Wednesday, August 23, 2006

Market Psychology: Why It Is Important

We are accustomed to thinking about the psychology of the individual trader, but the marketplace has its own psychology that makes a significant difference in its price behavior over the short-term. The best measures of market psychology are not polls of trader sentiment, as what traders say and what they do can be quite discrepant. Rather, we want to track what traders are *actually* doing in the marketplace.

Readers of this blog know that I track several measures of market psychology, including the NYSE TICK (number of stocks trading at their offer price minus those trading at their bids), the volume of ES futures trading at the offer vs. bid, and the put-call ratio. I have dealt with the TICK and the volume of futures at offer/bid in recent blog entries; let's look at the put/call ratio for an example of why market psychology is important.

In general, I've found that the put/call ratio does make a difference in modeling future results. I use the equity put/call ratio to eliminate the bias of hedge trading among index options. The past two days in the S&P 500 Index (SPY) have been flat from the open of day one to the close of the second day. The two-day equity put/call ratio, however, has been quite high at .92. (Thanks to Jason Goepfert of the Sentimentrader site for pointing out high put buying in the current market). In fact, of the 66 flat two-day periods in SPY that we've had since 2004 (periods that have neither risen nor fallen more than .10%), the present one has the highest equity put/call ratio of all.

In general, since 2004, flat two day periods have tended to resolve to the upside. This fits with my Micropsychology Modeler results posted to today's Trading Psychology Weblog, which found that periods of low trending among stocks have also had favorable short-term results going forward. Specifically, SPY has been up three days later by an average of .21% (41 up, 25 down), notably stronger than the average three-day gain of .075 for the entire sample (N = 660 trading days).

If we break the sample of flat two-day periods roughly in half, however, based on equity put/call readings, we see a significant pattern. When the put/call ratio has been above .70 during a flat two day period, the market has been up two days later by an average of .37% (20 up, 12 down). When the put/call ratio has been below .70 during a flat two-day period, the market has been up two days later by an average of only .01% (16 up, 18 down).

Indeed, when the put/call ratio during a flat two-day period has been above .80 (N = 8), the market has been up the next day and two days after on seven of those occasions.

What that tells us is that market psychology matters. When traders are bearish (buying a high proportion of puts to calls) and cannot push the market lower, it's telling us something important about the market's latent strength. Think of a flat market as a dead-even tug-of-war and the put-call ratio as an indication of who is putting all their resources into the rope-pulling. If you put all your effort into pulling a rope and can only get a standstill, it's only a matter of time before fatigue sets in and you're the one getting tugged.

=========

Additional analysis posted 7:00 AM CT: Tempering the above finding is the finding that two-day flat periods in which total equity option volume is below the 40-day average perform worse in the near term than two-day flat periods in which total equity option volume is relatively high. Specifically, when we've had a two-day flat period and low total option volume (N = 36), the next day in SPY has averaged a loss of -.08% (13 up, 23 down). When we've had a flat two-day period and high relative option volume (N = 30), SPY has averaged a gain the next day of .19% (19 up, 11 down). While the ratio of put volume to call volume tells us something about sentiment, the total option volume tells us about the speculative interest of options traders. Low speculative interest during a flat period, such as we've seen in the past two days, has been associated with subnormal returns in the short run.

The natural question to ask is, "What has happened when the put/call ratio is high in an environment of low speculative interest?" This has happened during 23 two-day periods since 2004, and returns have been subnormal the next day (average change -.03%; 9 up, 14 down), but better over two days (average change .15%; 13 up, 10 down). Five of the six very high put/call ratio two-day periods (over .80) were up two days later.

Bottom line: Bullish indication two days out, not necessarily for today. In terms of market psychology, sentiment is important, but so is the willingness of traders to speculate. Total option volume is as important as its distribution between calls and puts.