Is This Possible

New comment
 

I backtested an EA with two MT4 Terminal same broker same feed 89% Modeling Quality but each one gives different results that is one shows positive result one shows equity down to zero! my question is -- is this possible is this normal am i doing something wrong or Baktest are just garbage results if that's the case then i would say to every one to stay away from systems or system sellers which just show backtest results.

One thing more one MT4 Terminal shows 490 trades one shows 180 trades.

 
EACAN:
I backtested an EA with two MT4 Terminal same broker same feed 89% Modeling Quality but each one gives different results that is one shows positive result one shows equity down to zero! my question is -- is this possible is this normal am i doing something wrong or Baktest are just garbage results if that's the case then i would say to every one to stay away from systems or system sellers which just show backtest results. One thing more one MT4 Terminal shows 490 trades one shows 180 trades.

It should be something different.

89% is not 90%.

For example, in one Metatrader is not enought data for M5, for an other one - M15 for example.

If you have 89% modeling quality so it means that not enough historivcal data.

 

The other differences are the following.

If you are using MM if your EA so different deposit size will show you the different results (sometimes). And it will be completely different from backtesting without MM.

And as i know some EAs can not be backtested (it was coded to prevent backtesting or to show you very excellent backtesting results which is nothing with reality).

 
newdigital:
It should be something different.

89% is not 90%.

For example, in one Metatrader is not enought data for M5, for an other one - M15 for example.

If you have 89% modeling quality so it means that not enough historivcal data.

Sometimes we may see this differencies: press "File" (MetaTrader's menu), "Open Offline" and check "From" and "To" for both Metatraders.

Besides there is script to find the gap in historical data.

And there is EA to fill the gaps in historial data ( am not posting it because everything is in Russian language inside the code and probable nobody needs this EA).

 

Next Phase!

Thanks Newdigital for covering strategybuilder 101 for me and it helped and now let me tell what i have done.

1st. I downloaded MT4 from metaquotes.net and imported history from AlPARI on tools->options->charts both 999999999 for bars period converted to 5,15,30,60 and tested on H1 with modeling quality of 89.28% and the results were good

Next ( ofcourse i couldn't believe i could write such a devastating piece of software so i downloaded InterbankFX Demo 30 days directly from interbank and did the same above process ) but this time same modeling quality but account zero even if it misses some days from the previous MT4 which takes data from 18th June 2004 and this one 22June2004 the results are completely different i mean what i could conclude is that because of ALPARi DATAFEED ? maybe but then again when i test the moving average EA Test both of them shows same results which led me to believe then something is wrong in my coding anyways thanks ND for informative reply

 

Ok i solved the issue with strategy tester and i get 90% Quality on both platforms with same results as i understand the quality must be 90% not 89.82% but 90% for atleast some reliability i choosed to took 1 year instead of 2 years for both MT4 and it seemed it worked and today it gaved good results live demoing so now i believe there is not that unreliability with strategy tester but one needs to test with atleast 1 year data and must 90% not even 89.94%-82% as i understood.

New comment
Reason: