Same data, different MT5, different chart and backtest results

 

Both a friend of mine and myself are downloading data from  the same source (Tick Data Manager) and with the same settings and importing to MT5 as custom symbols. From what I can see the symbol settings are the same. However when running a backtest we are getting similar but different results.

On closer inspection of chart data, there are some noticeable differences in individual bars (especially on M! bars). So I can only assume that the slight difference in chart data is in turn slightly throwing off the indicator data resulting in some different signals.

Has anyone had experience with this and what may be the root cause of 2 instances of MT5 handling data from the same source differently?

I havent compared the actual data exported from Tick Data Manager side by side, so there is a chance the data being exported is slightly different each time for some reason...
The data is being exported as ticks with the same timezone and summertime settings. The time of the bars arent off, just the HLOC prices are slightly off on enough bars for it to be obvious.

Would be greatly appreciated if anyone was able to shed some light on what might be happening here?

Creating and testing custom symbols in MetaTrader 5
Creating and testing custom symbols in MetaTrader 5
  • www.mql5.com
Creating custom symbols pushes the boundaries in the development of trading systems and financial market analysis. Now traders are able to plot charts and test trading strategies on an unlimited number of financial instruments.
 
NSTprop: Both a friend of mine and myself are downloading data from  the same source (Tick Data Manager) and with the same settings and importing to MT5 as custom symbols. From what I can see the symbol settings are the same. However when running a backtest we are getting similar but different results. On closer inspection of chart data, there are some noticeable differences in individual bars (especially on M! bars). So I can only assume that the slight difference in chart data is in turn slightly throwing off the indicator data resulting in some different signals. Has anyone had experience with this and what may be the root cause of 2 instances of MT5 handling data from the same source differently? I havent compared the actual data exported from Tick Data Manager side by side, so there is a chance the data being exported is slightly different each time for some reason...

The data is being exported as ticks with the same timezone and summertime settings. The time of the bars arent off, just the HLOC prices are slightly off on enough bars for it to be obvious. Would be greatly appreciated if anyone was able to shed some light on what might be happening here?

The answer is within your description:

  • Either the source data is different, or ...
  • the method used to import and produce the Custom Symbol, was not exactly the same.
Only you and your friend are in a position to answer that question, by analysing the data first and then discussing the methods used.
 
Fernando Carreiro #:

The answer is within your description:

  • Either the source data is different, or ...
  • the method used to import and produce the Custom Symbol, was not exactly the same.
Only you and your friend are in a position to answer that question, by analysing the data first and then discussing the methods used.

Thank you for the reply, I'm in the process of collecting the specific window of data from the original source to compare. But what you are saying that assuming the same data and process to produce the custom sybol, MT5 has no room to build the bars from tick data differently? Assuming the source data matches and custom symbol were built the same it would be built exactly the same across any instance of MT5?  

 
NSTprop #:Thank you for the reply, I'm in the process of collecting the specific window of data from the original source to compare. But what you are saying that assuming the same data and process to produce the custom sybol, MT5 has no room to build the bars from tick data differently? Assuming the source data matches and custom symbol were built the same it would be built exactly the same across any instance of MT5?  

Yes, assuming both the data and the method are the same, and that platform conditions and context are also the same.

However, you have not explained the context nor the method.

Are you both using the same broker and account type or different ones?

Depending on the method you used, working with a different broker or different account type may affect the contract specifications for the Custom Symbol produced.

Since you have not explained neither the context nor the method, I am unable to offer a definitive answer.

EDIT: There is also the question of which settings were used on the Strategy Tester!

Reason: