NOTE: There will be an update of the Trading Psychology Weblog tonight.
Thursday was a very broad rally in a market that had been sharply lower for the month. Specifically, the S&P 500 Index gained over 2% on the day after having been down about 4.8% during the previous 20 sessions.
I went back to 1996 (N = 2629 trading days) to see what happens after we get a 2%+ rally after a 20-day period in which the market has been down over 4%. This occurred 40 times. Five days later, the S&P 500 Index was up by an average of .55% (24 up, 16 down), stronger than the average five-day gain of .16% (1434 up, 1195 down) for the entire sample. Twenty-three of the occasions took place during the 2000-2002 bear market, and, overall, the strong up days after 20 days of weakness tended to be followed by considerable volatility.
The ratio of up volume to down volume on Thursday was huge: over 22:1. That is the strongest concentration of volume we've seen since 1996. Interestingly, when the market rallied strongly after 20 days of weakness *and* volume was highly concentrated in the advancing stocks (N = 14), the next day in the S&P averaged a loss of -.41% (6 up, 8 down), considerably weaker than normal. Similarly, when we've had a strong rally in a down market and the day's Arms Index (TRIN) has been less than .40 (N = 16), the next two days in the S&P have averaged a loss of -.32% (6 up, 10 down).
In summary, strong up days in a down market, in which volume is highly concentrated in the advancing issues, actually yield subnormal returns in the short run. Moreover, many of these big up days during falling markets occurred during bear markets and did not signal an end to the bear. While strong up days *in general* tend to produce favorable near-term results, panic buying/short-covering after a period of bearishness has not produced favorable returns over the next couple of days.