Stocks quoted in this article:
I often mention that I use the CBOE Volatility Index (VIX) closing greater than 20% above its 10-day simple moving average (SMA) as a trigger point. I also use it as a proxy for declaring VIX overbought, which (to me) is bearish for volatility and bullish for the market on the theory that VIX tends to mean revert.
But, it's important to always note that 20% is somewhat an arbitrary threshold. I personally use it because it's a nice round number and it gives a reasonable number of signals, maybe three or four per year, typically. Clearly, if you lower to 15% or 10% above the 10-day SMA, you'll get more signals, and 25% or 30% you'll get fewer. And, it's entirely possible either case suits you better (it's also entirely possible that you don't view VIX as a contra-tell, but that's a subject for another day).
Anyway, my point is that VIX action relative to its moving average is best viewed as a continuum, rather than a series of hard-and-fast actionable signals. If you believe overbought VIX will mean-revert, then it's relatively better to go long the market/short volatility when VIX is 15% above its 10-day SMA than 10%, relatively better at 20% than 15%, etc.
I bring this up thanks to a friend who asked me the other day how close VIX was to being overbought. Given that on the surface, absolutely nothing had happened in the market to that point, I figured VIX wouldn't say much. But, when I hit it up, it was around the low ebb of Monday's market decline, and VIX was 15% above the 10-day. If you take a more aggressive approach to VIX mean-reversion plays, that's actionable.
And I have to admit, I was pretty surprised. Yes, the market had drifted a bit until Tuesday -- it was really a tiny drift. The peak close in SPDR S&P 500 ETF (SPY) was 200.50 on Sept. 3. The low close was Monday at 198.98. That was a whopping 0.75% drop over the course of nine trading days.
That's barely more than a rounding error. Yet VIX made a real move, at least in percentage terms. It closed at 12.36 on Sept. 3, and 14.12 on Monday. That's about a 12.5% pop. On average, VIX moves about six times the pace of SPY (i.e. -- a 1% SPY loss would translate to about a 6% VIX pop). It's very variable, though. It's more of an exponential relationship than a linear one. That is, if SPY drops 2% today, VIX would likely lift way more than 12% -- probably something like 30%.
But, a 0.75% move over the course of two weeks? That would generally translate to almost nothing in VIX. Yet we see a decent move and a decent-size stretch from its moving average.
Subjectively, that was a bit of an overreaction to a nothing move. Maybe it's predicting we do see a more sizable sell-off at some point. But, another way to look at it is that traders/investors are over-preparing for the next sell-off, and will have some ammo ready to fade the move -- which often means the move won't happen.
Disclaimer: Mr. Warner's opinions expressed above do not necessarily represent the views of Schaeffer's Investment Research.