-- P-l '1 i

J_L!-lfitt;-tJ : !,,,
5-12
194. lUI 1942 r 19f3 19U
!
 (App. A8-p. 6)

1---+---l------1----1.- -.--+--+--+--+--+--+--+--+--+--+---1---+--+--+---1

-----f- .-- .-.f-.- ..-1-- _ . _ . __ -- . - __ l . - -. 1--- .- - ---

1---+-_+-.. 1---. - - - . - - l.- I- -1--.-1----- . - - - .- - 1--.--. -e·-

-- ._- - .. _. 1----+-+---1---1 - f - - - f--.I----

-_ ... ~.

r1r
"q
1801---I--·-l---- , - -

II

~.I
~IJ~ I---.J---l---- -1----+--" - --.- . - .--+----1---+".--+..- - - 1 - . - . 1---1--.-
160 ]

~'-_I____I---I--_I____I-__I_-_I_--I-__I_-----l-+-I__+-+_-+-_+-+_-+__+-__I
150 r,

1---+----1---.- --- . - . ---1----1----1----1- - - ' - -.I---!--.--I--+--+--+---I------1

1401---+----1----1--·- 1...----1---. 1------l----+-----1I----.J---1------1~-_J_----I-+-+-_+--+-+-____I

I---+----I---I---+----I---I--_J_----I---I---I--I--- 1---. \--- --1--+----1---+---1------1----1

1~ 0 I---+---l----__I---l-----I--.I----<- ----1--.-1------1- ----I---I------If.-- - - I--- _J_---I---I- ----4---1-----4

1-----+--1--1_---+--1------4--- - - 1--- - - - -.\-- - ! - - ~- - . + - -1-- --1--.!--

120 I--_J_---l---I---_J_---+------Ii------l-- ---l------II------I----l------II---_J_---l---I--- +---l---+---+---1

I---+----l---I---+-- .- -1-----1--+-1_---1--+-1------1-·-+-1 - - - - f-- e---I--

1I01---+----l---I----1----I----I----1-----'I--·-I-- - - -- ---- - - - -1--

'MAl
1-----'---- - - - - --'---'-----'--.--1----1-- - - - . , - - - - - .---'---
lU5 1'46
I'" I,.. 19f'
 (App. A8-p. 7)

-----~

.. _. -- .. _. f--

-_. --

400 I - - - 1- .. ---- _.-- . -1--- - 1-. ... -- -_.- -_._.-- -

~Oi
_.. ........ - - --- -- - _. f---
~
L_ _._ ... _.. -1-- /---- 1---+-.- .- I - - ---- -
~

r 1--

- - - - --_.
_... --

.".
- _.- .. _.- - -I-
.
ilso
~.
--- .. __.. --- - - _._,..- -- - ...
350 ~

1---. _.. ---- - 1--- - - - 1--- -- -- -

-- - _. .. _-,.- _.-
-~
. -- I---- 1--- I--- ---

I--- .. .- -_.- ----. -- --_. - -

-
I~'

+
1--. ---_.- ._- --- - - - -- - -
p'
300 -.' J! 300-

tI'
'p

1---1--- .. -- -
j IV
r"- .- .. - - - - 1- 1 I~ ~ ~~ ~~ '-. f - ---
,~ IIIJ~ rr~~ 1r~- - - _.
1~

~_L- ... .- '-

.~ --- I--- ._-
p ~'i 'r
.~
r-~t
250 _ .. .- -_.. _.- f- _.. .. - I-- 250

rr- --1---
"---- -_. -
I
_. -- f--. -- -_. .. _..-
I---- .- 1-- -- -

~ 1--- ._-- I - - - __ I . -
·'0- I-- --

L I
~ f -.- I--- ._+-- - --- /---- +-- _. -- f - - 1--- I-- . -l - - -

-- t-f-· '.- 1--- .... - f - - - _.

-
--.~

200 ._- - - .- 1---.- I--- ..- ' - - .. - ---- _.

~-~- - 1---.. .- .. - - - I- - - /-- --

19 0 ---1--- . I--- - - - -.. -- --- _ ..

- _. _. . ~- --- I- --
MAl
18 0 1-- .. .. ._ .. _- f - - - . -- ' - - - -
1951 1951 1952 1953 1954

5-10
 (App. A8-p. 8)

800 r----. --'---'

1'1 0 0-

700 -- I- -00
I

~~ ~
I,
A~"
-
I
600 1",00

.- -- - - . -1-.- I-- ~
,~
•

500
J

,j~
, b.

;
ld

- d
~I
j~1
I-- -. f - -

500

~
.. _-f - -
~ ~ .~ ~d,
J m ,1Jl~ ~, .~
~
,~ ~I'
.J
~II IJI ~ ~W
450 --450
lr~, lu
gn --

IlII' IU I

400
N'W .- - . -- - -- - - _. f-- --- -_.- _. -_ .. . _- - 400
- _. _._. -_.. 1-- .. .- f-- -- -- - - ..- - f-- .-

-- - 1---. 1--.. . - -- _. - - _ .. -- f - - -

-_. L - __
1--- .. .. .. --- ~ - -- I--.
- - -
-' - ' - f---. -- .. - -- . ---

350 ...- -- --
._--- '--

MAl
- '---- .- .L- ~.~-~ .. - -- -_.
Ins 1956 1917 1911 1919
 (App. AS-p. 9)

800 ,Jr-
1----1tI._+--+-I----:
I ~~'

J-1I!I4A"'--.J-.llI'.!i--l---.,II----J..--+- -+-- +----1- --II- I- e-

MI ~
I"
1
I---+-----.jl---!jlrlll-~-- I--- - - I-- - - I-

1960 1961

5-8
 (App. A9-p. 1)

Appendix A 9:
DOW THEORY DATES AND DATA -

The data on the following two pages are the dates and
prices for the primary and secondary movements of the mar-
ket as reported by Robert Rhea, one of the most authoritative
interpreters of the Dow Theory.

The sources include his books and his market letters.

