I agree with not shortening biotechs. I have Industry!=BIOTECH in the buy rules in one of my short models. Also don’t short the BESTOGA stocks: BEER, SPIRITS, TOBACCO, and GAMING.
The model below is short Russell3000 stocks with market cap>700-million and is permanently hedged long with SSO at 75% of equity. It can’t always find 5 stocks to short, so currently only short 3 stocks, as one can see.
And also from Jan-2015 model performance is extraordinary, showing a return of 80% for the year.
Unfortunately this model cannot be featured as a smart alpha model, because shorts are not allowed.
Here is a combo of my two short models. Interesting that from Jan-2014 onward the combo makes money again. Reason is that the S&P500 was carried by the FANG stocks, and the broader market did not participate much. So the models could find stocks to short. (FANG stands for Facebook, Amazon, Netflix and Google.)
The second combo shows performance of a combo with the same two short models, but both of them hedged with long 100% SPY. Note the low draw-down for the hedged combo and the high annualized return.
Current market conditions are uncertain. A hedged short model should perform well now.
I’m like the idea of adding a short model to a book just as a hedge. I think maybe the simplest method to effectively execute might be an inverse sector rotation model that gets long into an ultra short secter ETF whenever your favorite market timing indicater kicks in. Whether it’s tech in 2000, financials in 2008 or energy today, there always seems to be one secter driving the race to the bottom during a significant downturn.
