Different results from strategy tester between different builds of MT5.

 

Hi everyone,

I run some backtests for some assets a few months ago with mt5 build 1940.

When i run the same tests with same EA and parameters using the same brokers in a newer build (1978) i get different results for some of the assets.


More specifically i run backtests with these assets: AUDUSD, EURUSD, GBPUSD, USDCHF, USDJPY, XAUUSD, OIL, BTCUSD using every tick based on real ticks in order to test the behavior of my EA.

I get different results from my original tests with these assets: USDCHF, USDJPY, XAUUSD and OIL .


I reinstalled mt5 and run the same tests in build 1940 again for these 4 assets and i got the previous results for the 3 of them.

The tests with OIL procured the same results as in the latest build instead the ones that i got with build 1940 the first time.


Does anyone know what is the cause for all that?

Is there a chance that my EA is the problem?


Thank you for your help!

 

You say tests are different: Result_1940 != Result_1978.

To understand, you need to provide complete information to reproduce the situation.

 
Looks to me like it's a bug in the new version, maybe something slipped in the regression tests.