These data are charted in Figures 14B and 14C in Chapter

14.
 [App, A9-p. 2)

DOW THEORY - DATES OF PRIMARY AND SECONDARY

MOVEMENTS - from Robert Rhea -
(Prices are D-J Industrial Closing Prices. )

-.-8/10/96 29.64 ~11/19/09100.53

I 11/9/96 44.08 (1) 2/8/10 85.03
gt 4/19/97 38.49 ~ 3/8/10 94.56

T-
::: 9/10/97 55.82 7 / 26 /1 0 73.62
I 3/25/98 42.00 8/17/10 81. 41
8 / 26 / 98 60.97 9/6/10 78.35
10/19/98 51. 56 10/18/10 86.02

+- 4/4/99
5/31/99
9/5/99
76.04
67.51
77.61
~
S
12/6/10 79.68
2/4/11
4/22/11
86.02
81.32

+'"
~ 12/18/99 58.27 6/19/11 87.06
~ 2/5/00 68.36 9/25/11 72.94
5 / 1 5/ 0 0 56.62 9/30/12. 94.15

+-
~
_
6/1/00
6/23/00
7/23/00
9/24/00
6/17/01
59.38
53.68
59.02
52.96
78.26
~
~
12/11/12 85.25
1/9/13
3/20/13
4/4/13
6/11/13
88.57
78.25
83.19
72.11

+ +-
12/ 14/01 61. 52 9/13/13 83.43
9/19/02 67.77 12/24/14 53.17 [a]
12/15/02 59.57 1/23/15 58.52
~ 2/16/03 67.70 2/24/1554.22
~ 8/8/03 47.38 ~ 4/30/15 71.78

t: +
8 / 17 / 0 3 53.88 ::: 5/14/15 60.38
11/9/03 42.15 12 / 27 /1 5 99.21
1/27/04 50.50 4/22/16 84.96
~ 3/12/04 46.41 11/21/16110.15
~ 12/5/04 73.23 2/2/17 87.01
12/12/04 65.77 ~ 3/20/17 98.20
4/14/05 83.75 ~

+
11/8/17 68.58
5/22/05 71. 37

t
l l / 2 3 /1 7 74.23
1/19/06 103.00 12/19/17 65.95
3/5/06 92.90 2/19/18 82.08
4/3/06 98.19 4/11/18 75.58
~ 7/13/06 85.18 . 5/15/18 84.04
~ 10/9/06 96.75 ~ 6/1/18 77.93
§

