The way we make our money in option selling is from time decay. The premium of the option we sell is fully comprised of "time value" and as this erodes, our realized profit increases. So determining when to sell options such that the time decay is maximized is very important to us.
Every textbook on options will have this graph. If you naively based your option selling on this graph, you would say it's best to sell options with 1 to 2 months until expiration. It looks like most of the time value is still in the option and it doesn't begin to drop off rapidly until after this point.
It doesn't help when the authors of The Complete Guide To Option Selling, who are supposedly the experts on option selling, reproduce this graph in their book and on their website
here, and say "Notice that the value decays the fastest during the last 30-60 days of the option's life". They then go on to advocate selling at 3 - 5 months to expiry, which confuses readers since it seems to contradict what they just said and what the graph shows... but it's all because
the graph is a lie!
What these graphs fail to tell you is that they are showing the typical behaviour of an *** At-The-Money *** option. We are selling Far-Out-of-They-Money options, and the difference is huge!
As a simple gut-check that the graph must be wrong, look at any far out-of-the-money option with 30 days remaining until expiry, and you'll see that it is practically worthless. What was once worth maybe $400 with 90 days remaining is lucky to be worth $50 with 30 days remaining. Anyone who has sold options knows that there is no value left by the time you hit the 30 day mark.
So what does a typical out-of-the-money option time decay chart look like? Here's one I put together a few months ago, showing an 82 cent put option strike and 106 cent call option strike for the Canadian Dollar when it was around 94 cents. I simply looked at these strikes in different option months to get an idea of how the price would change over time, assuming the futures price and sentiment didn't change.
Well well well... the curve has completely changed! You could infer from this graph that the farther from expiry, the faster the time value decays -- the exact opposite of what the standard option time decay chart implied!
Now before anyone goes and starts selling options that are 1 year from expiry, please note that:
- the farther away from expiry, the more time the commodity's fundamentals have to shift and the more time the futures contract has to move towards to your strike price. E.g. the CAD could easily move to 0.82 over a full year. Thus, you should be selling farther out-of-the-money the farther you move out in time, which means you won't get as much premium, so it may not be worth it.
- If there is a move against your position, these long-out options often start to behave a little more like At-The-Money options, because if people are thinking there is a shift in the trend, the premium could remain high for a long time. You can't easily wait out these moves. E.g. from 12 months left to 10 months left, an option premium may not change that much depending on how the future's sentiment has changed. However, with a 3-month-to-expiry option, the option almost has to really start losing it's value within 2 months, and you'll have a 3 month maximum time to wait it out.
- The graphs don't always look like this. For example, with Sugar, the futures contracts show that the prediction is for sugar to drop drastically in price in about a year. So the call options that are a year out are actually not much different in price than the ones that are 4 months out.
Based on the trades I've done, the optimal time-to-expiry to trade seems to be around 90 to 130 days to expiry.
I seem to be able to find good value in this range at strikes that seem reasonably safe. With less time than this, the premium is already dropping too much and commissions will eat up a bigger percentage of the premium, or I have to trade too close to the money or overposition myself to get a good premium. With more time than this, it can be hard/frustrating to wait out moves against my position. Also, since more people are trading the closer months, it is harder to trade the farther out months with less liquidity and crazier bid/ask spreads.
As a side note, one rule of thumb I try to follow is that for any option I'm about to sell, I should expect it to halve its value in 1 month. So If I'm looking at a June option, I check out the May contract, and it should currently be at about half the premium of June's at the same strike price. This is just one of many rules of thumbs I follow to help me confirm that the trade I'm making is a good one.
