I am a newbie , been in the forex world for 7 months now and very interested in automated trading . I have an EA i found somewhere and loved how it worked and tweaked it to add risk and money management settings based on some calculations i like to make and use for money management when trading forex .
Below are the results of the EA back tested from 1st July to 4th July , i wanted to back test it from 1st of July to 1st of August but my data got stuck after that and the ea couldn’t go on so the last series of trades opened were stop closed at a loss . It took $500 to $1,330 in 4 days , impressive yes but i don’t like to call it profitable yet. It uses the much hated martingale strategy yes, but i like to keep an open mind on everything in the forex market .The EA has a small take profit , right after spread is catered for , whatever comes after is picked as profit . It uses no indicators .
I would like interested members to critique and give me their opinions on the results , and i will share it with one of the members here to perform further analysis and testing for me . LORD-EA results.zip (93.9 KB)
Interesting EA. One of the things I noticed by the strategytester report:
Modeling quality is N/A. Which could mean that you don’t have the proper history data to properly test your EA. In my opinion modeling data should be 99%, at least. This way you could properly backtest your EA. But this is definitely not proof that it will perform in a forward test…!
If you don’t have such data please let me know. I can test drive your EA with tickdata from 2012 till now on all currencies if necessary.
Hi Lakoenza, Thank you , i have tried to back test and getting 99% modeling quality but i am not good at doing that at all . I would appreciate your help . Send me your email and i’ll send you the full EA with whatever else i have in mind about it .
I went through a tutorial and did a ‘newbie’ back test . I tried to download back test data but from the look of it , it seemed very awkward with gaps and the EA worked on it but kept jumping price and looked very unreliable . I decide to use a simulator , i ran it and it collected price data from my broker demo that looked much more reliable and natural even though the EA couldn’t go through all of the data found . I ran it from 15th July to 17th July and was able to get at least a 90% modeling quality . The EA took $500 to $605 and the results are attached below .
Are back test data downloaded really reliable as they look because i believe they are ALWAYS downloaded from meta quotes servers and that the 4 digit broker data set downloaded from Meta quotes contains MANY errors such as major price gaps in lower time frames .
Would back testing an EA using the daily data be more reliable and good for analysis on how an EA works in daily market conditions than back testing over years as this only seems to hide bits of bad behavior about the EA?
All sort of criticism and debate is welcomed . I only get better from being wrong
Simple question, does your EA run every tick or once per bar. Also there seems to be a lot of activity from the bot (mod orders) your broker won’t like that.
Hi Bobbillbrowne
The Ea runs every tick actually . Thank you for the link to the thread , reading it to grab the nice points and tips you’ve given there. So apart from the modeling quality , just a quick look at the results and how i go about testing the EA , what do you see as wrong and something that needs a deeper look at ?
Man o man, bro, so many dangers. Tick based bots are neigh on impossible to perfect unless it is pure price action driven. Then there’s the question of data quality. Any variance in backtest data and your brokers live feed including spread pretty much voids your results. And of course past performance is no indicator of future results. Just check out the threads (including my own) that have died.
With that said there’s no better way to test the validity of a strategy. Your goal should be to trade the system manually. And believe me after hours of back testing you’ll know your strategy very well. Remember it’s just and expert “advisor,” you are the expert “trader.”
Martingale system is absolutely profitable, but it need huge balance to make safe our account live. In liteforex, I use this trading sytem too…
Thanks anyway for sharing your robot sir, I will test it first.