t
~~~ ~~~7 ~;: ~~
10/18/18 89.07

1
2 / 8 / 19 79.15
11 /15/07 53.00 7/14/19 112. Z3
1/14/08 65.84 8/20/19 98.46
2/13/08 58.62 11/3/19 119.62
5/18/08 75.12
gt 6/23/08 71.70 (Note: [a) This is
::: 8/10/08 85.40 a new series, to

1
9 / 2 2 / 08 77.07 be comparable.
11/13/08 88.38 multiply preceding
2/23/09 79.91 figures by 0.7339 )
11/19/09 100.53

Fig. A9A
5-6
 (App. A9-p. 3)

DOW THEORY DATES (cont'd)

T 1l / 3/
12/22/19
119.62
103.55
9/ 3129 381. 17

T
10/4/29 325. 17
1/3/20., 109.88 10/10/29 352.86
2/25/20 89.98 11/13/29 198.29
tl:l 4/8/20 105.65 4/17/30 294.07
III 5/19/20 87.36 6/24/30 211.84
.,
~
7/8/20 94.51 tl:l 9/10/30 245.09
(1l
8/10/20 83.20 .,
~ 12/16/30 157.51
9/17/20 89.95 2/24/31 194.36
12/21/20 66.75 6/2/31 121. 70
5/5/21 80.03 6/27/31 156.93
8/24/21 63.90 10/5/31 86.48
12/15/21 81. 50 11/9/31 116.79
tl:l 1/10/22 78.59 1/5/32
-
1=
5/29/22
.4--6/12/22
96.41
90. 73
3/8/32
7/8/32
71.24
88. 78
41. 22
10/14/22 103.43 9/7/3Z 79.93
tl:l 11/27/22 92.03 2/27/33 50.16

t
III tl:l
.,
~ 3/20/23 105.38
e. 7/18/33 108.67
7/31/23 86.91 - 10/ZI/33 83.64
8/29/23 93.70 161.99
4/6/3.
10/27/23 85. 76 4/29/36 143.65
2/6/24 101. 31 3/10/37 194.40

+
5/20{24 88. 33 6/14{37 165. 51
8{20/24 105. 57 ~ 8/14/37 190.02
10/14/24 99.18 .,
~
11{24/37 113.64
3/6{25 125.68 134.35
1/11/38
3/30/25 115.00 3{31/38 98.95
2/11{26 162.31 4/16/38 121.00

-
tl:l 3/30/26 135.20 5/31/38 107.74
-
1= tl:l
8/14/26 166.64 1= 8{6/38 145.67
10{19{26 145.66 -1--9/26/38 129.91
10/3/27 199.78 11/12/38 158.41
10{22/27 179.78 til 1/26/39 136.42
(1l

5/14/28 220.88 .,
~
3/10/39 152.Z8
6/18/Z8
11/Z8/28
201. 96
295.62
1tl:l
4 / 8 / 39 121. 44
6/10/39 140. 14
lZ{8{28
Z/5/29
5/27/29
257. 33
322.06
293.42
-e. 6/29/39 130.05
7/22/39 144.71
8/24/39 131. 33
9/3/29 381. 17

Fig. A9B
 (App. AIO-p. 1)

Appendix AIO:
DATA AND CHARTS - BULL AND BEAR MARKETS -
PRIMARY SWINGS AND SECONDARY REACTIONS -
1897-1963

The charts which follow outline all the bull and bear
markets since the beginning of the Dow-Jones Industrial
average in 1897. The Primary Swings and Secondary
Reactions are included.

The swings are based on these definitions, which

were developed in Appendix A6:

(1) The great primary Bull and Bear Markets include

all swings of the D-J Industrial average with an ampli-
tude of more than 40% or a duration of one year or
more.

(2) The Primary Swings or Legs are smaller swings

in the direction of the primary market, which are term-
inated by Secondary Reactions.

(3) Secondary Reactions are all contrary swings within

the bull and bear markets which exceed 5% in amplitude
and one week in duration.

The size of a bull or bear market can be measured

by three yardsticks (These are tabulated on each chart):

First: How far did it swing? This can be measured

as a percentage. The charts have been put in this order.
The biggest swing is first; the shortest is last.

Second: What was the duration? How many days?

In these charts the duration is calendar days. (See Ap-
pendix A13. )

Third: How many Primary Swings or Legs? In

these charts, the primary swings or "legs" are solid
lines; the secondary reactions are dashed. If a move
in the primary direction is part of a secondary, it is
dashed. The horizontal scale is only an approximate
expression of time. The curve moves forward one space
for each price reversal. The vertical scale is logarith-
mic, so that all of the charts are visually comparable.

The current market is charted on Figure AIOH.

Space is included to permit up-dating. A suggestion:
Trace this chart on thin paper and lay it over the other
charts for comparison.
5-4
 (App. AIO-p. Z)
Bull Markets:
400 - - - - - - -

.. (I)

8/2411921 - 9/3/1929

2932 Day.

J, J', ~,
496.51 %
21. Primary Leg.
~J'J;
I G ,

~I
,
I

250t---------------- -_ ... -

200t------------ -------

150 I - - - - - - - - - - - t '.....II ,.+iic--

1:f---7~~ \1~'
~ -- ---------- - -

,
I ,

J,
,
III ~I
I"
! I X
,
I
, .1 I -----
~~ !
..
!4
o
- - - - - - - - - . - - - - .. _ - - - - _ .. - - - - - - - - -
80
'"
!
..
o
IE

"
701-+-------- -------- - - - - - - - - - - 1

60L-------------------------!

Fig. AIOA
 (App. AIO-p. 3)
Bull Markets:

Bu II - {2l
212711933- 3/6/1937

1468 Days
287.70%
13. Primary Legs.
200

150 ---

r," I1
100
t

•• 8 "~
~
'i '.
''.''
,
II,
I

, "II
I 1\
90
I I
"II
" •
W:,,
~
, "I:
• R.
I'"
-. 8
I
I
I
I

'
&

J
V~
80

70

~
60 I
I

5o IA

Fig. AIOB
5-2
 (App. AIO-p. 4)
Bull Markets:

Bu II - (3)

6114/1949 - 4/9/1956

2491 Days
224.13%
13 Primary legs.

600 ~
Bull (4)
11/911903 -1119/06

802 Days
5001---------------+--1 144.37%
7. Primary Swings
110

100
•
400 - - - - - - - - - . -

...
0: 90

~ _.
C

I - - - _ .__ ._._-+---~--
'"
Z I
I
il:
o __._ .. ~ ,
.... \_.-
z 80 r -~
I
2

'"
!!!
'"
[
300 I
701 II .
!

250

200

150

Fig. xroc
 (App. AlO-p. 5)
Bull Markets:

Bull (6)
Bull (5) 7/30114-11/21/16
4/2 B/42 - 5/29/46 845 Day,
