Trader's Ruin

Status
Not open for further replies.
Quote from virtualmoney:

I mean express edge in terms of C,X & U of the inequality equation above, and then setting it >0,..., what does it all mean?
It doesn't mean anything to me. Why would you want to do that? :confused:

To repeat:

RoR = ((1-edge)/(1+edge))^((C-X)/U)

where edge = (pW - q|L|)/U

and U = (pW² + qL²)^½


RoR < 1 if edge > 0 and C > X
 
Quote from virtualmoney:

+1...but how to determine how much?

Since I'm an engineer, not a mathematician - I use the back test and try to validate it in live trading. My system breaks down at about 8x. 3x seems to be the limit of my stomach, however, and leaves a good margin of safety.

It seems that other trend followers (Eckhardt comes to mind) do it this way, as well. He has some interesting comments about this subject in Schwager's New Market Wizards.

:D
 
Quote from Wide Tailz:
Since I'm an engineer, not a mathematician - I use the back test and try to validate it in live trading. My system breaks down at about 8x. 3x seems to be the limit of my stomach, however, and leaves a good margin of safety.
It seems that other trend followers (Eckhardt comes to mind) do it this way, as well. He has some interesting comments about this subject in Schwager's New Market Wizards.
:D

Thanks for the info. Here's some interesting reading material

http://www.forexfactory.com/showthread.php?t=275566
http://forex.knowforfree.com/improve-forex-strategies-with-consecutive-losses/
http://www.optionetics.com/market/articles/21296
 

I've tested scaling down position size based on consecutive losers and it actually hurt the performance. For some strange reason, my algo likes more leverage with more losers, probably because it just likes leverge. Again, it breaks down at triple the leverage I intend to use, so this may be why. I tried all kinds of creative money management and it had the highest annual return as a percentage of largest drawdown with a constant leverage. I believe this is the ratio that determines how much leverage is possible.

I'm sitting at about 2.5. Classic trend systems seem to be around .5-.8
 
Status
Not open for further replies.
Back
Top