MT5 strategy tester - low number of passes in the forward assessment phase - page 2

 
Stanislav Korotky #:
I suppose you can start with a self-made (semi-automatic) analysis of matching settings (confirmed hits) between top 10% of full and top 25% of genetic optimization. I think it will give you enough material to estimage overfitted status. Also I doubt MQ will change the procedure in near future.

Well the goal is not to perform an extra fast genetic optimization just because there's an issue in the slow complete process.

And even, if the 10% in the slow complete opt is overfitted it's just useless to spend time optimizing this way and to analyze the results, as they wouldn't bring anything relevant. Then let's just focus on several fast gen opts lol.

All in all, I'm sorry but users expect something else from a slow complete optimization and related forward assessment. So yeah, it's mainly about a setting/option allowing to control this percentage.