1492 Day, 110.13%
128.69% 9 Pri. leg,
7 Primary Swing, 120
220
~ 0
.;
K I
200 ~'"
:I
•
1\1
~
/'.
0 -; 100
l- S 18
.:
.
>
w , ~,~
t I 1~I
0
90
v~
150
7'
~, ,
BO

/", ~
,
I

"~
I

70
.
II:
;0
0
l
•
:t:
r;0 ,,
I

.
:II! I
I

60
..
0
w
~
fOyjl
100

I
50

90 ,-

Fig. AlOD

6-1
 (App. AIO-p. 6)
Bull Markets:

Bull (7)
Bu II (8)
4/19/1897 - 9/5/IB99 Bull (9)
7/8/32 -9/8/32
62. Day s 11/15/07-11119/09
869. Days
IDI.64% 96.64% 735 days
G. Primary Swings I. Pri. Swing 7_ Primary Swings

80 89_68~
90

\IJ
110

0
70 80 100
~ "g f.j
0-
:>
90
"J
z
J,
70
,
;0
Q
60
.."',. ,
I 80
\ ~
•
<i

I
'"
0-
I

I i
on 60
• 0"
50
~ •• 5 70

11 'I"'

~ ~
60 "A
40 V

35·---- 50

Fig. AlOE
 (App. AIO-p. 7)
Bull Markets:

Su [I (10) 8 u II (" )
12119117-11/3119 10/22/57-11/15/61

684 doys 1485 doys

6. Prj mory Swi ngs 8 Prlmory Sw jngs
81.38% 75.68"
130~------1

800

ifJ
700
'J e
100 ------+-J--l---j
Jj
",.
, II. "~
1\
,\

I' ,r V ~
90 1-------+-------1 600 1/1
W' \~
II
I',
500

;.l
60 L - - - - - - - - 400 l - . - - - .

Fig. AIOF

6-3
 (App. AIO-p. 8)
Bull Markets:

Bull (12)
3/31/3B-1I110/3B

224 Days Bull (14)

Bull (13) Bull (15)
GO.42%
11113/29-4117130 9/24/00-6/17101 9/25/11- 9/30/12
3 Primary legs
155 Days 266 Days 371 Days
160
4B.00% 47.77'" 29.08 %
2. Pri. 'egs 5. Prl. 'ells 2 Pri. legs

140
~ 300
80

\ l\
.' I
~ I

I 100
701---1Yt+.LJ-l
~
120
\A ...
01
:.
I
I
90

60 1-+-----1

100 80 .f'
90 50'----------l 70

Fig. AIOG
 (App. AIO-p. 9)
Bull Markets:

The Currenl Markel--

Begins: 6/25/62, 2. PM
DJI=52669
1200

1100

1000

900

800

700
~.!~..
..
I!:c

."..
"
600 1-1.- I--I .
z
z
:I z

-J
500
, , ,,

Fig. AIOH

6-5
 (App. AIO-p. 10)
Bear Markets:
3-00

250
\ r
813.
(I)

4/17/1930 -7/811932

Doys
%
" ,
II I 616.7Z

~'"~
II. Primory Legs

200

I
,

\A.
aI
," e",!
:
I
I

.-,1 ' 200

0 1,
...
..'"
:>

\
10
~::i
~ l
.,,,.
150
k' J",
i .
I
I
r 90
~
I I

:y
I
.,
...
.
0
80
:\.~
150
!
'"..
i~

:\'.k!.\ :\..
I
l-
i l,~
70
I I 'J ~

I
,
I I : I
.,
I
I

100
I ; ~
I
I
60
I
I
90

1 I
100
.1 : 50
80
",
"
til
I
90
(2) (3)
Y
70 3/6/37 - 3/31/38 1119/06- 11/15/07
390. Dovs 665. Doys
9G.93'r. 94.34%
S. Prlmory Legs 7. Prlmorv Lege
6-0

50
~
Fig. AIOI
,
,
I

40
 (App. AIO-p. 11)
Bear Markets:

400
8
., ~
~

~
~ ~
350
70
\1\: i
D ~

300
60
f\Vv
~b
u
"
,

250 .
,l

•
50

200
- !\
601------------+ 40
(5) (6)

11/3119- 8124/21 6117101· 1119/03

(4) 660 Days 875 Days
9/3/29-11/13/29 87.20% 85.67%
71.Days 9. Primary Legs 9. Primary Legs
91.84%
I. Primary LegS

IE
;
160
..~- 110

150
\~! 9 ;\\:\,
:E

..... -
.J

.,
\1 D ~ I 1001-1-. .- -

r. r \ f\
14 0
~
:,,"~ \.~ .
1\ ,I ;;:
13 0

,
",:l
I
I S 90 f-lL--\+--"---t--t---+

~ I~ :~
..... ,
I
120 ~
II ~
"z I
Oz 80 f - - - - - - ' - - t - -
z.,
;;;10
0:
6 ,,I
110 ...
ii:z
mO ~

100 .. .
tic
WI
~
701---------i'+-i

.
ii!
90
(7) (8)
11/10/38 -4/28/42 11/21116 - 12119117
1265. Days 394. Days
70.48 % 67.02%
Fig. AIOJ
II. Prj mary Legs 5. Primary Legs
6-7
 (App AIO-p. 12)
Bear Markets:

80
8 110/---------1
8

,l
\9
1\
70
I
70
70
,'.1 ~
~
~
~ I K I

,~ t I
,, X~ 50
8

i\
I
60
I II' 60

& i'" I ~
~ 6

50 70f----------1
50
50 (II) (12 )

11/15/61- 11/19/09 - 9125111

6/25/62 675. Days
(9) (10)
222.Days 37.83%
9/8...32 - 2/27/33 9/5/99 - 9/24/00
40.03% 5. Primary Legs
172. Days 385. Days
I. Leg.
60.85 % 46.54%
5. Pri mary Legs 5. Primary Legs

220 100 540

Q

200
\~ 90
500

~ ~
190
~ II I,o •: \ "
, Q
180

17 0 " ~ ,," "

:k " 0 ,I ~
V\ ....
...
....
... N

-"
16
H Y'i
• ~
& "
0 \ 400 - ..
~
.. N
. -
~w
... on
150

(13 ) (14) ( 15)

5/29/46 -6/14/49 9/30/12 -7130114 4/9/56-10/22/57

1112 Days 668. Days 56 I. Oays
31.91 % 3183% 24.39%

6. Primary Legs 5. Primary Legs 5. Primary Legs

Fig. AIOK
 (App. AIO-p. 13)

The dates and prices are tabulated in the figures

which follow. These data may be useful to readers with
computers - please send me a copy of your findings!
These data, incidentally, are not easy to obtain. While
the Dow averages are a matter of record, the terminal
dates for swings, under any specific set of definitions,
require a careful review of the record.

The prices used in the tabulations are closing prices

from 1897 until the publication of the hourly average
February I, 1932. From that point onward, the hourly
figures are used for all terminal points.

6-9
 (App. AIO-p. 14)

DATA - Bull and Bear Markets. Primary SWings,

T 4 /1 9/ 97
9/10/97
11/8/97
38.49
55.8Z
45.65
Secondary Reactions
I 4/7/04
g' 5/18/04
49.98
47.43
::: IZ!5/04 73. Z3
~ 12/15/13 75.Z7
~ z/3/14 83.19
4/Z5/14 76.97
tll 1/7/98 50.67 IZ/1Z/04 65.77 6/10/14 81. 84
E: 3/Z5/98 4Z.00 4/14/05 83.75 7/30/14 71. 4Z
4/1Z/98 46.3Z 5/ll/05 71. 37 IZ/14/14 56.76 (a)
4/Z1/98 43.Z7 8/Z3/05 8Z.82 12/24/14 53.17
8/Z6/98 60.97 9/7/05 78.60 1/23/15 58.5Z
10/19/98 51. 56 1/19/06 103.00 Z/z4/15 54.ZZ
4/Z5/99 77. Z8 3/5106 9Z.90 4/30/15 71. 78
5/31/99 67.51 4/3/06 98. 19 5/14/15 60.38
6/1Z/99 73.08 5/3/06 86.45 6/ZZ/15 71.90
