I can imagine how it is possible to hedge out contango (or backwardation) in the Vix futures through calendar spreads at a ratio. If the contango between months 1 and 2 is 10% and the contango between months 2 and 3 is 5%, couldn't we hedge out the contango to zero by going: Month 2: - 1 (+10 % contango) Month 3: + 2 (-10% contango) This should be possible at any given time if we set up the appropriate ratio. But what exposure to the Vix does that provide? Is the above position essentially long the Vix market, or short? If Vix spikes, are we more likely to get a payoff, or take a loss? The CBOE publishes a crude map of Vix futures beta relative to the spot index. http://www.cboe.com/products/vix-index-volatility/vix-options-and-future Looking at this chart, one month out gives a beta of about 50 (which I assume means the futures contract moves by half the amount of the Vix), while two months out the beta is about 40. That means the above spread should be net long, as the net beta would be +30. Of course, the hedge would destabilize with the passing of time and shifting of the term structure, so adjustments to the ratio would have to be made as necessary to maintain neutrality of contango/backwardation. How does that analysis read? I suppose there are two questions: can we indeed hedge away contango, allowing for expression of a Vix outlook irrespective of the term structure; and upon neutralizing the curve, how would the spread perform in response to Vix fluctuations?
How would you forecast a change in the term structure? And, how would you handle the drag of the steepening curve?
Hey, I'm trying a similar method right now to test a hedge strategy using call ratio spreads on the VIX. Different in the details to your method, but similar desired ends. A lot of the futures stuff is beyond my knowledge, but certainly interested to see what others have to add here. You done any live testing with this strategy? I'm trying verticals, but I'm intrigued by your approach. Certainly open to something that gets rid of the natural decline of the futures. My goal is specifically VIX exposure against potential systemic moves down to hedge other positions that are too expensive to hedge with SPX debit spreads.
I do attempt to speculate on the future shape of the curve and have been successful with spreads taking that object as a focal point. I don't know if it's just my dumb luck, though. I look for lumps and bumps in the curve and bet on a prettier curve, buying low and selling high with the expectation of a smoother curve. Beyond that, I don't try to forecast the curve. I certainly can't predict steepening into backwardation or any major changes. I would not deal with the front month, as that one is wild. From month two and onwards the moves are more measured and take place relatively in tandem, making a spread more viable. That's in my limited experience, anyway.
Note: The link I posted above in the original post was inaccurate. This one seems to be correct: http://www.cboe.com/products/vix-index-volatility/vix-options-and-futures/vix-weeklys
The short answer for what you're trying to achieve is to reverse the calendar (long the near, short deferred), operate across the shorter maturities, ratio the contracts according to recent curve behaviours (beta to spot VIX may not be as reliable these days). Employ a type of time stop: if your expectation or thesis doesn't play out in a determined time period, close out.