Adam Warner’s “Just Because the VIX is Low, Doesn’t Mean It’s Time to Buy”

This is a follow up to Adam Warner’s excellent post “Just Because the VIX is Low, Doesn’t Mean It’s Time to Buy”.

My takeaway from Adam’s post is that when the VIX is especially low (below 15), as it is now, it leads investors to take long VIX positions (ex. long VXX) because surely, the VIX can only go up from here, right? But even at these depressed levels, VIX futures still tend to overestimate the future realized VIX, making most of those long VIX positions losers.

Here I put some numbers behind Adam’s thoughts.


The graph above shows VIX futures (30-day constant maturity) minus the subsequent 30-day realized VIX. Positive values indicate that VIX futures overestimated the future realized VIX, benefiting short VIX positions (ex. long XIV or short VXX). Negative values indicate futures underestimated the realized VIX, benefiting long VIX positions (ex. long VXX).

To match Adam’s post, I’ve shaded red those periods when the spot VIX was above 15.

That’s a lot of data to show something very simple…

Regardless of whether the VIX is low (below 15) or not, VIX futures have overestimated the future realized VIX a bit more than 70% of the time. To Adam’s point, even when the VIX is especially low, VIX futures still tend to overcompensate for future volatility.

What differs when the VIX is either low or not, is the magnitude with which VIX futures tend to over/underestimate.

When the VIX was below 15, the median difference between VIX futures and the future realized VIX was 1.0, with a maximum value of 4.9 and minimum of -8.5.

When the VIX was above 15? A median of 2.7, with a maximum of 21.1 and minimum of -52.3.

What does that mean in plain English?

VIX futures will tend to overestimate the future realized VIX about an equal % of the time regardless of whether the VIX is high or low, but when the VIX is high, that tendency becomes much more volatile, with VIX futures either over- or underestimating by a much greater margin.

At higher VIX levels, shorting the VIX (ex. long XIV or short VXX) to capture VIX futures’ usual premium over the realized VIX becomes much more profitable on average, but much more dangerous for any one single trade.

A big thank you to Adam, one of my favorite VIX bloggers, for his thoughts.

Click to see Volatility Made Simple’s own elegant solution to the VIX ETP puzzle.

Good Trading,
Volatility Made Simple

Wonk note: For more on the relationship between volatility, the VIX, and VIX futures, I highly recommend our post Four Graphs to Rule Them All.

Posted in Volatility Mechanics.