6/ZZ/99 68.84 6/6/06 95.21
§ ~~;~~~15
67.88
9/5/99 77.61 7/13/06 85. 18 96.46
10/Z!99 70.95 ~ 10/9/06 96.75 11/9/15 91. 08
11/18/99 75.93 ~ 3/Z5/07 75.39 J2/Z7/15 99.Z1
vz! 18/99 58.Z7 5/3/07 85.0Z 3/z/16 90.5Z
tll 1/Z/00 68.13 5/Z7/07 77.30 3/16/16 96.08
(1l

...III 1/11/00 63.Z7 7/6/07 8Z.5Z 4/Z2/16 84.96

Z/5/00 68.36 8/ZI/07 69.Z5 6/1z/16 93.61
3/9/00 61. 11 9/6/07 73.89 7/13/16 86.4Z
4/6/00 66.15 11/15/07 53.00 10/5/16 104.15
6/Z3/00 53.68 lZ/6/07 61. 77 10/13/16 98.94
7/Z3/00 59.0Z lZ/17/07 56.85 II/ZI/16 110. 15
9/Z4/00 5Z.96 1/14/08 65.84 lZ/Zl/16 90.16
11/ZO/OO 69.07 Z/10/08 58.80 1/3/17 99. 18
lZ/8/00 63.98 tll 8/10/08 85.40 Z/2/17 87.01
lZ/Z7/00 71.04 § 9/Z2/08 77.07 Z/ZO/17 94.91
g' 1/19/01 64.77 11/13/08 88.38 IJ:I 3/1/17 91. 10
::: Z/15/01 70.78 Z/23/09 79.91 ~ 3/Z0/17 98.Z0
Z/Z8/01 67.00 6/5/09 94.46 ... 5/9/17 89.08
5/01/01 75.93 6/Z1/09 89.66 6/9/17 99.08
5/~/01 67.38 10/Z/09 100.50 9/4/17 81. ZO
6/l7/01 78.Z6 10/Z3/09 95.70 9/Z5/17 86.0Z
7/15/01 69.46 11/19/09 100.53 11/8/17 68.58
7/Z9/01 n.94 Z/8/10 85.03 II/Z3/17 74. Z3
8/6/01 69.05 3/8/10 94.56 lZ/I9/17 65.95
8/Z6/01 73.83 5/3/10 84. n Z/19/18 8Z.08
lZ/1Z/01 61. 61 5/ZI/10 89.66 4/11/18 75.58
4/Z4/0Z 68.44 ~ 6/6/10 8Z.05 5/15/18 84.04
~ 6/Z4/0Z 63.67 ~ 6/ll/10 86.Z8 6/1/18 77.93
~ 9/19/0Z 67.77 7/Z6/10 73.6Z 10/18/18 89.07
tll
lZ/15/0Z
z/16/03
59.57
67.70
-1-'0/'8/10 86.02
lZ/6/10 79.68 E: II/Z5/18 79.87
lZ/10/l8 84.50
4/13/03 60.79 ztct i: 86.0Z Z/8/19 79.15
4/Z1/03 64.56 3/4/11 81.80 6/5/19 107.55
8/8/03 47.38 6/19/11 87.06 6/16/19 99.56
8/17/03 53.88 9/Z5/ll n.94 7/14/19 lIZ. Z3
10/15/03 4Z.Z5 ~ 10/14/11 78.66 8/Z0/19 98.46
10/Z7/03
11/9/03
lZ!Z9/03
tll 1/6/04
45.41
42.15
49.35
47.07
t- 10/Z7/ll
9 / 3O/1 Z
3/Z0/13
tll 4/4/13
74.82
94.15
78.Z5
83.19
11/3/19 1I9.6Z
IZ/ZZ/19 103.55
tll 1/3/Z0
(1l
109.88
III Z/II/ZO 90.66
E: I/Z7/04 50.50 ~ 6/11/13 n.1I
83.43
... Z/ZI/ZO 95.63
I 3/11/04 46.41 ... 9/13/13
Z/Z5/Z0 89.98
Fig. AIOL 6-
 (App. AIO-p. 15)

DATA - Bull and Bear Markets, Primary Swings,

Secondary Reactions.
4/8/20
5/19/20
105.65
87.36
I 3/25/29
bl 5/4/29
297.50
327.08
I
g'
8/25/33
9/6/33
105.07
98.24
7/8/20 94. 51 §: 5/27/29 293.42 ::: 9/18/33 106.66
8/10/20 83.20 18/3/29 355.62 10/3/33 92.47
~9/17/20 89.95 8/9/29 337.99 10/10/33 99.88
~ 11/19/1.0 73.12 9/3/29 381. 17 10/21/33 83.64
12/4/20 77.63 B=;r 11/13/29 198.69 12/11/33103.36
1l/21/20 66.75 ~ 12/7/29 263.46 12/20/33 94.56
2/ 16/21 77.14 ::: 12/20/29 230.89 2/5/34 111. 16
3111/21 72.25 4/17/30 294.07 2/10/34 103.61
5/5/21 80.03 5/3/30 258.31 2/19/34 109.21
6/20/21 64.90 5/29/30 275.07 5/14/34 90.02
8/2/21 69.95 6/24/30 211. 84 5/18/34 96.15
8/24/21 63.90 7/28/30 240.81 6/2/34 91. 09
5/29/22 96.41 8/12/30 217.24 6/19/34 100. 79
6/12/22 90.73 9/10/30 245. 09 7/26/34 85.51
9/11/22 102.05 11/10/30 171. 60 8/25/34 95.71
9/30/22 96. 30 11/21/30 190.30 bl 9/17/34 86. 10
10/14/22 103.43 12/16/30 157.51 ~ 1/7/35 106.17
11/27/22 92.03 12/20/30 169.42 2/7/35 100. 11
3/20/23 105.38 bl 12/29/30 160. 16 2/18/35 107. 17
5/21/23 92.77 Z 1/8/31 173.04 3/18/35 96.44
5/29/23 97.66 ... 1/19/31 161. 45 9/18/35 134. 59
7/31/23 86.91 2/24/31 194.36 10/3/35 127.38
8/29/23 93.70 4/29/31 143.61 11/20/35149.05
10/27/23 85.76 5/8/31 154.41 12/19/35138.94
2/6/24 101.31 6/2/31 121. 70 3/6/36 159.13
5/20/24 88.33 6/27/31 156.93 3/13/36 150.42
8/20/24 105.57 8/6/77 133.77 4/6/36 162.45
10/14/24 99. 18 8/15/31 145.80 4/30/36 142.03
3/6/25 125.68 110/5/31 86.48 8/10/36 169. 50
3/30/25 115.00 10/9/31 116.79 8/21/36 160.80
11/6/25 159.39 1/5/32 71.24 3/6/37 194.47
11/24/25 148.18 11/15/32 85.88 6/17/37 164.07
2/13/26
bl 3/3/26 '''.08
144.44 1'/10/32
3/8/32
71.48
89.31
8/14/37
10/19/37
190.02
118.37
~ 3/10/26 153. B 6/2/32 44.09 10/29/37140.47
3/30/26 135.20 6/16/32 51. 04 11/23/37113. 52
4/24/26
5/19/26
8/14/26
10/19/26
12/18/26
144.83

166.64
137.16
145.66
161.86
"1"0'"
7/8/32

9/15/32
9/22/32
10/10/32
41. 03
80.68
65.82
75.64
58.47
tiltil 12/8/37
IlJ 12/14/37
... 12/21/37

~ 1/12/38
130.66
122.46
129.98
118.23
""'/37 134.76
1/25/27 152.73 ~ 10/20/32 65.92 2/4/38 117. 50
10/3/27 199.78 ~ 11/3/32 57.56 2/23/38 132.41
10/22/27 179.78 11/12/32 68.12 3/31/38 98.75
1/3/28 203.35 12/3/32 55.48 4/18/38 121. 27
2/20/28 191. 33 12/15/32 62.55 5/2/38 109.85
6/2/28 220.96 12/23/32 56.40 ~ 5/10/38 119.64
6/18/28 201.96 1/11/33 64.86 ..... 5/31/38 107. 16
11/28/28 295.62 2/27/33 50.16 145.70
12/8/28
2/5/29
2/16/29
257.33
322.06
295.85
3/16/33
4/4/33
7/18/33
63.48
55.36
109.43
1""38 9/14/38 131.17
9/21/38 139.47
9/28/38 128.07
3/1/29 321. 18 7/22/33 88.42 11/10/38 158.41
6-11 Fig. A10M
 (App. AID-p. 16)

DATA - Bull and Bear Markets. Primary Swings,

Secondary Reactions.

145.67
I 10/20/47 185.66 3/9/60 597.60

I""""
1/5/39 154.94 Zlll/48 164.77 4/18/60 632.68
1/26/39 136.42 ~ 6/14/48 193.81 5/2/60 599.61
3/10/39 152.40 ~ 9/28/48 175.98 6/9/60 659.69
4/11/39 120.82 10/23/48 190.30 til 7/25/60 600.61
