We’ve tested 16 simple and effective VIX trading strategies on this blog that, based on all of my readings both academic and in the blogosphere, are broadly representative of how most traders (of the quantitative, swing-trading variety) are trading VIX products.
Even though those strategies are keying off of completely different metrics, ranging from the shape of the futures term-structure to moving average crossovers, most are at their core, really variations on a theme: momentum/trend-following.
Nearly all of the strategies tend to trade long volatility when the VIX is rising, and short when the VIX is flat or falling (as opposed to mean-reversion strategies like this one which are actively betting against the prevailing trend).
That means that all of those strategies are going to agree more often than not. There’s nothing inherently wrong with that. It’s tough to argue with a long XIV/short VXX position when (for example) equities are in an uptrend, VIX futures are strongly contangoed, and a juicy VRP all exist concurrently.
But that also means that judging one strategy versus another is really about understanding the narrow periods of time when these different approaches don’t agree. This also makes designing strategies more difficult as the sample size of truly relevant data is much smaller than it would at first seem.
The table below illustrates how frequently our representative sample of simple VIX trading strategies agreed with one another. Fifteen trading strategies (1) are listed (see legend below table). Find where a given pair of strategies intersect on the table to find out how often their respective positions agreed.
Key: (1) First vs Second Month Futures, (2) VIX vs Front Month Futures, (3) VIX vs 1-Month Constant Maturity, (4) VIX vs VXV Indices, (5) V&M’s VIX:VXV Ratio, (6) DFTB’s StDev Strategy, (7) DFTB’s Spread Strategy, (8) DDN’s Momentum Rotation, (9) DDN’s VRP Strategy, (10) 10/100-Day MA Crossover, (11) TWP’s Quadratic Fit, (12) NAS’s VIX Futures Momentum, (13) S&P 500 50/200-Day Crossovers, (14) Brute Force Optimized VRP, (15) LI’s Bollinger Band Strategy
* * *
Strategies agreed with one another on average 79% of the time. The two individual strategies that disagreed with one another the most often, still agreed 65% of the time.
Again, these strategies are meant to be broadly representative of how most traders are trading VIX products. The key takeaway is that even though these strategies are keying off of completely different metrics, most are really variations of momentum/trend-following, and as such, will agree with one another more often than not. One of the keys to designing VIX trading strategies is better understanding the narrow periods of time when strategies disagree.
* * *
When the strategies that we cover on our blog signal new trades, we include an alert on the daily report sent to subscribers. This is completely unrelated to our own strategy’s signal; it just serves to add a little color to the daily report and allows subscribers to see what other quantitative strategies are saying about the market.
Click to see Volatility Made Simple’s own elegant solution to the VIX ETP puzzle.
Volatility Made Simple
- The TM RSI(2) strategy was excluded from this analysis as it trades far too infrequently to make for an apples-to-apples comparison.
- I took some mathematical liberties when deciding whether two strategies “agreed”. If strategy X was 100% long VIX and strategy Y was 100% short VIX, I counted that as full disagreement. But if strategy X was in cash, and strategy Y was 100% long or short VIX, I counted that as only a half disagreement. I did this to compensate for the fact that some strategies are long/short and some short vol-only.