Does breadth tell us anything that price doesn’t?
Commentators frequently note the strength or weakness of a market based upon
the number of advancing versus declining issues. It sounds very impressive
if the Dow Jones Industrial Average rose 80 points with advancers leading
decliners by a 2:1 margin, and it’s scary to hear about markets that are down on
negatively skewed breadth. But does breadth tell us anything that price
doesn’t? My stats show that daily advance-decline figures correlate
slightly over .80 with price change for the 732 trading sessions between
January, 2003 and November, 2005. That means that nearly two-thirds of all
the ups and downs in the breadth measure are a mere reflection of the fact that
the market has gone up or down during that period.
The past six days in the market have seen a relative decoupling between
breadth and price change. The S&P 500 Index (SPX) has barely budged
over the six day period, but advances have outnumbered declines during that same
time by more than 2000 issues. This has only occurred four times since the
start of 2003, and the market was up six days later after each of those
occasions.
That got me thinking.
What if breadth means little on very strong and very weak market
occasions–everyone knows that breadth will be lopsided when the market is up or
down big–but has value as an indicator when the market is relatively
flat? Interestingly, when I looked at the 128 occasions when the six-day
S&P 500 was up or down within half a percent of breakeven, the correlation
between price change and breadth was only .41, about half of the correlation for
the overall sample. This means that price change only accounts for about
16% of the ups and downs in breadth during relatively flat markets–far less
than the two-thirds for the general sample. It would appear that breadth
is not a unique indicator when the market is up or down by a lot, but becomes a
unique variable during relatively flat periods such as the past six days.
Statisticians call this a non-linear relationship.
That’s fine, but does the unique indicator possess practical predictive
value?
I found 26 occasions in which the market was up or down within a tenth of
percent of breakeven over a six-day period–a very flat market similar to the
current one. I then divided the 26 occasions into halves based on their
breadth readings, creating a strong breadth flat market group and a weak breadth
flat market group. The average price change for the two groups was almost
identical (.02% vs. .01%), but the strong breadth group averaged 1880 more
advances than declines. The weak breadth group averaged 417 more declines
than advances. In case you haven’t noticed, this is as close to a
laboratory experiment as we can get in the market. By statistically
controlling the price change variable, creating two equivalent groups, we can
now determine the unique contribution of breadth to market prediction.
It turns out that, for the sample overall (N = 26), the market six days after
a flat six days achieves an average gain of only .02%, with 16 occasions up and
10 down. That’s not a tradable edge. When we compare the strong and
weak breadth groups, however, a different picture emerges. Six days after
a strong breadth flat period of six days, the market was up 10 times, down 3 for
an average gain of .32%. Six days after a weak breadth period flat period
of six days, the market was up 6 times and down 7 for an average loss of
-.25%. It appears that, when buyers are acquiring the broad list of stocks
even as the large cap S&P is flat, the six-day outlook for the S&P is
better than when buyers aren’t taking the initiative.
No one likes bad breadth, and flat markets appear to be no exception.
And that could be a breadth of fresh air for the next six trading days.
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.
He is currently writing a book on the topics of trader development and the
enhancement of trader performance.