6/10/39 140.33 11/30/48 171. 20 § 8/24/60 641. 56
6/30/39 129. 36 1/7/49 182.00 9/29/60 568.12
7/25/39 145.43 6/14/49 161. 10 10/17/60 597.52
9/1/39 129.20 6/12/50 228.44 10/25/60 566.05
9/13/39 156.00 7/13/50 197.11 5/22/61 710.93
~ 9/18/39 147.78 11 /25/50 235.62 6/19/61 676.46
~ 10/26/39 155.60 12/4/50 222.33 9/7/61 730.62
11/30/39 145.33 2/13/51 255.96 9/25/61 691. 86
1/4/40 153.02 3/15/51 243.39 11/15/61 737.50
1/15/40 143.41 5/4/51 263.64 Bear 6/25/62 526.69
4/9/40 151. 60 7/2/51 241.70 8/23/62 619.97
6/10/40 Ill. 29 9/17/51 276.39 tll 10/24/62 552.92
9/5/40 134.11 11/26/51 255.82 g,
...... 6/5/63 730.73
9/13/40 127.54 1/23/52 275.45 7/22/63 686.55
11/8/40 138.47 5/1/52 255.78 10/29/63 762.09
117.71 tll 8/11/52 280.62
2/19/41 11/22/63 711. 49
4/4/41 124.80 § 10/23/52 262.62
5/1/41 115.02 1/5/53 293.79
7/22/41 130.71 6/10/53 261. 36
12/23/41 106.22 8/13/53 277.57
1/6/42 114.48 9/15/53 255.11
4/28/42 92.92 1/3/55 409.52
4/6/43 136.97 1/18/55 386.41 ( Footnote:
4/13/43 130.23 3/4/55 420.07 (a) This is a new
7/14/43 145.59 3/14/55 391. 36 series; to be com-
8/2/43 134.00 7/27/55 469.74 parable, multiply
tll 9/20/43 142. 18 8/9/55 447.25 preceding figures
§ 12/1/43 128.89 9/23/55 487.45 by 0.7339 )
7/10/44 150.57 10/11/55 435.87
9/7/44 142.93 1/3/56 489.45
3/6/45 161. 82 1/23/56 460.34
3/27/45 152. 13 4/9/56 522. 18
6/26/45 169. 15 5/28/56 467.58
7/27/45 160.32 8/9/56 521. 39
2/4/46 207.24 t:l:I
10/1/56 468.38
2/26/46 184.80 CD
P>
11/7/56 499.72
5/29/46 212.50 ... 11/29/56 463.06
7/24/46 194.90 1/4/57 500.65
8/14/46 204.84 2/12/57 454.82
9/19/46 165.17 7/15/57 521. 27
9/26/46 174.96 10/22/57 419.79
10/30/46 161. 10 12/4/57 449.93
11/4/46 175.34 12/17/57 425.65
t:l:I 11/22/46 162.96 2/4/58 458.65
CD
...P> 2/10/47 184.54 tll 2/25/58 435.07
4/15/47 166.16 § 11/17/58 568.35
5/15/47 174.58 11/25/58 540.52
5/19/47 161. 84 8/3/59 679.52
7/25/47 187.05 9/22/59 615.76
9/9/47 174.62 1/4/60 685.83
 (App. All-p. 1)

Appendix All:
BIBLIOGRAPHY :

Alexander, S. S. Price Movements in Speculative Markets-

Trends or Random Walks; Industrial Mgt. Review, May 1961.
Ayres, L. P. Turning Points in Business Cycles,
Macmillan 1940
Allen, L. B.: Method for Stock Profits Without Forecasting.
Doubleday 1962
Barnes, L.: Your Investments, Am Research 1964
Barron's - National Business and Financial Weekly
Baruch, B.: My Own Story, Holt 1957
Bean, L. H. : How to Predict the Stock Market, Luce 1962.
Bernhard, A.: Evaluation of Common Stocks, Simon and
Schuster, 1959
Black, H.: Watchdogs of Wall Street, Morrow 1962
Bolton, A. H.: Elliott Wave Principle, Bolton Tremblay 1960
Bolton, A. H.: Introduction to Bank Credit Analysis, Bolton
Tremblay 1958
Bratt, E. C.: Business Cycles and Forecasting, Irwin, 1948
Brooks, J.: Seven Fat Years, Harper, 1954
Burns and Mitchell: Measuring Business Cycles, NBER 1946
Campbell, D.G.: Let's Take Stock, Bobbs Merrill 1959
Carpenter, H. G. A Successful Investor's Letters to His Son,
Simon and Schuster, 1934.
Carpenter, H. G.: This is Investment Management, Scarbor-
ough Press 1946
Cobleigh, 1. U.: How to Get Rich Buying Stocks McKay 1959
Cootner, P. H.: Random Character of Stock Market Prices,
MIT Press 1964
Cragg, A.: Understanding the Stock Market, Garden City 1929
Crane, B.: Sophisticated Investor, Simon and Schuster, 1959
Croxton and Cowden: Applied General Statistics, Prentice
Hall, 1939.
Dahl, C.: Consistent Profits in the Stock Market, Tri State
Offset, 1951
Darvas, N.: How I made $2,000,000 - Am Research 1960
Dewey, E.R. and Dakin, E.F.: The Science of Prediction,
Henry Holt 1947.
Drew, G.: New Methods for Profit in the Stock Market,
Metcalf Press 1941
Edwards and Magee, Technical Analysis of Stock Trends,
Magee 1948
Elliott, R. N.: Nature's Law - 1946
Elliott, R.N.: Articles in Financial World, March-August 1939

6-13
 (App. All-p. 2)

Elliott, R. N.: The Wave Principle. 1938

Engel, L.: How to Buy Stocks. Bantam. 1953
Epstein, R. C.: Making Money in Today's Market,
Economica 1959
Fisher, P. A.: Paths to Wealth Through Common Stocks.
Prentice Hall 1960
Fisher, P. A.: Common Stocks and Uncommon Profits.
Harper, 1958
Fritz, S. and Shumate, A. M. : Making the Dow Theory
Work, Russell 1960
Granville, J. E. : Strategy of Daily Stock Market Timing
for Maximum Profit, Prentice Hall 1960
Gutman, W. K.: You Only Have to Get Rich Once,
Dutton, 1961
Hamilton, W. P. : Stock Market Barometer. Russell,
1922/60
Hardy, C. O. : Odd Lot Trading, Brookings Institution.
Investors Intelligence: Encyclopedia of Stock Market
Techniques, Inv, Intelligence 1963
Jacwin and Costa, How to Accumulate Wealth Through
Stock Speculation, Kent 1958
Jiler, W. L.: How Charts Can Help You in the Stock
Market, Commodity Research, 1962
Kelly, F. C.: Why You Win or Lose, Kelly, 1930
Krach, E. T.: Bond Values and Yields, Cosmopolitan
1945
Leffler, G. L.: The Stock Market, Ronald, 1957
