Stop relying on luck, use this 75% correct pattern
The foundational insight of behavioral finance is that people display
consistent information processing biases that distort their decision making
under conditions of risk and uncertainty. One of the most common biases is
the assumption that the present and future will mirror the past. This bias
alone has enriched many a casino, as gamblers who experience a string of (lucky
and random) wins become convinced that they have a “hot hand” and
raise their bet sizes. We see the same thing among traders, who become
bold after they win, only to cut their size after losses. The result is a
roller coaster, ensuring that market participants trade most aggressively just
when they’re due for a fall.
A variation on this bias is the assumption that markets will continue moving
in a given direction after having moved that way in the past. Once we see
several consecutive rising or falling bars on a chart, it is hard to not see a
trend in the data. That identification of a trend is like the gambler’s
perception of the hot hand. We assume that the market is unusually hot or
cold and bet that this will continue. But do we typically see continuation
in market moves?
To be sure, there are traders who have succeeded at trend following, as
Michael Covel has noted in his book. Invariably, however, the strategies
used by the trend followers include, not only algorithms to help them separate
genuine strength/weakness from random market runs, but also money management
methods to preserve capital when trends are not your friends. This is
important, because it turns out that reversion to the mean–the tendency of
markets to return to their price averages after straying from them–is the norm,
not the exception.
I recently conducted some research on mean reversion, examining market data
in SPY
(
SPY |
Quote |
Chart |
News |
PowerRating) since October, 1999 (1499 days). I found that 1184 of those
1499 days–about three-quarters of all occasions–returned to the previous day’s
mean price at some point during the trading day. The results were almost
identical for the Dow Jones Industrial Average
(
DIA |
Quote |
Chart |
News |
PowerRating) and the NASDAQ 100
average
(
QQQQ |
Quote |
Chart |
News |
PowerRating). Since the beginning of 2005, the results have been similar: 151
out of 184 in SPY have returned to their previous day’s midpoint.
I then wondered if this pattern of mean reversion occurs over larger time
frames as well. Since the start of 2000 in SPY (300 weeks), 231 weeks have
returned to their prior week’s midpoint. For QQQQ, 250 of the 300 weeks
return to last week’s midpoint. Monthly data? Since 1996 in SPY, 81
out of 116 months have returned to their midpoints–including eight of the last
nine in 2005.
If your market indicators, intuitions, and research help you identify
strength and weakness as markets stray from their means, you have the basis for
high probability trades and trading systems. Markets do exhibit hot hands,
but like most hot things, they cool over time. About 75% of the time, by
my calculations.
Brett Steenbarger
Brett N. Steenbarger, Ph.D. is Associate Clinical
Professor of Psychiatry and Behavioral Sciences at SUNY Upstate Medical
University in Syracuse, NY and author of The
Psychology of Trading (Wiley, 2003). As
Director of Trader Development for Kingstree Trading, LLC in Chicago, he has
mentored numerous professional traders and coordinated a training program for
traders. An active trader of the stock indexes, Brett utilizes
statistically-based pattern recognition for intraday trading. Brett does not
offer commercial services to traders, but maintains an archive of articles and a
trading blog at www.brettsteenbarger.com.