Big VIX Declines In A Down Trend - A Short Trade Set Up on SPY 11-7-16

I posted recently that there was likely to be a sharp rebound in SPY based on the multiple days in a row below the lower bollinger band.  We are seeing that today with SPY up nearly 2%.  So the question now is really about how far the rebound is likely to go and what is the next tradable move.

Currently around noon EST today the VIX is down about 17% from last session's levels.  When I backtested SPY (going back to Sept 1995) for performance following days where the VIX dropped more than 15%, the results showed negative skews looking forward all the way out to 6 months.  There were 39 instances in this scan.

This means greater MAX losses on average compared to MAX gains over the time period.  The skew was most negative over the first couple weeks.  So this should give us some perspective that these types of big moves in the VIX are in markets that are in the midst of choppy sell offs or abrupt rebounds and possibly short covering rallies.

I went further and looked at times when the VIX dropped over 15% and also SPY gapped up more than 1% as it did today.  There were 19 instances in this scan.  The result was even more negative in the skew.  Additionally on a CLOSING basis, all forward time frames from 1 day out to beyond 2 weeks (and even at the 1, 2, and 3 month forward times) showed negative average returns from the signal day.

What this indicates is that these rallies tended to make most of their gains off the bat, and then stocks began to decline again.  Of course there were some instances where stocks continued to rally nicely.  But I am referring to the averages of all instances here.  The strongest negative skew looking ahead was at 2 weeks forward.

I also went ahead to look at time where the MACD (daily and weekly) were both down when the gap up and big VIX drop occurred.  This left 15 instances and eliminated 4 more.  This scan fits our current market where the big jump is coming right off a low as opposed to a market that has already begun to turn up or is in a consolidation.

Here again the skew is even more negative.  And the peak of that negative skew is at 2 weeks forward.  Closing returns are again sharply negative on average looking ahead in coming days.  At 7 trading days ahead, 11 out of the 15 instances closed negative relative to the signal day.

Now in looking at past instances most of these situations were pretty choppy.  This makes the option play of buying puts less ideal.  But the expected values in shorting the stock or using an inverse ETF appear to be very good on average.

The best play from my perspective appears to be to short SPY (or buy inverse ETF) with a paired limit exit gain of 9.25% or even 15% and then a stop loss of a corresponding amount.  Obviously it would be almost  incomprehensible for a 10% gain SPY over the next couple weeks.  But as a simple strategy that allows for an expected value of around 3% on shorting SPY.  This would equate to closer to 10% expected value on a 3x inverse ETF like SPXU.

From my current perspective it seems that I would want to be short before the election results came out if stocks hold up into tomorrow.  Based on these results I will determine a strategy to enter an inverse position.  Possibly  enter half a short position today at the close, and then wait to enter a second half position based on future action or an actual short term technical topping signal like a 15 minute MACD divergence.

Pete