Forums  > Basics  > Simulated option expected value on long-dated vs selling of shorted dated  
     
Page 1 of 1
Display using:  

contango_and_cash


Total Posts: 141
Joined: Sep 2015
 
Posted: 2021-06-13 12:25
I was having a debate with a friend, and I may be a dumbass here.

The friend, who has traded options profitably for quite some time, said:
"It is better to sell shorter-dated options because of (time-varying) vol and delta".

While I don't disagree with his point, I was arguing that it may equally make sense to sell options that have positive expected value to the seller by way of current pricing (simulation-based pricing).

For this exercise, let's assume the simulation routines are reasonably robust (merton jump, sampling with replacement or copula-based simulation all yield positive expected value results).

I used a hyperbolic example of: "I'd rather sell a 10$ 3-month option than a 0.1$ 1-month option".

Now, of course, the simulation routine I'm suggesting is agnostic to time-to-maturity. I may sell a 5-day option or a 50-day option with equal conviction.

Of course, as time evolves, and odds change and underlying move, this problem continues to reset (and then you wind up in continuous-hedging-world).

In general, I'd be curious to hear how others have approached this problem.

FWIW, I am referring to strictly selling covered-calls for the time being. The problem doesn't require that, but just as a matter of fact.

Previous Thread :: Next Thread 
Page 1 of 1