Strategy tester Results

 

Dear folks, All EAs that am testing since last 20 days, they are giving me consistent losses in strategy tester for April 2019 and after. Is there a problem with the market or my strategy tester?

It cannot be a co incident that every EA is failing from April 2019 on my strategy tester.   what can be the reason for this failure?

 

Only STRATEGY TESTER? NOT MARKET? 

Only two cases: (possible)

1. Over optimizatión or natural robot issue performance. 

2. Failure / incorrect data download (spread, type data, failure data, corrupted (false data), tick to open, copy data into folder, combinated data,). 

Test:

Install NEW MT5, download NEW history data by broker since MT5.  

 
sjmotos:

Only STRATEGY TESTER? NOT MARKET? 

Only two cases: (possible)

1. Over optimizatión or natural robot issue performance. 

2. Failure / incorrect data download (spread, type data, failure data, corrupted (false data), tick to open, copy data into folder, combinated data,). 

Test:

Install NEW MT5, download NEW history data by broker since MT5.  

Maybe thats the reason, i will install new MT4 platform of a different broker than the one am currently using, and see how it goes

 
sjmotos:

Only STRATEGY TESTER? NOT MARKET? 

Only two cases: (possible)

1. Over optimizatión or natural robot issue performance. 

2. Failure / incorrect data download (spread, type data, failure data, corrupted (false data), tick to open, copy data into folder, combinated data,). 

Test:

Install NEW MT5, download NEW history data by broker since MT5.  

Issue seems to be the EAs since they are free EAs.  i opened new demo account with new broker and data was freshly downloaded this time but same results.

 

Totally normal. Make new backtest and manage risk (drawdown max do not go over 10% safe system). 

Example: 

(year,month,day)

2018,6,1 -> 2018,9,1 backtest & optimization, save results config.

2018,9,1 -> 2019,3,1, ONLY backtets previous config. 

If there are many parameters and it is optimized a lot, most of the future results are bad or flat (different from the past). The solution is to apply data mining to find the most appropriate results, or to make a more complex or multiple system.

Do not forget commissions or spread. (Some brokers charge the commission apart from the spread, it can affect).

 
sjmotos:

Totally normal. Make new backtest and manage risk (drawdown max do not go over 10% safe system). 

Example: 

(year,month,day)

2018,6,1 -> 2018,9,1 backtest & optimization, save results config.

2018,9,1 -> 2019,3,1, ONLY backtets previous config. 

If there are many parameters and it is optimized a lot, most of the future results are bad or flat (different from the past). The solution is to apply data mining to find the most appropriate results, or to make a more complex or multiple system.

Do not forget commissions or spread. (Some brokers charge the commission apart from the spread, it can affect).

yes i know about all that but this time it was funny to see 3 diff EAs with diff trade logics performed exactly the same only since April 2019.  I have seen the BUG in two EAs now. Its fault of EAs, thank you very much for your help sjmotos.

 

If you are using mt4, run the backrest and open up the chart and go through the trades

Check if your price data is having any issues

Mt5 should be more reliable with the price data

 
I use mt5 to optimize my ea and trade on mt4
Reason: