How do you know you are not overfitting your strategies
I don’t know. But I spend a lot of time trying to find reasons for my strategies not to perform. The nice thing about automation is that I have plenty of time to worry about improving the strategy development process as opposed to worrying about specific trades.
There are a few ways I go about it none of them bullet proof:
1. The strategy has to work on baskets on random stocks, various stock universes
2. I try to minimize the number of variables and backtest on walk-forward out of sample rolling windows where I try not to pick any parameters
3. If I do optimization i look for consistent intervals and try not to pick the peaks so to speak.
4. I look for consistent results over various trading regimes.
If you have any other suggestions I’d be glad to hear them.