One that isn't curve fitted, which most optimized strats are IMO. However, I'll grant that codes can help speed up the process of finding one. But my bone of contention is that many think that coding is synonymous with trading. Running a bunch of scripts won't magically give you a money-making ATM.
Your opinion is based on your (very limited) own knowledge and/or experience.
Your so called "curve fitting" confirms this. You see coding as curve fitting, which it is not always, and in case of a good strategy it even has nothing to do with curve fitting.
I build a strategy with use of Excel and after that it was coded. I never tried to optimize parameters to get the best result, as that is curve fitting. I used it in a completely different way. A way from which you probable don't even know the existence. The basis is logical thinking.
The result is that my system is profitable in any kind of market (trending, sideways or anything else) although returns are not always as good. Trending markets are always more profitable than sideways markets. My system tries to detect the kind of "trending" we have and automatically adapts the system to that.
I can also use it in futures, crypto's, stocks or forex. This excludes already clearly the curve fitting as all these markets are different.
I also use the same system for longer than a decade with good results, again confirming that it is not curve fitted. It is a dynamical system, so self adapting on the results of the calculations. I can even change the parameters within certain limits without any, or at least a very small impact on the long term performance. And this impact can be positive as well as negative because some trades will be more profitable while others can be less profitable.
A curve fitted system will never work for a long time, and it needs to be curve fitted over and over again. Curve fitting is never ending as it incorporates data that is noise instead of good and relevant data. It does not filter out the noise.
Last edited: