Re: Watch HowardCohodas Trade Index Options Credit Spreads Quote:
Originally Posted by DionysusToast OK - so for each $8000 return we need $110,000 in the market - but as our win:loss ratio would be $8000:$110,000 - we'd need a larger account than just $110,000 as one loss would wipe us out.
As you say, you are never all in - only 80-90% in the market.
So - how large an account would we need here ? We can re-build the capital from a loss in approximately 14 months but in that 14 months, we'd need to not spend any money as it needs to go to re-building capital.
Obviously as the required account size grows, then the percentage return on the account would inevitably fall.
From the discussions so far, I would guess your probability calculation shows you that 2 consecutive losses would be a rare event. What's your maximum expectation of consecutive losses ? |
Just to review, the $110,000 account size to earn $8,000 per month is based on the following: - 10.5% account size growth in successful months
- 1 in 10 chance of an unsuccessful month
- 30% loss of entire account in an unsuccessful month (more on this later)
- No estimate of the probability of consecutive monthly losses
The 10.5% upper bound is based on my actual performance. My testing results show a higher return is possible. Since I'm in preproduction (small money) in some areas, the actual results may be somewhat diminished. And four months is not yet statistically significant.
The 1 in 10 chance for an unsuccessful month is below what I saw in testing, however it is more reflective in the Probability of Touching level I choose. I choose the more conservative number in my estimates to account for possible failures not anticipated.
The choice of 30% loss of the entire account rather than the funds at risk actually assumes two consecutive months of failure. That is because I generally enter a trade right after the expiration of current months options when the funds quarantined by margin requirements become available. The new spread is entered around 60 days before expiration in order to catch the time-value decay curve before the knee. Thus I am leapfrogging the following month's options, which I'm already in, for the one after that. (Gosh, that's a messy explanation. I'll have to work on making easier to understand)
I feel this is an overly conservative approach to estimating draw-down as I never saw two consecutive monthly losses in testing, but I like to stay on the safe side of things when discussing the possibilities.
I know that I did not answer some of your questions directly because I did not attack the problems as you stated them. If you feel more explanation is needed, ask away, and I'll do my best. |