To print article, click
Print on your browser's File menu.
Go
back
Posted
11/7/2002 |
The Speculator
Recent articles:
• Nothing pays
dividends like a new dividend, 10/31/2002
• 5 genuine buys
on a Street of impostors, 10/24/2002
• Higher
dividends are no magic bullet, 10/17/2002
More...
|
|
| sponsored by: |
 | |
The Speculator
Do
comeback kids keep going?
In this season when Angels, politicians and
stocks come roaring back, we wondered if comeback kids tend to keep
going strong. If only stocks did as well as World Series
winners.
By Victor
Niederhoffer and Laurel Kenner
Inspiring comebacks are everywhere these days. On
the baseball diamond in Anaheim, in the skies with the Concorde
flying again after a 17-month absence, and in the Senate in
Washington, D.C., where both former Sen. Frank Lautenberg and GOP
icon Elizabeth Dole will soon be taking seats.
Intimately
related to this, and reflecting social trends like a mirror, is the
spate of comebacks in the stock market. Many indexes around the
world are up 20% to 30% in the last month or so, including the Dax
in Germany (+23%), the ISE in Turkey (+37%), and even the
beaten-down-by-the-bond-market Bovespa in Brazil (+21%). Within our
shores, we find that the Nasdaq 100 ($NDX.X)
is the third best-performing major average this quarter in the whole
world with a 26% gain. Many stocks in the S&P 500 index ($INX)
(43, to be precise) that were money-losers for the year as of the
end of September 2002 came back to be winners at the end of October.
Notable examples are Paccar (PCAR,
news,
msgs),
the truck manufacturer, which was down 22% year to date in September
but ended October up 1.3% for the year; and Sara
Lee (SLE,
news,
msgs),
the food processor, which went from being down 18% for the year in
September to up 3% for the year in October.
On the other
hand, only eight stocks in the S&P 500 universe are experiencing
"letdowns" -- that is, they were profitable for the year at the end
of September but gave up their total gain (and then some) in the
month of October. Notable companies include Tenet
Healthcare (THC,
news,
msgs)
the hospital management company, whose stock went from being up 25%
year-to-date in September to being down 27% YTD in October; and
Amerada Hess (AHC,
news,
msgs),
the oil and gas company, whose stock went from being up 9% YTD
September to down 18% YTD October.
How will the comebacks and
letdowns perform in the future? Will the winners gain added gusto in
subsequent periods, and will the losers lose hope? Or will there be
letdowns and catch-ups? How does the market perform after comebacks
and letdowns? How do individual stocks perform after moving from
losers to winners, and vice versa?
Diamonds are forever
A good place we
found to start the consideration of comebacks and letdowns is in the
annals of baseball. Baseball is the quintessential American sport.
Its organization, culture and techniques frequently represent what
goes on in American life. Because baseball's rules and customs are
set with a view to encourage fan participation in the game, the
world of baseball provides a keen insight into consumer behavior --
and thus, into the American economy, several years into the
future.
We have studied this in some detail and find, for
example, that the prevalence of scientific baseball versus slugging
baseball is highly positively correlated with the returns in the
U.S. stock market in the subsequent year. (See our article from Oct.
11, 2001, "Are
home runs bad for stocks?"). A typical finding was that
following years in which home runs were at a five-year high, the
returns in the next year were some 10% per year worse than the
returns in "normal" years. The correlation between changes in the
number of home runs in a five-year period and the returns in the
stock market in the subsequent five-year period was a significantly
negative 0.20.
The Major League season just ended brought a
new crop of insights for stock-market participants. The one thing we
can say for sure about the 2002 season is that the Anaheim Angels
were the Comeback Kings. They came from behind to win in 43 of their
162 games; that's a comeback in 26% of their games. In 22 of those,
the Angels won with their last at-bat. Furthermore, their win in
Game 6 of the World Series after coming back from a 5-0 deficit in
the seventh inning was the biggest comeback ever for a team facing
elimination. Only the 1919 Chicago White Sox and the 1925 Pittsburgh
Pirates (who both came back to win from 4-0 deficits) come close to
a comparison. An Anaheim Angel also won the Comeback Award from
Major League Baseball this year.
Do
comebacks carry forward?
Comebacks are great morale
boosters. But does this boost carry forward into productive
performance in the next game? In other words, do stocks that come
back have increased or decreased resilience the next day?
To
answer these questions, we studied the performance of S&P 500
Index futures over the last six years from the end of 1996 to the
end of October 2002. We divided the trading day into nine intervals,
to create a rough correspondence with baseball's nine innings.
These, we defined as The Open, 10 a.m., 11 a.m., noon, 1 p.m., 2
p.m., 3 p.m., 4 p.m. and The Close. We looked at all occasions when
the S&P 500 index came back after being down during a day at any
one of the first eight intervals to be up at the close. We
eliminated all overlap in these calculations.
We found that
the index showed a comeback on 17% of all days that it was behind at
one of the first eight intervals, through the close that same day.
