Forex Software — Deviating results in backtesting

Re: Deviating results in backtesting

null@example.com (Popov) — Thu, 06 Jul 2023 05:01:23 +0000

> I resolved the issue by setting the commission per lot correctly when using data export.

Good job!

You can also do it in EA Studio in the "Data" -> "Symbol Settings" tab.
Select the desired symbol and change the "Settings source" to "Custom symbol settings".
You can play with the values to see how they affect the backtest.
You can always return back to the original settings.
The symbols with overwritten parameters are marked with an asterisk.

Re: Deviating results in backtesting

null@example.com (Averell) — Fri, 30 Jun 2023 11:51:28 +0000

UPDATE:

Thanks for replying so quick, Popov!

I resolved the issue by setting the commission per lot correct when using data export. Newbie mistake.
All results now correspond between EA Studio and MT.

Wow, what a tool!

Re: Deviating results in backtesting

null@example.com (Popov) — Thu, 29 Jun 2023 04:35:30 +0000

Please share your Expert Advisor, and we will test it.

In the meantime, you can compare the deals between EA Studio and MT.

The easiest way to do it is to use the charts.
In the EA Studio's Indicator Chart, you can go to the previous or next deal using Ctrl + arrow.

You can also use the Journal to see where the trades deviate.

Deviating results in backtesting

null@example.com (Averell) — Wed, 28 Jun 2023 20:40:28 +0000

Hi.

I have uploaded history from ICMarkets to Expert Advisor Studio.
Meaning Expert Advisor Studio and ICMarkets MT5 backtester are using the same data.
However when I apply the strategy to the MT5 backtester I`m not getting the same results as the generator, and mostly only around 50% trades taken. Trading dates are the same.

Two last pair I noticed it are GBPJPY and EURGBP.

Any idea why?