Existing EA to calculate profit based on given buy and sell dates.

 

To back test my strategy is really hard, since its discretionary it is hard to this manually as there are always different ways of seeing things.

In order to find the best way to make the system follow rules instead of my thoughts, I would need to play around with rules and create entries in back test mode.   These rules will effect my trading, limiting potential profit and loss but creating a more consistent growth. 

In order to find which rules will work the best, I will need to go back and back test a bunch a bunch of sets of rules. and see which is best.

This can be a hard task as it is all manual. I would like to test out as many set of rules as possible and not settle on one due to the work involved. 

Can I not instead feed MT4 a set of buy and sell dates ?

 

So mt4 would get data like this :

jan 3 2013  buy on the open.        jan 5 2013 close on the close

 

jan  7 2013 open on the open       jan 8 2013 close on the close

jan 13 2013 open on the open      jan 15 close on the close.  

 

 So "open trade date" would be on column A  and "close trade date" would be on column 2. (adding time would be nice too)

This data would be some how given to mt4 and it will calculate the profit that would of been made if those instructions would of been followed.   (I would imagine vie a spread sheet)

From there I can see the data on the system like max draw down, % win, ect adn get results.

Also, as a bonus, It would give an output in spread sheets that would calculate the actual profit per trade and enter it in column C.

From there I could run a simple EA like a Bollinger band EA, pull out all the dates of each trade and put them in excell and then see if results are better when both my manual system and the bollinger band system are giving a buy signal. 

In essence, I would be optimizing by adding EA's and then seeing the dates the EAs are agreeing and only take those trades.

But that part is down the line and should be a bit easier.

Any thoughts, I am really stuck here. Thanks  

 
temptraitor:


To back test my strategy is really hard, since its discretionary it is hard to this manually as there are always different ways of seeing things. 

You don't have a Strategy., you are guessing and calling it "discretion".
 

I would say that is about accurate at this point, 

I am trying to make a system that is solid using programming.  

 
temptraitor:

I would say that is about accurate at this point, 

I am trying to make a system that is solid using programming.  

OK,  good.  Design your system,  test it manually to prove it is consistent then you will be in a position to consider coding it.
Reason: