Message Font: Serif | Sans-Serif

No. of Recommendations: 5
Elan,

I can understand why someone would choose to ignore the Maximum Drawdown, but why would you consider it bad information if assessing a screen? It would seem to provide some real world information of how painful a screen can be.

Eric took the words right out of my mouth -
http://boards.fool.com/Message.asp?mid=17743901

To illustrate the point, I ran a little experiment with Excel's random number generator. I simulated 150 monthly returns of a screen with CAGR=42 and GSD=35. (The respective monthly average return is 3% and standard deviation is 10%). I then identified the maximum drawdown. I repeated the same exercise ten times. Here are the max drawdowns from each of the ten runs -
-48
-39
-35
-35
-58
-50
-31
-38
-37
-39

As you can see the numbers are all over the place. But they were sampled from identical distributions. The 'screen' with -31 is statistically identical to the one with -58. Their chances of reaching a particular 'max drawdown' in the future are identical.

Elan