Just a ma crossover. Ma 200, or 180, 220, 223 etc it doesn't really matter. I think every line is possible. It's the risk and monymanagement which can make or break it. And being creative in how to make the losers less losers. Just my experience and this is position trading, larger timeframes only.
I think you could also look at the homo- or heteroscedasticity of your returns, it could reveal interesting things on how to make your method more adaptable (if you are interested)
You need to analyze in more detail where the gains originate from. Even with a 50/50 win/loss ratio if your gains on average exceed the average loss then you are ahead. But this has to originate from somewhere. If you don't have any edge whatsoever then as you hinted at, your recent profile is purely random. But if your profitable positions for whatever reason run further while cutting losses early this could be because of some sort of edge you are exploiting, even unknowingly. Note here that risk management (the fact you decide to cut losses early and let gain run) is not an edge in itself ever. I strongly suggest you run more extensive back tests to investigate. For example, if the market exhibits momentum with statistical relevance and you are able to exploit such moment (meaning you got an edge) then you can figure out at what level you cut losses and by how much you let profitable positions run. But you need that edge. Without edge you got nothing. Not in trading, not in gambling.
YOu know what I learned after testing out my bot using all different settings: Averaging in will kill you eventually You can make random entry work if have multiple entry tries and you close your trade at the end of of the day. OR trading is the purest trading method known to mankind
Thanks. That makes me feel better. But how do I know it is not random due to small sample size? Sometimes when on a roll I made good money playing the roulette betting on red.