Back Tests on Days When the VIX Increases More Than 20%

Click on Stats to Enlarge

The VIX is an implied volatility index, based on actual option prices trading in the market.  So a big jump in the VIX shows a real money indication of anticipated increased future volatility.

Monday was an interesting day in the VIX as the VIX jumped about 23%.  That is not very common, but also is not the really interesting thing.  What is more interesting is that going back to September 1995, where I have data to backtest, the VIX has never increased over 20% on a day where SPY was note down at least 1%.  However, I looked for days when SPY was down less than 2% to give a sensible comparison, and the results for the next 1 and 2 weeks were both consistently bullish with good profit factors in both options and stocks.

I ran another simple scan which looked at VIX up over 20% but the SPY close was not below the lower bollinger band.  About half of the days had SPY closing below the lower bollinger band.  And the scan results are shown above for this scan.  Notice the very consistent wins here - over 90% with correspondingly high expected values and Kelly Bet %.

So relative to Tuesday's open in SPY June 17th 208 calls, the option opened at 2.75.  And so a limit gain of 40% would be about 3.85 as a limit order.  There is not much difference at 4 or 5 days ahead, so choosing this week's expiration seems sensible.

If moving out to next week's expiration, the optimal limit order on the 208 call would be 80% gain.  And the EV is only about 13% and Kelly bet is 16%

The instances where the next day, in this case Tuesday, closed down, 5 out of 8 closed up more than 1% over the following 3 days.  This suggests there may be a tendency to rebound into the end of this week.  So buying here could still be justified with a smaller position.

Pete