To quote my very first boss, RIP: "There are priorities in this business. First, don't go to jail. Second, don't get fired. Third, don't lose money. Lastly, make money"
Certainly a choice as long as the weeklies cover the earnings at least the day after. Look at the last Amazon surprise. Doesn't happen often, but a huge surprise. If you had just sold premium to capture the big vol. in the front month options, you experienced a ton of pain.
If this was the only event, sure it hurts, but as long as you controlled the size (read - was not stupid) and have plenty of other events coming up, it's a blip on the screen. If that was your "big day", then yeah, it would be mighty painful.
In my opinion, holding ANY option position (long or short) through earnings is pure gambling. Earnings are too unpredictable. Yes, there are companies that most of the time move more than the implied move - however, all it takes is one cycle where it doesn't move, and the straddle can lose 70-90% easily. We implement a strategy of buying a straddle few days before earnings and selling just before earnings are released. I described the strategy in How We Trade Straddle Option Strategy article. We have been doing it for the last 6 years with very consistent results.
Hmm. When you say "we have been doing it" do you mean that you actually transacting or that you have been paper-trading it at mid? You are paying a round trip in terms of transaction costs, you are delta neutral and your holding period is fairly small. To benefit, the earnings move has to be really mis-priced for you to reap any vega benefits. Let's go through some back of the envelop calculations. You have a stock that costs a 100 dollars and you have an option that's expiring in 10 days. Let's assume that ambient volatility on that stock is 2% and the currently priced-in earnings move is 10% - that makes implied volatility of 59% annualized and premium of 9.4. If the earnings are under-priced by 2% (a fairly rare occurrence) your expect the implied vol to reprice to 74% (accounting for lower number of ambient vol days) so you sell your straddle for 10.55 (some decay included). Sounds nice, but if you introduce transaction cost of just 2 annualized vols, your lose half your expected P&L - you buy the straddle for 9.75 and sell it for 10.30. Considering that you also might lose some of your vega due to the spot move etc, I don't think the trade works that well.
In general what I have noticed is that while the IV does increase semi-consistently prior to earnings, this is mostly a result of the option pricing model factoring in time decay especially with little time to expiration. In other words, while the IV does increase prior to earnings (and you say you don't hold through earnings so let's ignore the actual earnings release move) the PREMIUM in those options hold steady, containing in their implied move the impact of the anticipated binary event. TL/DR, IV increases but for the most part option premium does not increase prior to earnings (at least in any predictable manner)
Is that a sound strategy? Why don't they suggest buying lottery tickets...provided you buy at least one big winner?
@sle When I say "we have been doing it" that means live trading, not paper trading. Our full track record - https://steadyoptions.com/performance/. Also, ss I mentioned, we always sell before earnings, earnings move is not relevant for this strategy. "Considering that you also might lose some of your vega due to the spot move etc" - actually if the spot moves, it increases the gains, not loses them. @TradingDemystified You are correct that IV increase does not always outpace the negative theta. However, in many cases it does - especially if you enter at favorable prices (compared t previous cycle prices). But what is more important that the strategy can make money not only from vega but also from gamma. If the stock moves, many times it's enough to book gains and re-position on new strikes. Sometimes we were able to book gains 2-3 times on the same stock in the 7-10 days period before earnings. And most important, the risk of those trades is very low. Even if IV increase is not enough, it is very rare to lose more than 7-10%. And winning ratio is around 75-80%.
That's not necessarily true. Your PnL will be path dependent based on the direction on the moves and your hedging. That's true even if the stock has realized higher than ambient implied (the whole premise of "free gamma before earnings"). If you are stating that you are making most of the money on vega (*) the fact that you are losing vega on your strikes can not be positive. Unless I am missing something, like you are saying that the trade is to make money on gamma and vega gains are secondary. (*) I dispute the whole vega ramp-up story in a modern market-maker environment. Every market maker has a statistical model of earnings and other recurrent events and adds special weights to these days. Yes, "nominal" implied volatility does increase before the earnings but that's just because ambient days are falling out. E.g. if you have a 10 day option with a 10% event day and 9 2% ambient days it will have an implied roughly equal to 59 and next day, as long as event is still upcoming, the implied will be a bit over 61. That's simply because the event is dominating the implied. You see that especially strongly in biotech stocks where the term structure is sometimes pricing a single 20-40% daily move.