JW Trading Journal : 2007-03-31 : Further OEX Analysis

Unfortunately, further analysis from the past few days has came to some undesirable findings.Whereas the OEX gives me much more upside potential, it's a double edge sword. In order to protect the downside not to exceed the maximum drawdown that I can tolerate, I HAVE TO set stop loss points if I want to use OEX instead of SPY or DIA as the underlying instrument.

This, however, goes against the core idea of the model itself which is aiming at high accuracy of the "outcome" when closing the position. Higher number of positive outcome enables constant compounding. The swing of OEX would stop me out a lot more times than the number of incidents that a non-stop model would have a 100% loss.

Using DIA 121 CALL for example. Current ask is 3.50 and delta is 0.808, which I like. It will expire in 20 days. The trade from non-stop model would only have a POTENTIAL 100% loss when the Dow suddenly having a total of more than 430 point drop within "1 to 4" trading days AND if not able to recover in the coming 20 days before it expires. Between the day that it cut-through my threshold and day #20 (expiration), I still have the flexibility to close the position out at any recovery after the big drops.

From the cumulative $ return point of view, because the fact that I was stopped out so many times from the OEX model, it wasn't compounding enough times to have the same outcome as the non-stop-non-OEX model. If I tweak it to have a comparable cumulative $ return, then I'll be forced to accept the maximum draw-down of more than 70% of the total capital, which is probably unacceptable to anybody. Why so high of a max drawn down? It is because the total capital is now being divided into fewer trades in order to "make more money back from the bigger winner" when it is proved to be a good trade at the end. For lack of better expression, "you can't eat like a bird and shit like an elephant".

Anyway long story short, when considering my own personal risk-reward on the per trade and cumulative $ return basis, I probably will stick with DIA and SPY options for a while for the original model. But I do feel OEX have its advantages as I was modeling and testing it. It's just that it will need an entirely different game plan than what I have for now.

Overall I have learn a lot in the OEX modeling process. Besides, I did make that money with OEX Put. LOL! :)