Loeb, G. M.: Battle for Investment Survi val, Simon and
Schuster 1957
Magee, John:General Semantics of Wall Street, Magee 1958
Mayer, M.: Wall Street Men and Money, Harper 1961
Merrill, Arthur C.: Investing in the Scientific Revolution,
Doubleday 1962
Mills F. C.: Price Quantity Interactions in Business
Cycles, NBER 1946
Moroney, M. J. :Facts from Figures, Penguin 1951
Neill, H.: Tape Reading and Market Tactics, Forbes
1931
Nelson, S. A.: ABC of Stock Speculation, Nelson 1900
Nelson, S. A.: ABC of Wall Street, Nelson 1900.
O'Connor, Wm: Stocks. Wheat and Pharaohs, Wener,
1961
Preston, C.: World of Wall S1. Journal, Simon and
Schuster 1959.
Rhea. R.: Dow Theory, Rhea 1932
 (App. All-p. 3)

Rhea, R :Dow' s Theory Applied to Business and Banking,

Simon and Schuster 1938
Rhea, R.: Story of the Averages, Rhea
Russell, R.: Dow Theory Today, Russell 1958
Sergeant, W. P.: Stock Market Behavior, Exposition
1957
Schabacker, R. W.: Stock Market Theory and Practice,
Forbes 1930
Schaefer, E. G.: How I Helped More than 10000 Investors
to Profit in Stocks, Prentice Hall 1960
Schwed, F.: Where are the Customers I Yachts?
Simon and Schuster 1940
Shulsky, S. :Stock Buying Guide, Arco 1959
Schultz, H. D.: Bear Markets, Prentice Hall 1964
Stabler, C.N.: How to Read the Financial News, Harper
1932
Stansbury, C. B.: Dow Theory Explained, Russell, 1938
Stearns, L: How to Live With Your Investments, Simon
and Schuster 1955
Sterling, D.: Wall Street, Doubleday 1955
Wise, T. A.: The Insiders. Doubleday 1962
Wright. W.: Forecasting for Profit, Wiley 1947
Wyckoff, R. D.: Method of Trading and Investing,
Wyckoff 1931
Wyckoff, R. D.: Wall Street Ventures and Adventures.
Harper 1930

6-14
 (App. A12-p. 1)

Appendix A12=
HOW TO MAKE THE CHI SQUARED TEST:

How significant are the findings in this book?

One of the simplest checks of significance is the "Chi

Squared Test". You will find a discussion of this test
in all standard statistical text books. We are including
below an example of this test which can be applied to
many of the findings in this book.

The formula is:

Where: -c': Chi Squared

o/' -
! = "Total of"
0= -Ob s e r ve d Result
E= Expected Result

For exarnpl er on the day before Christmas, In the

last 67 years, the market rose 49 times and declined
18 times. Is this significant?

In the 67 years, the market rose on 53.16% of all

the days in the period, and declined on 46.84% of the
days. If we multiply these percentages by 67, the ex-
pectation for the day before: Christmas is UP: 35.6
DOWN: 31. 4 Substituting these values in the formula:

(49-35.6)2 + (18-31. 4)2 = 10.7

35.6 31.4
 (App A12-p. 2)

Evaluation: Many of the findings in this book are of

the type stated in the example. The answer is a per-
centage up plus a percentage down, with a total of l~
For this simple type of problem (the statisticians call it
"one degree of freedom"), the results can be evaluated
as follows:

If Chi Squared is 3.84, the probability of getting

this Chi Squared or higher is 50/0 or one in twenty. This
is rated "Probably Significant" by some statisticians.

If Chi Squared is 6.64. the probability of getting this

value or higher is 1. 00/0 This is rated "Significant".

If Chi Squared is 10.83, the probability of getting

this value or higher is 0.10/0 or one in a thousand. This
is considered "Highly Significant".

In the example above, the Chi Squared is 10.7 •

This difference from expectation is close to the 0.10/0
level, and the result could be rated "Highly Significant".

6-12
 (App. A13-p. 1)

Appendix A13:
THE ELAPSED TIME CALCULATOR:
- - A tool for students --

ELAPSED TIIiI CA LCULATOIl

-DAY
DAY OF THE YEAR
OF
MONTH MAR. APR. MAY JUNE JULY AUG. SEP. OCT. NOV. DEC. JAN. FEB.

I 1 32 62 S3 123 154 IB5 215 246 276 307 33B

I
2 2 33 63 94 124 155 IB6 216 247 277 306 339

i
3 3 34 64 95 125 156 IB7 217 248
__•.
...
27B 309 340

I 4
5
4
5
35
36
65
66
96
97
126
127
!!I7
158
18B
IB9
218
219
249
250
279
280
310
311
~I
342
I 6 6 37 67 98 128 159 190 220 251 2BI 312 343

7 7 38 68 99 129 160 191 221 2&2 2B2 313 344

8 8 39 69 100 130 161 192 222 253 283 314 345
9 9 40 70 101 131 162 193 223 254 2B4 315 346

10 10 41 71 102 132 163 194 224 255 265 316 347

II II 42 72 103 133 164 195 225 256 286 317 348
12 12 43 73 104 134 165 196 226 257 287 318 349

13
14
13
14
44
45
74
75
I 105
106
135
136
166
167
197
19B
227
226
25B
259
28B
289
319
320
350
351
~ I
15

16
I!I

16
46

47
-+ . -76

77 I 108 1'38
1----
168

169
199

200 230
229 260

261
290

291
321

322
352

353
17
16
17
18
48
49
I 78
79
I 109
110
139
140
170
17\
201
202
231
232
262
263
29l:
293
323
324
354
356

19 19 50
---r'--j-- r---;l
80 II I \41 172
---- f - - - -
203 233
--_.
264 294 325 356
20 20 51 B I
B2
I 112
II 3 I
142
143
173 I 204 234 265 295 326 357
21 21 52 174 205 235 266 296 327 35B
. - f------- f----- --l---- . .._----
22
23
22
23
