Is the advisor suitable for real life? - page 51

 

In the meantime, I am slowly preparing an EA for the real world. I have implemented the mechanism of virtual trading, EA recording and restoring into a file, and trading functions.

So far we have the following (successful forward from the 62nd trade):

 
Dserg:

In the meantime, I am slowly preparing an EA for the real world. I have implemented the mechanism of virtual trading, EA recording and restoring into a file, and trading functions.

So far we have the following (successful forward from the 62nd trade):


There are too few deals.

Z.I. The idea of MM depending on balance is very good, but I have decided to use this way to avoid optimization trap. We should not achieve the best results of the tandem EA + lot regulator (virtual trade), but make the lot regulator stop trading reliably when the balance drops. This combination will reliably work with any pair (as a result check).

 
Dserg:
Yes, that's right. There are two bars on the deposit curve in pips. If they cross downwards - change lot from 1 to 0.01.
Here is the function, if anyone is interested: This module can be used in any Expert Advisor practically. This F should be multiplied by lot.

It's been a while since the branch was opened.
I wonder how this method proved itself in practice?
 

Useful idea, thanks Dserg! It is suitable for practically any TS, except unambiguously losing of course, allows for the same drawdowns to significantly increase the traded volume, while always adding time to make decisions in case of a prolonged drawdown, of course the most advantageous are not losing, but "earning nothing" TS, but this is a drawback of the system, and for the more robust ones it is just a good insurance.

One of the examples, though not with disconnection, but with minimum lot. There's a lot of trades wanted...:o)

Running with a standard lot:

And with MM on balance:


 
Prolon:

Useful idea, thanks Dserg! It is suitable for practically any TS, except unambiguously losing of course, allows for the same drawdowns to significantly increase the traded volume, while always adding time to make decisions in case of a prolonged drawdown, of course the most advantageous are not losing, but "earning nothing" TS, but this is a drawback of the system, and for the more robust ones it is just a good insurance.

One example, though not with disconnection, but with minimum lot. There's a lot of trades wanted...:o)

Running with a standard lot:

And with MM on balance:



And moss add virtual trading - there are libraries on the forum too. There's something here https://www.mql5.com/ru/articles/1441
 
YOUNGA:
And moss add virtual commerce - there are libraries on the forum too. There's something here https://www.mql5.com/ru/articles/1441
Sure mokhnache, who can argue, we're working on it! Thanks for the link, maybe we can pull something useful from there too...
 
Prolon:

Useful idea, thanks Dserg! It is suitable for practically any TS, except unambiguously losing of course, allows for the same drawdowns to significantly increase the traded volume, while always adding time to make decisions in case of a prolonged drawdown, of course the biggest winners are not losing, but "earning nothing" TS, but this is a drawback of the system, while for more robust ones it is just a good insurance.

One of the examples, though not with disconnection, but with minimum lot. There's a lot of trades that wanted...:o)

Run with standard lot:

And with MM on balance:


Have you done a comparison of results on a site that has not been optimised?
Such a comparison would be very revealing.

 
Basically these two pictures that I posted, the run for probably 6 years, some of the trend areas were optimized there, one year at most (I do not remember) and only in one of the directions. The rest of the mess was what the TS showed on the rest of the history. This is just an example of the application of what Dsergsuggested , although all this has already been covered in the first pages of this thread, there is still a lot to think about .
 

As you know, tester results and real trading results can differ significantly. I would like to know what tester results can be trusted.

I have only one Expert Advisor that has been in the real trade for a long time.

I have run and optimized it for last year - eurobucks, standard lot - 0.1.


 
Why aren't there any shorts?