Aurix Exchange

Differencce between backtest and live demo test!

shtcwl

New member
Hey guys,

I am new to FX trading and the forum. So far this forum is the best one I have experienced and many thanks to the members.

I have been testing an EA on my live demo account for two days and the EA only placed one trade. HOwever, just now I run the back test on the same EA with same setting and there are multiple trades in the same period of time!!

Why there are such big difference in backtesting and live demo testing? Is there any special setting I could have missed? Or if it is always like this how much confidence I can put on the backtesting results?

Can experienced MT4 users share your experience with backtesting and live demo testing?

Many thanks

SHTCWL
 

saxon

Member
It is a complicated answer, but it is essentially that backtest data is not exactly the same as real live feed data. Also, in backtest, the EA can open orders instantly, but it takes 2-3 seconds in real life to full an order.

The best way to test an EA is to forward test a demo account - it is still not 100% accurate, but it is close to real performance.
 

shtcwl

New member
Thanks Saxon,

So, should I assume that the ES is running correctly on the chart as long as there is a smily face on the top right hand side?

SHTCWL
 
Thanks Saxon,

So, should I assume that the ES is running correctly on the chart as long as there is a smily face on the top right hand side?

SHTCWL

yes you should, but for errors check the 'experts' and the 'journal' tab - post any error messages here for an explanation. Also check that you've uploaded any relevant indicators to the correct location also - or it won't trade.

Is it any particular EA we can help you with (believe me.... most of us have tried them all so far!)
:)
 

shtcwl

New member
yes you should, but for errors check the 'experts' and the 'journal' tab - post any error messages here for an explanation. Also check that you've uploaded any relevant indicators to the correct location also - or it won't trade.

Is it any particular EA we can help you with (believe me.... most of us have tried them all so far!)
:)


oh, it is the Forex Combo. I have placed the "Kernel32.dll" under the Library folder and I did not notice the requirement for any customized indicators in the code. I hope there is no hidden tricks as the EA runs great on backtest without errors.

from your experience, in live test does the ES trade as often as it did in backtest?

Thanks

SHTCWL
 
oh, it is the Forex Combo. I have placed the "Kernel32.dll" under the Library folder and I did not notice the requirement for any customized indicators in the code. I hope there is no hidden tricks as the EA runs great on backtest without errors.

from your experience, in live test does the ES trade as often as it did in backtest?

Thanks

SHTCWL

Typically, results will vary between brokers and also between the live and demo accounts - often demo accounts use ECN / STP / Raw fulfilment so that you often get better results and tighter spreads from your demo experience (to entice you into getting a live account) - this happens with most of them I'm afraid as they are mostly unregulated.

I use the same bots on several FX broker accounts and the difference in the results is frightening, what's profitable on one is terrible on another. Which broker are you using?
 

shtcwl

New member
Typically, results will vary between brokers and also between the live and demo accounts - often demo accounts use ECN / STP / Raw fulfilment so that you often get better results and tighter spreads from your demo experience (to entice you into getting a live account) - this happens with most of them I'm afraid as they are mostly unregulated.

I use the same bots on several FX broker accounts and the difference in the results is frightening, what's profitable on one is terrible on another. Which broker are you using?

Really!?? I am testing the ES on an Alpari demo account. I heard that Alpari's review is not that great and I hope the backtest on its historical data is not total waste.

I used to trade stocks where the historical data is pretty acurate and never thought about the huge difference among FX brokers.

Do you have any good FX brokers to recommend?

Thanks

shtcwl
 
Really!?? I am testing the ES on an Alpari demo account. I heard that Alpari's review is not that great and I hope the backtest on its historical data is not total waste.

I used to trade stocks where the historical data is pretty acurate and never thought about the huge difference among FX brokers.

Do you have any good FX brokers to recommend?

Thanks

shtcwl

I believe that Alpari NZ is better historical data than Alpari UK, as it doesn't use metaquotes (which is full of holes) - Alpari default demo is ECN so be careful when you go live - especially if you're using scalpers! If your profit target is 20pips or so you should be fine (in the case of the Combo strategies) I use IamFX, Thinkforex, FXPrimus and have blown accounts with Alpari and AVA - and I'm in no rush to re-upp either of these. I have a soft spot for Iamfx - but it's commission based - I think that ThinkForex is fair (but watch your leverage requirement) and like FXPrimus because they give me 500:1 leverage, although I think they have gone for my stops a couple of times (maybe I'm just paranoid!)

Hope that helps! :) - Don't forget the 'thanks' button ;)
 
H

hermes

UK brokers are exactly so screwed up as US brokers (Active Trades,UK). EA never runs on live account so successfully like in backwards testings. Demo testings are also not good. Best live micro/mini account. Disconnect the EA and open new chart and attach it again,that sometimes wake up the sleeping EA. Broker's machination is another possibility. EAs don't like fixed spreads. Variable are better or speak to your lawyer!
Hermes
 

Aurix Exchange
Top