In the December 2002 issue of Technical Analysis of
Stocks & Commodities magazine, Jay Kaeppel presented a system of being in
the market only when historical seasonality is positive. Mr. Kaeppel used
four tendencies:
1. The two days immediately prior to an
exchange holiday (New Year's Day, MLK Day, Presidents' Day, Good Friday,
Memorial Day, July 4th, Labor Day, Thanksgiving and Christmas).
NOTE: This has shifted somewhat over the years, and we have made the
appropriate adjustments beginning in the 2008 calendars.
2. The last trading day of the month and
the first four trading days of the next month.
3. November 1st through the 3rd trading
day in May.
4. The most favorable 15 months of the
48-month Presidential election cycle. This begins on October 1st two
years prior to each Presidential election, and ends on December 31st of the
following year.
Each tendency gets a score of 1. When you have
two or more tendencies working for you at the same time (i.e. a score of +2 or
better), market performance over time is greatly enhanced.
Mr. Kaeppel tested this system on the Nasdaq from
1972 - 2002 and found that readings of +2 or greater outperformed non-seasonal
days by a significant degree. He also found that readings of 0
underperformed all nonzero seasonal days by a likewise significant
degree.
The calendars below shows the score for each of the
upcoming months. If the day is highlighted in red, then the score is
"maximum bearish" by being a 0. If it is highlighted in green, then it is
more bullish than a usual day by being scored a 2, 3 or 4.
If you see a 
or 
then there is some extra seasonality evident that day, either negative
or positive, respectively. For example, the day after an options
expiration tends to be negative, so you will see a on those days, as well as some holidays which show a negative bias
when traders return from the break.
Below the calendars, you will see some performance
statistics from trading on days with each of the possible "scores".
|
AUGUST 2008 |
| Mon |
Tue |
Wed |
Thu |
Fri |
| |
|
|
|
1
1 |
|
4
1 |
5
1 |
6
1 |
7
0 |
8
0 |
|
11
0 |
12
0 |
13
0 |
14
0 |
15
0 |
|
18
0 |
19
0 |
20
0 |
21
0 |
22
0 |
|
25
0 |
26
0 |
27
0 |
28
0 |
29
2 |
|
SEPTEMBER 2008 |
| Mon |
Tue |
Wed |
Thu |
Fri |
|
1
MKT
CLOSED |
2
1 |
3
1 |
4
1 |
5
1 |
|
8
0 |
9
0 |
10
0 |
11
0 |
12
0 |
|
15
0 |
16
0 |
17
0 |
18
0 |
19
0 |
|
22
0 |
23
0 |
24
0 |
25
0 |
26
0 |
|
29
0 |
30
1 |
|
|
|
|
OCTOBER 2008 |
| Mon |
Tue |
Wed |
Thu |
Fri |
| |
|
1
1 |
2
1 |
3
1 |
|
6
1 |
7
0 |
8
0 |
9
0 |
10
0 |
|
13
0 |
14
0 |
15
0 |
16
0 |
17
0 |
|
20
0 |
21
0 |
22
0 |
23
0 |
24
0 |
|
27
0 |
28
0 |
29
0 |
30
0 |
31
1 |
|
NOVEMBER 2008 |
| Mon |
Tue |
Wed |
Thu |
Fri |
|
3
2 |
4
2 |
5
2 |
6
2 |
7
1 |
|
10
1 |
11
1 |
12
1 |
13
1 |
14
1 |
|
17
1 |
18
1 |
19
1 |
20
1 |
21
1 |
|
24
1 |
25
2 |
26
2 |
27
MKT
CLOSED |
28
2 |
|
DECEMBER 2008 |
| Mon |
Tue |
Wed |
Thu |
Fri |
|
1
2 |
2
2 |
3
2 |
4
2 |
5
1 |
|
8
1 |
9
1 |
10
1 |
11
1 |
12
1 |
|
15
1 |
16
1 |
17
1 |
18
1 |
19
1 |
|
22
1 |
23
1 |
24
2 |
25
MKT
CLOSED |
26
2 |
|
29
1 |
30
1 |
31
3 |
|
|
PERFORMANCE DATA
The table below uses the
S&P 500 from 1950 through mid-2004, and shows the market?s performance for
days with each of the four possible total scores.
|
S&P 500 Performance by
Seasonality Index
1950 - 2004 |
| |
0 |
1 |
2 |
3 |
4 |
| Avg
Ret |
-0.03% |
0.01% |
0.08% |
0.23% |
0.21% |
| %
Pos |
50% |
51% |
56% |
62% |
70% |
| #
Days |
3604 |
5647 |
3555 |
860 |
33 |
We can see from the table that
on a day when none of the four seasonal biases are present, the S&P actually
showed a negative average return, and was positive on almost exactly 50% of the
3,604 days.
As we notched more biases in
our favor, the S&P?s performance improved, slightly at first and then
dramatically. By the time we had 3 out of the 4 biases present on a given
day, the S&P showed an impressive average return of 0.23%, with 62% of the
days being positive. Those rare ?4? days, when all biases were present,
also gave an impressive performance, with even more of the days being
positive.
Let?s say that in 1950, you
had $10,000 to trade and decided to buy the S&P 500 (cash index) at every
open and sell it at every close, using all your money. Your $10,000 would
have grown into just under $2 million by now. Now let?s say that when a
?0? day arrived, you decided to stay in cash and not trade that day.
Interestingly enough, even
though you were in the market only 74% of the time, your $10,000 would have
still grown into just under $2 million. Now let?s make it really
interesting?when you saw a ?0? day, you still went to cash, but if it was a ?3?
or ?4? day, then you decided to play the odds and you leveraged your bets
2-to-1. In that case, your $10,000 initial stake would have grown into
more than $13 million, a return of over 133,000%.
| |
Buy Every Open,
Sell Every Close? |
Buy at Open and
Sell at Close, IF Day is not a ?0?? |
Buy at Open and
Sell at Close, IF Day is not a ?0?, and Double Down if it is a ?3?
or ?4?? |
| $10,000
Becomes? |
$1,994,652 |
$1,997,418 |
$13,320,444 |
| %
Return |
19,847% |
19,874% |
133,104% |
PLEASE
NOTE that this
is NOT a trading system. We have made no adjustments for dividends,
slippage, commissions, interest on cash balances, etc. Also, for the open
price I used yesterday?s closing price, so gap opens are not accounted for ? not
that it matters, since it wasn?t really even possible to trade the S&P 500
itself until the futures market came along in the early 1980?s.
By watching when
the market may or may not have a positive bent to it, we can adjust our
expectations accordingly. While it is difficult to implement this type of
information in practice, we do suggest that especially shorter-term traders
watch the Seasonality Index daily, and when we get a ?3? or ?4? day, it probably
pays to be more aggressive than usual on the long side.
It would be
foolhardy to suggest going short on ?0? days, but we do think it is more
difficult to make money on the long side on those days than others.
Longer-term traders may want to keep track of what the upcoming month may hold
as far as these scores go, and when they begin to add up, it can pay to be more
aggressive on the long side.