Optimization results inconsistent with backtest

 

Hi,

It suddenly appear that on my MT5 with a new adviser, after fast test, I got a list of optimization results, when I double click one of it, the result on backtest  and graph doesn't match the profit result on optimization results.

Here is the backtest result when double click above Pass:4 504

Here is the graph:

All three parts are inconsistent. I tried delete cache from data folder, under tester, but it still doesn't work. Can anyone help me on this?

 
I can see the following on your last image:
the trades were closed due to stopping back testing.
Means: if you stop backtesting (or if the backtesting was stopped by itself) so all open traders are closed because of that (forced to close).

And it may be the other reason ..
 

No, Sergey

It is not stopped, it is balanced to under 0, so automatically stopped. That is confusing me, because the optimization results shows a profit $299352.10, but when backtest, it cannot complete the whole time period, and got under negative on balance and equity before the end. That is something looks like stopped.

Any idea?

 

I explained about what I see from your last image.

As to backtest ... it depends on many factors: number of ticks, the settings, the time of starting the backtest (this time to be the same is very important for martingale EAs and for scalpers), quality of the data, and more.

Besides, if you are using MT5 strategy tester with "every tick based on real ticks" so this kind of backtest is very close to reality (it is almost same as you are trading by this EA).

 
I think the backtest when I double click from optimization results should have all the same set when I run optimization. So, the results should match. And I usually have the same results on other advisors, but for this new one, it doesn't.
 

I'm having the same issue.. trying to work it out. 

Some forums suggest checking that you use an actual spread instead of 'Current'. That still doesn't work for me. 

 
Same problem, this is my optimization, when I double click on this result, the backtest statistic is diffrent. I'm using FP Markets 

 
Luca De Andrea #:
Same problem, this is my optimization, when I double click on this result, the backtest statistic is diffrent. I'm using FP Markets 

What kind of optimization did you do ? Fast Genetic or Slow Complete ? On your computer agents or using network agents (Cloud?) ?
 
I have had similar issue when using real ticks.
 

I have same issue. After doing some digging the issue for me seems to be the  MQL4 Cloud Network.

I ran a small fraction of the mismatched results using local agents and the problem went away. Let me know if this is the same for anyone else.  

Reason: