I'm writing this prior to the market close, so it may not precisely match conditions at the close. Nonetheless, I think the underlying concept may have utility.
Basically, I decided to stay short the market through the day as long as two conditions held: 1) emerging markets (EEM) were underperforming the U.S. market (SPY); and 2) VIX continued to make higher highs through the day. My reasoning was that, as long as the market was trading on fear--avoiding volatility and risk--we were likely to see a flight from equities overall.
Do fearful markets have a short-term bearish bias? I went back to March, 2003 (N = 815 trading days) and found 76 occasions in which SPY was down between 1.0-1.5%. Two days later, SPY was up by an average .26% (39 up, 37 down).
When we split this sample in half, however, based on VIX changes, a pattern becomes clear. When VIX rises strongly during the SPY decline (as has happened today), the next two days in SPY average a subnormal gain of .01% (15 up, 23 down). When VIX only rises modestly--or falls--during a decline in SPY, the next two days in SPY average a gain of .52% (24 up, 14 down).
It appears that the bearishness which leads to an extreme bidding up of options premiums during a decline tends to carry over in the short run, producing subnormal returns. When the market declines by an equivalent amount but options players do not bid up premiums aggressively, we're much more likely to get a reversal.
The willingness of market participants to take on risk may be an important variable in determining the likelihood of continuation of market moves.