How do you define good EA?

Hey there, total noob here:)

I’ve been into trading fx for the last half-year and just started going into EAs. How do you define a good EA, what are some good stats of a successful EA when back- or forwardtesting? I know the question is a bit vague, but that’s the point, to generate some discussion, what are your personal criteria for a successful EA. Some screenshot of graphs/stats would help as well!
Thanks for your opinion in advance:)

Yasen

Im going to stab this, because I spend the entire day perfecting mine.

#1- Drawdown, most important.
A- tells you your entry is off, I aim for no more then 5%
#2- adaptation,
B- Does it run profitable during off season? ( Summer )
#3- MUST be 85% or better W/L ratio

#4- NET profit MUST be no less then 75% Of Gross Profit

#5- Money Management, Seems it should be first, but I beg to differ, Get the wins, then adjust varibles for MM.
A- I personally risk X% per trade, and always have just 1 trade open at a time.

Im currently almost done with my EA, just a few minor tweaks, 1 YR time Net Profit of $268,050 with $500 deposit,

Are these results from back or forward testing and on what tf’s does it run? Share a bit more on that, it seems like a success story:)

Yes, this is backtested, from March 13, 2009 - thru March 22, 2010 1 year, 1H TimeFrame,

30 SL
40 TP

I cant tell you anymore, its a year worth of work, X marks the spot,

I don’t understand, why are those number so high, is that in another currency? Other than that, wow…a nice percent of won trades there :slight_smile: Good job. Do you plan on selling this EA?

Hey Bud, that is EURUSD,

The lots get very high,

From March 19th, 2009 thru Sept, 2009 the lots reached 105 per trade, with a Balance of $670,000

After that, well, it gets nuts, the EA wont trade, if its not possible, margin wise, So it just keeps rolling, and when I stopped it on March 2010, the last lot size traded was 98,951

Im havent let it run till present day yet, I know, I know, why not? I dont know, Im happy with this much right now, But, Im going to do it right now, just o see the outcome, will post results…

And No, I wont be selling this…

Not saying its the cats meow, havent run live yet, but did let it run all day, and hmmmm, it was better results…

man, it wont let me trade farther after june of 2010, it just runs, but doesnt make a trade.

$602 Million was the final after 16 months,

dream dream dream

The first step is that you have to understand how the EA is taking trades. You have to know the EXACT setup and exits.

If you don’t know how it works, how do you know when it has stopped working?

Yes, most EA’s won’t give out their ‘secret sauce’. There are no shortcuts, learn to trade for yourself and stay in control of your money.

All above mentioned points are looking good… I have learnt a bit to start the trade with EA’s. I got some tips to select the best EA as well. Thanks all!

If something seems too good to be true…

Backtests can be fine-tuned to say just about anything. Don’t be impressed by people copying and pasting backtests on a website or a forum for that matter. There are a lot of scams out there. Live forward testing is what’s needed before any of it can become meaningful.

Yeah that’s a given of course! I am always skeptical regarding those too-good-to-be-true cases, but this doesn’t mean his EA is bad. Although his backtest might be invalid his EA results are nice anyway and will PROBABLY (but not necessarily) yeald good results in live testing as well.:slight_smile:

I don’t like to do back testing because it does not give you the real picture. It is OK to make initial check but
you have to test the EA live, which might waste your time but at least will give you the real picture.
When I am testing EA I want it to give at least the same amount of pips that I can make manually for 3 months in a row. Otherwise, I will not see the advantage in using this particular EA.

shady trader,
ea is just only a tool, for me there is no good or bad ea its depend on how the trader handle the ea (drive it, use the potential), do not believe all this wonderful backtest, its all rubbish!! a lots of free ea that can me you rich if you know how to handle it, most important get to know the characteristic of the ea and mix it with other tools. cheers!