DislikedSo.. I started to analyze in-sample vs out-sample data. This is the first version of anything but there are already some interesting data here.
What I have basically done is I sorted all strategies by in-sample fitness which is basically the best reward vs risk ratio I have found. I then split the data to 20 parts and printed some of the values for in-sample and out-sample.Ignored
To select best strategy I would sort strategies by Kelly ratio, which incorporates both win-rate and risk-reward ratio.
The best strategy by that criterion generally also has the best return to draw-down ratio.