Single EA strategy test not matching strategy optimization results

New comment
 
I ran the EA strategy optimization and then selected some of the individual results to run a single test. The single test results did not match the optimization results. The biggest difference was the drawdown% which is much higher with the single test. The number of trades was noticeably lower in the optimization results. I've seen a couple threads about this but none had a solution. Can someone help?
 
Christopher M A Holmes:
I ran the EA strategy optimization and then selected some of the individual results to run a single test. The single test results did not match the optimization results. The biggest difference was the drawdown% which is much higher with the single test. The number of trades was noticeably lower in the optimization results. I've seen a couple threads about this but none had a solution. Can someone help?

None has a solution because nobody ever provides all the needed information to reproduce the issue.

I can show you a lot of matching results between an optimization and single tests results, so it's not a general issue. You need to investigate, it could be a lot of things. If you need help, provide details.

 
Alain Verleyen #:

None has a solution because nobody ever provides all the needed information to reproduce the issue.

I can show you a lot of matching results between an optimization and single tests results, so it's not a general issue. You need to investigate, it could be a lot of things. If you need help, provide details.

Thanks for your response, Alain.

The problem is I have no idea where to even begin investigating, short of deleting all the EA code and starting from scratch. The only thing I've done so far is clear the cache, which did not change anything. Otherwise, it's not intuitive where this problem might be coming from. All the input parameters in the single test are exactly the same as in the optimizer.  I think myself and others with this problem have no idea where to even look because the optimizer is a black box (there are no logs or charts to reference). There is no documentation to even provide possibilities of where the differences might occur.

 
Christopher M A Holmes #:

Thanks for your response, Alain.

The problem is I have no idea where to even begin investigating, short of deleting all the EA code and starting from scratch. The only thing I've done so far is clear the cache, which did not change anything. Otherwise, it's not intuitive where this problem might be coming from. All the input parameters in the single test are exactly the same as in the optimizer.  I think myself and others with this problem have no idea where to even look because the optimizer is a black box (there are no logs or charts to reference). There is no documentation to even provide possibilities of where the differences might occur.

You still didn't give any details, not sure what you are expecting...

What test did you do ? What tester settings, what kind of optimization...

You can post the Tester logs, for both optimization and the single tests.

New comment