53
54
B3
84
114
115
144
145
175
176
I 206
207
236
237
267
26 B
297
29B
328
329
359
360
24 24 55 85 116 146 177 208 238 269 299 330 361

25 25 56 86 117 147 178 209 239 270 300 331 362

26 26 57 B7 liB 148 179 210 240 271 301 332 363
27 27 58 B6 119 149 IBO 21 I 241 272 302 333 364

28 28 59 89 120 150 IB 1 212 242 273 303 334 365

29 29 60 90 121 15 I 182 213 243 274 304 335 366
30 30 61 91 122 !!I 2 IB3 214 244 275 305 336
-- f------
31 31 92 153 184 245 306 337

To find numb" of days bel..,een two do tes: [Komple: April 24.1962 10 March 12, 1964

(I) Put down 365. for each ree.ae passed over. II) t- 355 (For feb. 28, 19631
(For Q leap yeor Feb. 29 use 366.) + 366 (For Feb 29, 1964)
(2) Add "Ooy of the year" for the second date. 121 + 12 ( For Mareh 12,1964)
(3) Sub'roci "Day of Ihe yeo," for the first dOle. 743 (Totol)

(4) Remainder 1'1 Ihe number of days between the two dotes. (3) - 55 ( For April 24 1
(962)

(4) 588 « Elapsed time In days)

COPYRIGHT 1962 Arthur A Merrill
 (App. A14-p. 1)

Appendix A14:
PRICE EARNINGS CALC ULATOR:
-- A tool for students --
EARNINGS
100 0.12

90 PRICE/EARNINGS 0.15

CALCULATOR
80
0.20

70

0.30

60
,, PRICE/EARNI NGS 0.40
,

50 100 0.50
90
80 0.60
70
60 0.70
EXAMPLE:
40 PRICE' $ GO. ' 50 0.80
EARNINGS· $4.00 " ... 0.90
40
ANSWER. ON 1.00
,,
MI DOLE SCALE:
... 30
PRICE/ EARNI NGS • 15.
...
-,
30 -... ... ... 20 1.50

~5' ., ,
,, 2.
25 ... ,
10
9
8
7 3.

20 6

5 ' ... 4.

5.
FOR PRICE SCALE OF 100-1000, MULTIPLY 80TH PRICE
SCALE AND EARNINGS SCALE 8Y TEN.
6.
15
FOR PRICE SCALE OF 1-10, DIVIDE BOTH PRICE 7.
SCALE AND EAAI'tINGS SCALE BY TEN.
a
9.
10.

COPYRIGHT 1959 ARTHUR A. MERRILL

10
10 15.
 (App. AI5-p. 1)
Appendix A15:
300 YEAR CALENDAR:
- - a tool for students --

- YEAR TYPE LETTER-

-- Go down column to first three digits of year

...--then across to column of f ourth digit I

and obta i n Year Type Lett e r.

~O I 2 3 4 5 6 7 8 9
175 A B A B C K F G A
176 I 0 E F N B C 0 L G
J E F G H C o E
177 A B
'78 M A B C K F G A 0 ,
179 E F N 8 C D L G A 8
180 C 0 E F N B C 0 L G
18 I A B J E F G H C 0 E
182 M A B C K F G A I 0
183 E,F N B C 0 L G A B
184 J E F G H C 0 E M A
L85 B C K F G A I 0 E F
186 N B C 0 L G A B J E
187 F G H C 0 E M A B C
188 K F G A I 0 E F N B
.189 C 0 -~ A B J E F G
190 A B C 0 L G A 8 J E
,191 F G H C 0 E M A B c
192 K F G A 1 0 E F N B
193 C 0 L G A B J E F G
194 H C o E M A B C K F
195 G A I 0 E F N B C 0
196 L G A e J E F G H C
197
198
0
, E
0
M A B C K F G A
E F N B C 0 L G
199 A B J E F G H C 0 E
200 M A B C K F G A I 0
201 E F N B C 0 L G A B
202 J E F G H C 0 E M A
203 B C K F G A I 0 E F
204 N B C 0 L G A B J E
205 F G H C 0 E M A B C

FIG.AI5A
 (App. AlS-p. 2)

CALEN DAR .... 1753 - 2059

TO USE:
FIRST: OBTAIN YEAR TYPE LETTER
FROM OPPOSITE PAGE,
'--THEN FIND MONTH,
/
GO STRAIGHT ACROSS TO YEAR TYPE LETTER.
THEN DOWN TO CALEN DAR.

JAN. 31d. E L G N D K A H C J B I F M 31.d. JAN.'

FE 8.28-29 B I 0 K A H E L G N F M C J 28·29FEB.
MAR.31.d 8 H 0 J A N E K G M F L C I 31.d. MAR.
APR. 30.d F L A N E K 8 H 0 J C I G M 30d. APR.
MAY 31.d 0 J F L C I G M B H A N E K 31.d. MAY
JUNE 30d. A N C I G M 0 J F L E K B H 30.d. JUNE
JULY 31.d. F L A N E K B H 0 J C I G M 31.d. JULY
AUG. 31.d. C I E K B H F L A N G M 0 J 31.d. AUG..
SEP. 3O.d G M B H F L C I E K 0 J A N 30.d SE P.
OCT. 31.d E K G M D J A N C I 8 H F L 31.d. OCT.
NOV. 3011 B H 0 J A N E K G M F L C I 30.d. NOV.

r
DEC. 31.d. G M B H F L C I E K 0 J A N 31.d. DEC.

.~
/

/ (~
~
5 M Tu W Th F 5 5 II Tu W Th F 5 5 II Tu III Th F 5
I 234~67 I 2 3 4 ~ 6 1 2 3 4 !i
8 9 10 II 12 13 14 7 8 9 10 II 1213 6 7 8 9 10 II 12
I~ 16 17 18 19 2021 14 I~ 16 17 18 19 2C 13 14 I~ 16 17 18 19
22 23 24 es 2627 28 2.1 22 23 24 25 26 27 20 21 22 23 24 as 21
129 3031 28:n 3031 27 28293031

5 M Tu \1/ Th F S 5 M Tu \1/ Th F 5 5 M Tu III Th F 5

2 I I2 3 4 1
3 4 !i 6 7 8 9 !i 6 7 8 9 10 II 234 5 678
10 II 12 13 14 15 18 12 13 14 15 16 17 18 9 10 II 12 13 14 15
17 18 19 2021 22 2~ 19 202122 23 242!i 16 17 18 19202122
24 25 2627 2829 3CI 26 27 28 29 30 31 23 24 25 26 2728 29
31 3031
-,

II Tu \II Th F 5
EXAMPLE:
JULY 4, 1776 I 2 3
COPYRIGHT 19~2

YEAR TYPE; H 45678910 ARTHUR A. MERRILL

DAY OF WEEK: THURSDAY II 12 13 14 15 16 17
18 192021 222324
2~ 2627 28 29 3031

6-8 FIG. AI5B

 About the author:

Arthur A. Merrill, BS, MBA, devoted 33 years to the

General Electric Company, in engineering. statistics,
planning, forecasting, and management. Since 1962 he
has been President of Merrill Analysis, Inc , , and has
edited a statistical letter "Technical Trends", to help
investors in their decisions. He is the author of many
articles and books. He is a Past International President
of SPEBSQSA, a member of the American Statistical As-
sociation, the Society for the Investigation of Recurring
Events and the Foundation for the Study of Cycles. His
home is in Chappaqua, New York.

6-6
6-4
6-Z