Page 1 of 1

unrealistic mt4 backtests with real tickdata and 99.90% modeling quality ( MT4)

Posted: Sat Dec 24, 2022 3:27 pm
by xzerax
can someone explaine this please ? the ea isnt this great but backtest is making it look great !
NOTE : the ea is winning this week and its my first week trading it .. but this backtest doesnt make since and is not realistic .. so can tickdata and modeling quality be misleading on mt4 ?

Re: unrealistic mt4 backtests with real tickdata and 99.90% modeling quality ( MT4)

Posted: Sat Dec 24, 2022 3:31 pm
by xzerax
example

Re: unrealistic mt4 backtests with real tickdata and 99.90% modeling quality ( MT4)

Posted: Sat Dec 24, 2022 3:59 pm
by Chickenspicy
xzerax wrote: Sat Dec 24, 2022 3:27 pm can someone explaine this please ? the ea isnt this great but backtest is making it look great !
NOTE : the ea is winning this week and its my first week trading it .. but this backtest doesnt make since and is not realistic .. so can tickdata and modeling quality be misleading on mt4 ?
Image
I had a lot of issues like this, forexflexea is a reputable one that’s probably the only one I’d ever pick, it costs though

Re: unrealistic mt4 backtests with real tickdata and 99.90% modeling quality ( MT4)

Posted: Sat Dec 24, 2022 4:09 pm
by xzerax
chickensword wrote: Sat Dec 24, 2022 3:59 pm I had a lot of issues like this, forexflexea is a reputable one that’s probably the only one I’d ever pick, it costs though
can you explaine ?
this is my own code in the pick it my ea .. but the thing is its working and im making money ... but the backtest just seems too good to be true .

Re: unrealistic mt4 backtests with real tickdata and 99.90% modeling quality ( MT4)

Posted: Sat Dec 24, 2022 4:39 pm
by Chickenspicy
xzerax wrote: Sat Dec 24, 2022 4:09 pm can you explaine ?
this is my own code in the pick it my ea .. but the thing is its working and im making money ... but the backtest just seems too good to be true .
¯\_(ツ)_/¯
Well I never actually coded an Ea but I’ve gotten a lot of mixed results before so I just stopped digging because of different symbols and dates & money management
, if it so happens the results are too good to be true that’s what I would recommend,
Spread test , slippage test, news test, different pairs and dates , all things that can affect the outcomes
I would just demo account maybe 4 pairs for a month but I haven’t been around eas in a while you probably have more information then I do

Re: unrealistic mt4 backtests with real tickdata and 99.90% modeling quality ( MT4)

Posted: Sat Dec 24, 2022 4:54 pm
by Chickenspicy
xzerax wrote: Sat Dec 24, 2022 4:09 pm can you explaine ?
this is my own code in the pick it my ea .. but the thing is its working and im making money ... but the backtest just seems too good to be true .
Or live test with very little money just to see how real market acts

Re: unrealistic mt4 backtests with real tickdata and 99.90% modeling quality ( MT4)

Posted: Sat Dec 24, 2022 5:06 pm
by xzerax
chickensword wrote: Sat Dec 24, 2022 4:54 pm Or live test with very little money just to see how real market acts
good idea .. will post updates after 2 weeks