Thread: How many trades until a system is proven, not just lucky?

A system I'm using has been profitable 3 weeks in a row. However, I have just traded it 50 times. When do you understand okay this system does work?

Originally Posted by ;

If your system works on D1 it must work on M5 also, not identically showing the same outcomes, but since the market is fractal-like, it should be somewhat similar.

The one thing that varies between time frames are the ranges.

Originally Posted by ;

quoteTo people posting that 5000 or 10000 trades are necessary, given the assumptions in this post, 1000 trades provides a very high degree of assurance, and somewhere between 1000 and 2000 trades the probabilities hit the asymptote, which means that you have a near certain statistical assurance that your win% will be quite close to the backtest. So raising the trade count above 2000 does not offer any more assurance from a statistical standpoint than 2000 trades (because maximum statistical assurance has already been reached). However, as a caveat, I have...

This is quite good to understand. I'll be honest and say I really didn't think too much about the 10,000 trade amount, but I had been thinking several thousand trades per market for its 5 markets. It would seem if you split the 10,000 trades by 5 markets that you get 2000 trades per market, which will be in the end of your 1000-2000 trade amount. I'm more of a discretionary trader (not certain if I can call myself a trader), however it's good to understand that after 2000 trades the gap gets fairly insignificant.

Additionally, you bring up a great point that the backtest has to be pure without any over / curve matching of the trades.

Thank You,
LVG

Originally Posted by ;

A system can produce major profit annually and fail miserably the following year. It will not create the machine poor or good. Most progr fall into one of two egories = tendency or range. Therefore the goal of a trader shouldn't be to loe a system. It ought to be to handle their transactions that more profit is generated than reduction in a manner and in accore with current and changing market conditions.

Hello
Thank you for your information. I never thought about this before the goal of a trader shouldn't be to loe a system. It ought to be to handle their transactions that more profit is generated than reduction in a manner and in accore with current and changing market conditions. When you said that managing the transactions were more important than finding a system, you referred to cash management? Can you please clarify more or inform me where I can find out more about this?
Thank you.

Originally Posted by ;

quote Hi Thanks for your information. I never thought about this before: Hence the goal of a trader should not be to find a system. It ought to be to manage their trades in a manner that profit is produced than loss in a manner and in accore together with current and changing market conditions. When you said that managing the trades were more important than finding a system, you referred to money management? Can you please explain more or tell me where I can find out more about this? Thank you.

I am not referring to money management in the sense of just how much to risk: xpercent per trade, for instance. I am referring to that instrument to exchange (expose yourself to) and when.