• Registered: 2017-08-04
  • Posts: 30

Topic: Difference between FXSB and MT4 backtest

i was backtesting a strategy on s and p 500. it is a simple mean reversion strategy which enters on lower band of bb and exit at swing high.. there is a huge difference in performance between fxsb and mt4. I see some of the trade results match to the 2nd decimal while others dont.

on further analysis , i see that the FXSB sis closing position every 4 hours (chart time frame) while MT4 is closing the position in between as well. am missing some setting here. the symbol and data was imported using the mt4 bridge.

have attached the strategy and journals would be delighted if experience members take a look.

Post's attachments

Trading.7z 83.01 kb, 9 downloads since 2018-01-18 

You don't have the permssions to download the attachments of this post.
  • Popov
  • Popov
  • Lead Developer
  • Offline
  • From: Bulgaria
  • Registered: 2006-11-30
  • Posts: 8,024

Re: Difference between FXSB and MT4 backtest

The produced expert advisors must show very near performance in the MT Tester.

Please examine the work of the custom indicator (Recent Swing High Low) you are using. It may produce different values for MT and FSB Pro.

Popov's Website

  • Registered: 2017-08-04
  • Posts: 30

Re: Difference between FXSB and MT4 backtest

Hi Popov i changed the strategy to daily and used a N bar exit along with bollinger band exit. ..but still i see extra trades or different closing in mt4 (not for all trades)..i imported the data and instrument using the fx bridge.

Post's attachments

Trading.7z 42.01 kb, 6 downloads since 2018-01-19 

You don't have the permssions to download the attachments of this post.
  • Blaiserboy
  • Blaiserboy
  • Junior Part-Time Aspiring Space Cadet
  • Offline
  • From: Toronto
  • Registered: 2010-09-06
  • Posts: 2,849

Re: Difference between FXSB and MT4 backtest

I suspect the data to be different.

Blaiserboy's Website

5 Reply by gs.sudharshan (edited by gs.sudharshan 2018-01-20 16:04:13)

  • Registered: 2017-08-04
  • Posts: 30

Re: Difference between FXSB and MT4 backtest

Hi Blaiserboy - i was searching the forum to troubleshoot the issue and found Popov's long on fractal on GBP strategy on daily gbp chart and compared the results for 2016 and found there is a difference in the time at which MT4 backtester is closing the trade.
For instace, there is a trade on 24th June 2016 which has been closed by MT4 on 27th June (when it is supposed to close on end of day).Am using pepperstone demo account. Stragely, the opening point is matching to the decimal or very smal difference and number of trades are matching as well.

Long story short i still dont know what is causing the issue.

Post's attachments

upper_frac.7z 100.47 kb, 3 downloads since 2018-01-20 

You don't have the permssions to download the attachments of this post.
  • Popov
  • Popov
  • Lead Developer
  • Offline
  • From: Bulgaria
  • Registered: 2006-11-30
  • Posts: 8,024

Re: Difference between FXSB and MT4 backtest

For instace, there is a trade on 24th June 2016 which has been closed by MT4 on 27th June (when it is supposed to close on end of day).

MetaTrader has a very big problem with Bar Closing. We fight with that every day (while their stuff doesn't care). This problem is discussed in the forum in details. In order to make the best for our users, we created EA Studio to close positions at Bar Open. We will do it in FSB pro in the next major update.

Popov's Website

  • Registered: 2017-08-04
  • Posts: 30

Re: Difference between FXSB and MT4 backtest

Thanks Popov.  I checked old threads and found that you have mentioned this is not just a backtesting issue but happens during actual trading as well. Any suggestions to minimise the impact of this.. ( i dont mind missing one trade here and one trade there).

Also, is there is list of must read forum threads which would help new users coming up to speed

  • Blaiserboy
  • Blaiserboy
  • Junior Part-Time Aspiring Space Cadet
  • Offline
  • From: Toronto
  • Registered: 2010-09-06
  • Posts: 2,849

Re: Difference between FXSB and MT4 backtest

All I can suggest is that you forward test... some of the strategies will work ok and some won't, I have experienced this myself quite a few times.

MT4 brokers are very industrious people, they do not reveal all of their abilities to us retail traders. You have no way of knowing as to how they may be influencing your trades.

Blaiserboy's Website

9 Reply by GD (edited by GD 2018-02-05 04:18:32)

  • GD
  • GD
  • Licensed Member
  • Offline
  • Registered: 2015-05-22
  • Posts: 444

Re: Difference between FXSB and MT4 backtest

Sometimes I have the feeling that differences exist because some problem of mq4 customized indicators
Sometimes big differences exist and sometimes do not. yesterday I found big differences by using QE.

I think to use control prices only O, H, L is the best.
I agree with Blaiser Boy about tricks from Brokers and use of higher TFs

Popov is using a few seconds a limit to EA as CLOSE.
It sounds enough good approach but finally what to believe?

  • footon
  • footon
  • ◄≡≡≡►
  • Offline
  • Registered: 2009-08-17
  • Posts: 2,969

Re: Difference between FXSB and MT4 backtest

GD, I expect a report with details how QQE has misperformed in a backtest.

11 Reply by GD (edited by GD 2018-02-05 14:59:16)

  • GD
  • GD
  • Licensed Member
  • Offline
  • Registered: 2015-05-22
  • Posts: 444

Re: Difference between FXSB and MT4 backtest

Hi footon

I tried to check different FSB made EAs in mq4 using  MT4 tester in control mode in order to make a portfolio
I had not any good coincidence of MT4 tester results with results of FSB Pro. I used H1 and 10,000 points
My broker did not have enough data for lower time-frames which could be a reason.
So, every EA did not performed well but most the one using QE.
Accidentally i did not keep this EA. I went to sleep...

  • footon
  • footon
  • ◄≡≡≡►
  • Offline
  • Registered: 2009-08-17
  • Posts: 2,969

Re: Difference between FXSB and MT4 backtest

Performance differences between 2 backtests can be due to a variety of reasons. It's my opinion that backtesting twice is not useful, there's no extra validation from that to gain. If strats works similarly in a backtest and in live trading, then that's an important aspect to look at.

QQE can be set up using exponential smoothing and if periods are large enough, then it is possible to have slight mismatches in one parameter reading because the EA is not using similar amount of bars compared to a backtest. That's all I can say for now.