So, I’ve written my own EA, but i need help with the drawdown of it. If there are any programmers on here, what drawdown do you normally aim for?
Here is a testing report. I am aware that this strategy won’t be a long term one but its not designed to be, its one I am writing for the experience of MT5 as it will probably become more commercially available within the next 12-24 months and I want to be ready for when it is.
Your comments are much appreciated.
Gary
Seems like you’ve got a decent system, 17% is not that much drawdown in my opinion. 65% win rate at a PF of 1.30 is a good start seriously. Besides a backtest, what else have you done to analyse the algo?
You’re 19, seeing your original thread; I am too. Nice to see some the same age as I and also working on algo trading. 
Best of luck,
Clark
Balance Drawdown is the difference between the last top level of balance and the last bottom level of balance. The same with Equity Drawdown.
That isn’t really a question for coders - it should be asked of traders. And since everybody has a different risk tolerance, there is no single “right” answer. You seem to be implying its too high, if so, you need to find another. Myself, I could stand 20%, but not much more.
Try to get higher PF by capturing more profit take pips. When simulating always good to remove the money mangement growing program. Stick to 1 lot size trading thru out then you can see the winning pips, PF, drawndown% and do some more modification (put back the money management later). You did more then 1k trades, is good but looking at your ticks is about 1 year for (1minutecharting). Try to simulate and test for more years. 2003 to 2008 is very trending. 2009 to 2011 is very sideways. If you achieve profitability in this 2 era, then your EA is good.
Thanks Clark. I am aware that the 17% isn’t alot of drawdown, its the equity drawdown of 45% which worrys me
I’ve taken about a month and optimized and these were the best set of results for Symbol and Variable Inputs and also took a fine comb through the result chart to see when orders were placed and whether they were complying with my system.
I tried a Virtual Trade method using the z-score to try and predict when the losing trades were going to occur, which worked to reduce the loss making trades but also resulted in missing some of the profitable trades. I’m just trying to figure out if anybody has any tricks to reduce their losers.
Also, are you coding mql language or C or a different one altogether?
Thanks
Gary
Oh and don’t worry guys, this EA is designed to only work well in downtrending currencies, its not designed to be a long term solution
