If you're a covered call writer willing to sell your stock then it makes sense, if possible, to time your call sales around earnings since the premium is larger due to IV expansion. The same holds true for a short put seller looking to acquire stock at lower than current price. Larger credit = lower purchase price. Selling spreads and condors also benefits from IV expansion because the nearer to the money strikes increase more thereby generating a larger net credit. Selling naked options pre EA is problematic because moves can be sizable and with a fixed potential profit versus a potentially much larger loss from an outlier on the other side - one big loss can wipe out many hard earned gains. Long strangles and straddles held through an EA is a bit of the reverse. They suffer dramatically from IV contraction so you'll have many small losers and an occasional big winner. It's a reasonable idea to buy these a week or two before earnings to capitalize on the IV expansion (sell before the EA) but you still need some price move to compensate for double sided time decay since IV expansion won't keep up with options hurtling towards parity as expiration gets closer. Apart from the covered calls and short puts for stock acquisition, all of these have the same thing in common - you need to have some sort of edge (I didn't) because there's always a scenario that defeats you, sometimes large, sometimes small. There's no free lunch with post EA volatility crush. Years ago I traded a lot of ratioed double diagonals for EAs. I looked for a set up where my post EA IV guesstimate modeled a 2:1 R/R profit graph (EA guess based on previous 3 quarters of the IV graph at IVolatility as well as looking at far months IV). The risk graph was W shaped so no move or a big move was profitable. But the two troughs of the W were not and that meant taking the loss or defending, hoping to get closer to break even. I got away from this because it involved combining earnings dates with IV (McMillan's weekly IV stats at his web site) and then looking them up one by one and then modeling anything that I recognized visually to be in in the ball park. Too much time spent for a lot of singles and occasionally, a lot of defense. These days, I limit earnings plays more to stocks that I own or trade.
Well, like anything, it’s only partially true. Yes, you can and will lose big on some trades, but you can diversify that problem away. Selling the earnings straddles is a statistical strategy that favors a smaller player. If you can afford to trade only a few contracts per event, you have like two thousands companies in your universe and get to play in about eight thousand events per year. That’s a pretty reasonable set of events. The crux is to develop a model that predicts the magnitude of the earnings move better than the market does. That’s where the alpha comes from.
What's the advantage of that? Lower IV in a later week/month? You still have a similar risk to reward ratio.
For a market maker it does make sense. However, if you are a net payer of transaction costs, simply trading the straddle is the way to go - otherwise, you are paying 3x the costs which erodes your expectation a fair bit. Also, chances are that you will get shafted on the Vega after the event.
Well, I am pretty sure MMs have done ok even without this strategy There are people who make a living trading event vol (I used to do that myself in certain classes of stocks) but spreading it in any way is a market maker type of trade.