Over the last 5 days, $VIX has gained more than 25%. Is a large gain in the volatility index bullish or bearish for the S&P 500 over the next 50 days?
The Rules:
- Buy $SPY at close when $VIX has gained more than 25% over the last 5 days.
- Sell $SPY at the close X days later.
- No commission or slippage included.
- All available $SPY history used.
The Results:
Some Additional Stats:
Next Day Winning Percentage: 58.93%
5 Day Winning Percentage: 69.62%
Median Trade After 50 Days: 2.74%
Average Trade After 50 Day: 2.26%
Number of Setups: 112
Number of Trades Held 50 Days: 44
While my breadth indicators are not quite signaling that a bounce is imminent, this study has yielded bullish results.
$VIX closed at 19.48 and hasn’t closed above 20 since July, 2012. Let’s see what happens over the next couple of days. Another $VIX spike and an extreme breadth reading will make for a great short-term bottom.
Interesting… I’d like to see this study completed using spreads in volatility… that is when imp vol in certain indices is higher relative to others w/ higher beta’s (e.g. VIX / RVX spread)
I don’t have enough data for options to test that, I don’t think. Unless there is a way to do it using the VIX/RVX. If you could be more specific I might be able to work it out.
Quite interesting that these VIX spikes have predictive ability so far out…though the 5 day limit seems optimal from a tradability viewpoint.
Here’s some stats from other symbols: http://qusma.com/wp-content/uploads/2012/12/vix-spike-stats-per-market.png
And a parameter stability check to see how things react if we change the minimum VIX % change and lookback period: http://qusma.com/wp-content/uploads/2012/12/vix-spike-parameter-stability-1024×419.png
Great blog Qusma! I appreciate your comments. Interesting that this effect is stable across so many other ETFs.
Isn’t this study dangerous as vix rising 25% also happened in october 2008 I presume?
Dangerous? I guess if, because of this study, you went 200% long with no plan for an exit, then yeah, it could be dangerous.
Every setup will have winning and losing trades. The edge, and from where it is derived, must be understood.
In this instance, yes, the $VIX rose more than 25% in 2008. But that is an anomaly. Will it happen again? Sure. Will it happen tomorrow? It could. Odds are that it won’t. You gotta play the odds.