How long is the ideal period of backtesting that considered to be able provide a valid result?
And particularly for the manual backtesting, how much the time needed to finished the manual backtesting process for that time period?
Quite honestly for me it depends on the amount of work involved in the backtesting.
if its a strategy that might offer an entry 2 or 3 times a week off the daily charts I would always go back at least 3 months. This is an initial test to understand what Iām testing and crucially for me it must show a win rate better than 50%. I know strategies with low win rates can be profitable - if I find one I would use it but I would not waste any time backtesting a low win rate strategy which I had devised.
After the initial test, I would usually forward-test either in demo or with mini positions another 3 months.
1 Like