As I reported recently on my research blog, I've found it worthwhile to track herd-like behavior in the S&P 500 Index by evaluating the proportion of volume that is going into advancing vs. declining stocks.

For example, going back to 2004 (N = 621 trading days), we've had 42 days in the S&P 500 Index that have been up by 1% or more. Overall, the market has been up by an average of .06% the next day (24 up, 18 down), not much of an edge over the average gain of .02% (343 up, 278 down) for the entire sample.

When we divide the strong up days in the S&P in half based on the ratio of advancing stock volume to declining stock volume, however, we see an interesting pattern. When the S&P 500 is up strong and volume is heavily concentrated among advancers (i.e., the herd is buying), the next day in the S&P averages a loss of -.07% (10 up, 11 down). When the S&P 500 is up strong and volume is not heavily concentrated among advancers, the next day in the S&P averages a gain of .19% (14 up, 7 down).

It thus appears that strong market rises are more likely to persist in the near term if they are not accepted by the herd.

A different pattern shows up for market declines.

Since 2004, we've had 47 days in which we've had a decline of 1% or greater. Overall, the next day in the S&P 500 Index has been up by an average of .04% (28 up, 19 down). Again, that's not much of an edge compared to the average gain of .02% (343 up, 278 down) for the entire sample.

Here, too, though we see a pattern. When the S&P is down sharply and volume is concentrated in declining issues (N = 23), the next day averages a loss of -.03% (11 up, 12 down). When the S&P is down sharply and volume is not concentrated in declining issues (N = 24), the next day averages a gain of .11% (17 up, 7 down).

What that means is that sharp declines are more likely to persist the next day if the herd is selling. When the decline is large but selling is not indiscriminate across issues, we're more likely to have a snap back the next day.

Most recently, I've looked at advancing vs. declining volume on an intraday basis in the ES futures. I compute whether each 15-minute time period in the market is advancing vs. declining and compare the volume of advancing to declining periods on a moving one-day basis. As the chart below indicates, concentrations of volume in advancing vs. declining periods have been leading to short-term market reversals.

When volume has been concentrated in advancing or declining periods and the market has been unable to make significant new highs or lows, that's telling us that volume cannot move price. My line of thought at those times is, "If this concentration can't bring us meaningful directional movement, probably nothing will." It makes sense to fade the herd in those circumstances. It's when those concentrations of volume break us out to new extremes that we can ride the herd's coattails in the short run.

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 and a blog of market analytics at www.traderfeed.blogspot.com. His book, Enhancing Trader Performance, is due for publication this fall (Wiley).