Bears who counted on a breakdown of a seeming head-and-shoulders pattern (another eloquent reminder of the folly of believing that shapes on charts dictate global capital flows) have been taken to the woodshed with a high momentum rise in stocks. Specifically, we've seen back to back strong days in the Demand/Supply indicator, which tracks the proportion of stocks closing above the volatility envelopes surrounding their short-term moving averages to those closing below their envelopes. Tuesday saw a near 10:1 ratio of Demand to Supply (upside momentum to downside momentum) and Wednesday was about 15:1.
When we've had back to back days of 5:1 (or greater) Demand to Supply going back to September, 2002 (when I began archiving these data), the next five days in the S&P 500 Index (SPY) have averaged a considerable loss of -1.73% (6 up, 12 down). That is quite a bit weaker than the average five day change of .06% (928 up, 767 down) for the remainder of the sample.
By the time we've had consecutive high momentum days, it appears that--in the short run--the bulls are all in and we've tended to give back some of those gains. Indeed, after a single day of greater than 10:1 momentum, the next four trading days have averaged a loss of -1.22% (10 up, 17 down). Chasing highs after several days of strength, overall, has not been a winning strategy in the short run.
A few people have commented that today seems unusual in that stocks are strongly higher, but the VIX has also moved up from yesterday's close. Normally, option-related volatility tends to drop as stocks rise: since 1990, the correlation between one-day changes in the S&P 500 Index (SPY) and VIX has been a statistically significant -.68.
I went to the all-purpose database and found only 13 occasions since 1990 in which SPY has been up more than 2% on the day and VIX has closed higher on the day. Two days later, SPY has averaged a change of -.32% (5 up, 8 down), compared with the average change of -.23% (87 up, 79 down) for the remainder of instances when SPY rose by 2% or more with a falling VIX.
That's a small sample and not a statistically significant difference, but it's certainly an indication that such short-term strength does not confer a bullish advantage in the near term.