Stocks quoted in this article:
Got this on Twitter the other day, in regards to my recent look at CBOE Volatility Index (VIX) futures:
Hi Adam. Interesting article. You seem to be saying that VIX is always a good short on any pop. Don't you think the low vol regime will end?
To which I replied:
I've never said on any pop, it's ultimately mean reverting, that's all
And he shot back:
True. My takeaway was that selling VIX is a sure win. But someday vol will actually rise and that strategy will blow up.
It then occurred to me that I better clarify my opinion on VIX, and tradable VIX.
Selling VIX (presumably futures, though I suppose it could mean volatility in general) into pops is NOT a sure thing. Nothing is ever a sure thing. Everything I ever write, every data study I ever provide, every chart I ever post all are intended to highlight probabilities, never certainties. Certainties don't exist.
When VIX makes an abrupt lift, it generally will extend enough above a mean where we may seek to derive a signal. I use 20% above the 10-day simple moving average as a general trigger point. But you can use more, or less, or Bollinger Bands, or whatever --it's more about the concept than anything else.
In my humble opinion, that generally becomes a time to look for mean reversion. That is, VIX will stall and stocks (which are presumably in decline) will stabilize.
But importantly, again, that's a probability, not EVER a sure thing. Over the course of time, data suggests that fading VIX-plosions works well over most time frames. There are often exceptions however, as some sell-offs persist, and crashes (however you define one) essentially all start from already-oversold markets.
In more statistical-speak, fading VIX pops will give you a positive expected gain, but you may have to withstand some scary drawdowns along the way.
As to whether I think the current low-volatility regime will end, of course I do. I know not when it ends; we can only answer that in hindsight. Long-term regimes tend to last about four to six years, and we're inexactly a few years into one. So perhaps in a year or so, VIX starts trending up.
It's important to remember, though, that long-term trend changes are, by definition, very rare. And I'm absolutely not going to be the one to declare some future VIX pop THE start of the new regime. I'll let someone else be a hero on that; I'd rather miss the first boat and adjust my strategy going forward from there.
And finally, as to VIX futures, that perpetual upslope in the term structure has proven wrong for the better part of five years now. That lasting VIX rally pretty much hasn't happened. It will someday. But just know that there's no real reason why the curve should slope up as much as it still does. The local, low-volatility regime mean of VIX is in the 16-17 range, depending on how far back you want to go. In my humble opinion, that's a fair level for an out-month VIX future -- not the 18, 19, and even 20 we've seen over the last few years.
But again, that's going to change at some point, too. Someone will absolutely buy a VIX future in the high teens and ride it and sell for a big profit someday. I'm just not prepared or inclined to guess when that happens.
Disclaimer: Mr. Warner's opinions expressed above do not necessarily represent the views of Schaeffer's Investment Research.