How much backtesting is enough?

If we talk about the strategy, it takes me about a month to evaluate its performance. At the same time, most often I test it on a demo, plus I evaluate how well it has worked in the past. During this time, I test it on various assets in order to understand which ones it is best suited for. If we are talking about some automated earning system, it seems to me that it takes at least a year, although it is relative, we change strategies periodically, because the market throws up new regularities and everything changes, so here and there. That is, today the system can be quite working, and in 2021 it will lose all sense. What if you don't have time to catch the moment when it happens and you lose your capital?

Hi guys, nice thread.. I would like to share my thoughts.

I see the point of people who are arguing that market is changing and that you need 1000 trades in oder to have a good idea about the profitability of a strategy, I see the point of people who argue that backtest is useless and I see the point of people that say only forward test is good.

My thought here is that market is cyclic, that's a fact, so basically there is always a trending or ranging situation. Markets changes their behavior with time but the trending or ranging situation are always there.

So let's say you watch a monthly timeframe, you see again trending and ranging periods, huge periods because we are on a monthly chart. Now here's the point, do you think that a trend on H1 with ranging on monthly chart would be the same than a trend always on H1 but with a trending monthly chart? I don't think so.

The point is this, I don't know what timeframe the strategy you are backtesting is using, let's assume it's a strategy that work on H1 tf, I wouldn't care about having 1000 samples trades, I wouldn't care about having 200 years of historical data, that would be an exaggeration and a waste of time.

I would only see how the strategy adapted during the changes of the market in the past and if it was able to remain profitable, if a strategy adapted well to different market major changes to me it is a valid strategy. Please remember that in these days people trade S/R more or less they used to do 30 years ago.

Market is cyclic, it repeat itself... Just like humans do.

Hope I've explained myself and that you found this interesting. Cheers!