Let's say I sell an American-style option. Either put or a call, it doesn't matter. I plan to keep the short option to expiration. I'm right about both direction and timing, and the option will actually expire worthless. Provided it is allowed to expire. However, the market moves against me in between the time I sell the option, and its expiration date. How can I figure out the odds of an early assignment ? What data should I be looking at to calculate those odds ? How far the underlying has moved from the strike ? The volatility ? How much time is left ? All of the above ? How often does this happen to you ? The particular cases I am worried about are for weekly options on SPY. Specifically selling cash-covered puts, or selling covered calls, on monday morning, with a friday expiration, all of which would be out of the money at the time of writing. Also, if you get assigned early, do all your contracts tend to get assigned at once, or partially ?
Risk of early exercise if option is ITM and time premium gone. Also occcurs if there's a pending dividend and the time premium is less than the dividend (see conversons). If time involved, must factor in carry cost. It's rare but I've had assignments when there was still a lot of time premium remaining. That's a good thing if you can unload the UL w/o equity loss and resell the option position.
As an option seller, early assignment is almost always a "good" thing. The other side of your trade decided for whatever reason they wanted to throw away what was left of the time premium in the option, and make a gift of it to you.
Unless it is you who got screwed. If you are assigned on a short call due to a dividend then there is no gift for you.
As part of my education process, I sit in on trading rooms from time to time. Basically this is a learning experience conducted by coaches. The other day I was in an option TR and one of the students asked your very question. The coach asked if any of us students had ever been assigned prior to expiration Friday. While a few of us had been assigned, not one of us had ever been assigned early.
Well, presumably you sold your call *expecting* either early assignment, or the drop in stock price that would follow the dividend. You wouldn't have been "screwed". Just as an aside, I get early assigned in commodity futures options regularly.
Yes, but you can get screwed overnight by price movement in your newly acquired position in the UL. As an aside, I once had some naked ITM MSO calls with 2 pts of time premium exercised early. Stock cooperated in the AM and I covered the stock and resold the calls. That's a rare event but even it can happen.
Or you can make money in your newly acquired position in the UL (as in your example). In the name of market efficiency, let's just agree the UL was pricing risk/reward correctly the previous day, and you didn't lose anything (in expected terms) from early assignment. On the other hand, being handed what's left of theta for free is absolutely a free gift, in expected terms.
spindr0, Thanks. Let's assume there is no pending dividend before the option expiration to take that out of the equation. Has anyone studied this to factor the probabilities of early assignments ? I'm looking primarily to sell weekly OOM SPY puts with 5 days remaining. About 98% out of the money. There is never much time involved, these options start at 9 days to expiration. My calculations show that even in extreme 2008-2009 bear market, the odds of assignment at expiration were under 10%. But I would like to figure out the odds of early assignment. If the stock never moves below strike before expiration, I can tell that the odds of early assignment of the put will be 0, as the intrinsic value will always be 0. That's the simplest - and best case. But the UL price usually moves up and down below or above strike before put expiration. Those are the cases where I'm a bit at a loss to assign probabilities of early assignment. In most cases, those early assignments would not be terribly negative, except for the commissions involved, if I continue to hold the UL until the time the put option would have expired worthless (ie. until the end of the week). But I still want to know the odds Has no statistician studied this ?
Thanks. Interesting data point. Was that for puts or calls, or for both ? What was the duration involved ?