Back testing is positive but on live account is negative. WHY?

To add comments, please log in or register
Yasir Kidil
80
Yasir Kidil  

I have an optimizated EA. It is profitable but on live account is negative. I don't understand. Is back testing totaly lie?


You can see the result: https://www.mql5.com/en/signals/451038

Trading Signals for MetaTrader 4: OTTCCI Demo
Trading Signals for MetaTrader 4: OTTCCI Demo
  • www.mql5.com
Symbol Deals Sell Buy Symbol Gross Profit, USD Loss, USD Profit, USD Symbol Gross Profit, pips Loss, pips Profit, pips MFE and MAE Distribution Point Graphs Maximum profit (MFE) and maximum loss (MAE) values are recorded for each open order during its lifetime. These parameters additionally characterize each...
William Roeder
20250
William Roeder  
Yasir Kidil: Is back testing totaly lie?
How should we know what you did?
  1. Did you use a valid spread (not current.)
  2. Did you download M1 history covering your entire test?
  3. Did you use every tick method?
  4. MT4 Strategy Tester : good practices, know-how and howtos. - Strategy Tester - MQL4 and MetaTrader 4 - MQL4 programming forum - Page 2
Yasir Kidil
80
Yasir Kidil  
whroeder1:

How should we know what you did?

  1. Did you use a valid spread (not current.)
  2. Did you download M1 history covering your entire test?
  3. Did you use every tick method?
  4. MT4 Strategy Tester : good practices, know-how and howtos. - Strategy Tester - MQL4 and MetaTrader 4 - MQL4 programming forum - Page 2

1. YES

2. YES

3. NO

Is the third option important? I use the "open prices only", otherwise optimization test time is being very long.

4. I looked. Good topic but it is not helpful for me because I already applied their said.

Edit:

By the way, my expert advisor works on M1, M5 and M15 charts and It opens a trade when the candle close. "Every tick" method is not for my expert advisor, I think.

Icham Aidibe
11680
Icham Aidibe  
Yasir Kidil:

I have an optimizated EA. It is profitable but on live account is negative. I don't understand. Is back testing totaly lie?


You can see the result: https://www.mql5.com/en/signals/451038

The most common error in these cases is about the trading conditions. I see a MT4 signals, it supposes you have 99% quality data and imported it in the platform with the right brokers' spec (including commissions).

I highly suggest you to switch to MT5.

Yasir Kidil
80
Yasir Kidil  
Icham Aidibe:

The most common error in these cases is about the trading conditions. I see a MT4 signals, it supposes you have 99% quality data and imported it in the platform with the right brokers' spec (including commissions).

I highly suggest you to switch to MT5.

You say it might be profitable if I transfer the codes to MT5, right?

Icham Aidibe
11680
Icham Aidibe  
Yasir Kidil:

You say it might be profitable if I transfer the codes to MT5, right?

I say by using MT5 you won't have to worry about brokers' spec & datas when testing

Marco vd Heijden
Moderator
12737
Marco vd Heijden  
Yasir Kidil:

You say it might be profitable if I transfer the codes to MT5, right?

The answer is no.

You should design it around live market feed.

Not the tester.

What makes you believe that if it is profitable in the tester, it will also be profitable on a live market ?

The tester is to test, the logic, to see if the code works properly.

It's a tool.

Yasir Kidil
80
Yasir Kidil  
Marco vd Heijden:

You should design it around live market feed.

Not the tester.

What makes you believe that if it is profitable in the tester, it will also be profitable on a live market ?

The tester is to test, the logic, to see if the code works properly.

It's a tool.

I aggree with you. It's a tool. How do big players like goldman, ubs, blackrock test the algorithm? How can I design it around live market feed? Could you tell some?

Marco vd Heijden
Moderator
12737
Marco vd Heijden  

Not really i guess they have their own engines that provide the historic data at a fast pace, i think that it should not be too hard to program something like that.

But i usually run it on a demo for a month or two and then it's clear for 99% of the algo's you put up for testing.

Alain Verleyen
38753
Alain Verleyen  
Yasir Kidil:

1. YES

2. YES

3. NO

Is the third option important? I use the "open prices only", otherwise optimization test time is being very long.

4. I looked. Good topic but it is not helpful for me because I already applied their said.

Edit:

By the way, my expert advisor works on M1, M5 and M15 charts and It opens a trade when the candle close. "Every tick" method is not for my expert advisor, I think.

Are you using a stoploss and/or takeprofit ?
Yasir Kidil
80
Yasir Kidil  
Alain Verleyen:
Are you using a stoploss and/or takeprofit ?
Yes. I am using but it is different.
My risk reward ratio is 1:2
TP parameter is optimizated. For example if optimizated TP is 300 points (30 pips) then stop-loss is 150 points because of RRR. 
Also, I am using trailing S/L and trailing T/P.
123
To add comments, please log in or register