I usually back test 3 or 4 strategy's for each equity, but I think Im partial to double calendars and atm calendars for the probability and risk reward.
Ratio spreads. If it goes opposite, your loss is limited or even a credit. Only downside is neutral action.
Quadruple bachelor dragonfly hedged by married puts on binary options on annual winter Groundhog event and double hedged by a certified local palm reader (female) - former voodoo pro.
Credit Spreads using SPY. Deep OTM spreads, Bid/Ask spread <.02, Delta <.10, >1.5 SD, >.25 per spread, risk <500. Considering that "earnings" happen almost on a daily basis throughout the trading year I have found it profitable/practical to not focus on one symbol but instead focus on the market as a whole. That's my favorite strategy, day in and day out.
Long Fly, although my FB put fly 129/124/119 is in jeopardy as AH price crashed through all the strikes.
you dont need backtesting for any earning play (gambles) it is hardly relevant you can do it before earning / through earning you can use calendar / fly / calendarized fly / double or triple calendar / etc doesnt matter i am sure someone can get lucky with this type of trade -- not a good and scalable business model though