Recall that the Anaheim Angels were able to come back and
win in 26% of all the games they played. Thus, the Angels were about
1 1/2 times as likely to come back (26% versus 17%) as the index. To
put it another way, stocks are just 65% as likely to show a comeback
during the day as the Anaheim Angels.
Stocks, as the world
recently learned, can come down from being up, as well as come back
up from being down. We defined a "letdown" as a day when the index
was up on any one of the intervals identified earlier and then
woefully declined to finish the day down. Letdowns, thus defined,
came on 33% of all days. However, since up days are slightly more
frequent than down days in our sample, the index showed a letdown on
a little more than 19% of all days. Thus, we find that stocks are
slightly more likely to experience a letdown than a
comeback.
| Comebacks and letdowns in S&P 500
|
| |
# Days |
% Total |
| Comebacks |
|
|
| Down at least once in a day but comes back to close
up |
286 |
17% |
| Down all day intraday and did not close up |
497 |
30% |
| Total number of chances given to make a
comeback |
783 |
47% |
| Letdowns |
|
|
| Up at least once in a day but letdown and closed
down |
315 |
33% |
| Up all day intraday and did not close down |
553 |
19% |
| Total number of chances given to make a
letdown |
868 |
53% |
| Closed unchanged |
71 |
|
| Days studied 1/96-10/02 |
1722 |
|
| Net chances to make either comebacks or
letdowns |
1651 |
100% | |
Now
that we have laid a statistical foundation for the prevalence of
comebacks and letdowns, we can turn to the prognosis. Are they good
or bad for the market? We looked at the losses and gains from
comebacks and letdowns in each of the eight intervals from the close
on the day of the comeback, to the close on the day after. In other
words, granted that the team performs very well on the day of the
comeback. How does the team perform the day after?
The
results show that stocks, as represented by the S&P 500 Index,
tend to let down the day after late-day comebacks. In particular,
there were 263 days when stocks were down as of noon but closed up
on the day. Of those 263 days, the market performance by the close
of next day was down 1.7 points. Taking account of the uncertainty
of this estimate, we can be 95% confident that the actual
performance of the market on days like this would be between -1.3
and -2.1 points. Thus, stocks tend to let down after a good
comeback.
This is somewhat contrary to the accepted view,
which always gets extraordinarily bullish after stocks rally and go
from being down in any period to up (or vice versa). In fact, many
of these days are what technicians might refer to as "a key
reversal", an event invariably posited as being highly bullish or
bearish. We cannot resist querying the extent of contempt that
people who posit bullishness or bearishness without a test after
such days, must have for their readers and customers.
The sad
part of our test is that the performance of the market after a
comedown is 50-50, about as close to a zero expectation as one might
expect after a random coin toss.
Next, we turn to the
question of the effect of comebacks and letdowns on individual
stocks. There are many ways to consider this question and we
encourage our readers to come up with their own methodology or to
suggest methods to us. We chose to define a comeback company as one
that was down year to date at the end of any given month, and up for
the year to date as of the end of the next month.
For
example, 3M (MMM,
news,
msgs)
stood at 118 at the end of 2001. At the end of September, it was
down to 110; it ended October at 127. So as of September end, the
stock was down 6.7% for the year; it was up 7.8% for the year by the
end of October. Thus, it was a comeback stock. As we reported
earlier, there were 46 comebacks within the S&P 500 complex, a
count far greater at the end of October than the count of comeback
stocks for the five months preceding it.
This is just the
reverse of the situation for letdowns. During previous months,
letdowns ran at a rate of 60-70 per month. In October there were
just eight.
Suffice it to say that our results as to whether
letdowns or comebacks are bullish or bearish for individual stocks
would be classified in the medical literature as "a negative study",
with the results not definitive in either direction. The comebacks
in September did quite well, but the comebacks in June fared quite
poorly.
We conclude from our work so far that, at the very
least, the comebacks for the S&P 500 index are bearish for the
following day.
Final
note
The subject of what baseball can teach us about the
market is a fascinating one, one that opens up myriad fruitful
paths. Three of the eagles that individually and collectively
possess that unique language of time and place that is so much
sharper and insightful than ours are our readers and friends Tim
Melvin, Tom Klosek and Tom Gross. They emphasize that the loose
structure of the game, no fixed time limit, a variety of ways to
score, the endless variations of strategy, the prevalence of streaks
and slumps, the reliance on math and statistics, the importance of
experience, the terrible unreliability of it all, correspond well to
success in life and markets. We have posted their thoughts on the
interactions and implications on our Web site. We encourage
reader response to all our articles and try to answer all queries.
We will send the list of comebacks and letdowns in individual stocks
to all who comment, critique or augment our article.
We would
like to thank our stockmetricians Gitanshu Buch and Patrick Boyle
for their statistical contributions to this report.
At the
time of publication, Laurel Kenner and Victor Niederhoffer owned the
following stock mentioned in this column:
Paccar.
|
