What timeframe do you use for trading? Were 4-6 years meant for a strategy with parameters which are never re-tuned? Or self-tuned during strategy's life?
Even using 10-12 years for backtest, you could see when equity curve went up and when down. So if it went up only once 8 years ago and then went sideways - that's a bad sign i.m.h.o. Chances are high, that there was a big change in those years and, yes, the strategy "doesn't really work" in a considerable part of backtest window. That's why I beleive that the window whould'n be too big for optimization; and that OOS test is a must.
Time span choosen for optimisation matters a lot. E.Chan:
The most basic safeguard against data-snooping bias is to ensure that you have a sufficient amount of backtest data relative to the number of free parameters you want to optimize. As a rule of thumb, let’s assume that the number of data points needed for optimizing your parameters is equal to 252 times the number of free parameters your model has. (This assumption of proportionality is not based on any survey of the vast statistical literature, but purely on experience.) So, for example, let’s assume you have a daily trading model with three parameters. Then you should have at least three years’ worth of backtest data with daily prices.
