I was doing some analysis on just how successful it would be to buy a straddle during earnings events. There's lots of literature that has been written and most of it seems to point to a bloated IV before the earnings announcement that never lives up to the worry, but I wanted to take a look for myself.
Using data from http://www.optionslam.com I was able to look at the last earnings cycle from April/May of 2014. They provide a straddle price of the closest expiration option, at the close before earnings, at the open and at the close of the day following.
If you had of bought a straddle and then sold at the open, here's the results over all 1090 stocks that they list,
The results are pretty telling, whilst you could have done well on some individual trades, the majority would have lost money as the stock didn't move enough and when combined with the IV crush, meant it was just not worth the premium paid for the straddle.
There is also data on end of the day which indicates that the stocks continue to move further in at least one direction, but not significantly.
The chart of the % returns on a bought straddle looks like:
The interesting thing from the graph is that there really isn't that many large losing trades, a couple of 400%, a couple of 200% and about 10 more than 100%. I was expecting to see a lot more outliers or black swans, which is the benefit of a bought straddle as the buyer is expecting or hoping for a large move in any direction.
So ... if it is a losing proposition by buying the straddle, then how about selling the straddle. Since there's always an option trader on the other side of every trade then someone must be selling these straddles that lost money for the straddle buyers.
My understanding of selling a straddle is that you would be taking on unlimited risk since the stock could in theory go to zero or climb to the clouds. But the data from this earnings cycle simply doesn't indicate that is the case. The biggest loss was a 400%, which would be tough on anyone's account, but not if only a small portion of each trade was allocated to each straddle sold.
The win/loss on selling the straddle is 4 out of every 5 which would help (psychologically) as we all love winning trades. Another argument in favour of this strategy would be the very short term holding period, basically overnight from the close to the following day's open. This would allow a number of positions to be put on at the same time, helping to average out the inevitable large losing trades.
My concern would be the holding/margin requirements, would one need a very large bank in order for something like this to work ?
I have begun some initial investigations into buying a really far out of the money strangle, i.e. something roughly at the strikes where the straddle would lose 200%. If the cost of this was 5% or less then it would be a hit but might make the strategy tradeable from a smaller account.
Also, the open price on options seem to have a very large spread until things settle down for the trading day. Is it realistic that the open prices on optionslam would represent the true price ? I'm guessing that they must be close given the closing day price is not that different on average.
Maybe this earnings cycle was abnormally quiet in terms of large moves. Although even in a high VIX environment, the straddles should become relatively more expensive, offsetting or even making the strategy more profitable.
Anything else I'm missing with this strategy, aside from the guts to trust the numbers for not just this earnings cycle but future ones !? I'm sure someone must have thought of this already so I'd be interested to hear if anyone else has crunched some numbers on it...
Disclosure: The author has no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.