My Best EA | Forex Factory

Options

Permalinks
Print Thread

Similar Threads

Which system is the best? Or which combinations of systems work the best? 46 replies

Newbie: Best broker, best platform - trading off charts? 4 replies

The Best of The Best Trade Follow up RSI 6 Divergence 7 replies

Best Demo platform = Best Real Platform? 6 replies

Best Broker with Best Math Processor? 3 replies

Attachments: My Best EA
Exit Attachments

My Best EA

  • #1
  • First Post: May 10, 2011 4:21am May 10, 2011 4:21am
What is best EA should be?

I've developed many ideas into EA and truely some are craps & some are profitable but not last long. they're not adapt to market changes.

For some EAs, they are profitable during the correct trend and worst at ranging or MC at reverse trend
For some EAs, profitable only at the ranging only, worst when trend occurs.

I do believe other developers also facing similar experiences.

So recently I had a new thought:
Forget about other trade secrets, one need only a good system to deal with 3 things(TREND, RETRACE,REVERSE TREND)
Yes, only this 3 main things..other than MM

To develop EA that can adapt to market change automatically so called
1. Follow trend , sure profit
2. Retracement , still profit
3. Reverse trend , have to dealing with some floating but finally still end up with profit

Had tested in the month of Jan 2011 & Feb 2011 market conditions ( I like this 2 months because it consist of all the 3 above).

results so far:
with depo amount of $1000, start lot size 0.1 (10 cents/pip on account leverage 1:500)
Jan gain 50%, with maxDD 50%
Feb gain 60% with maxDD 48%
March still testing...results so far 28% 10 days trades...

may be due to a lot of calculation involve, I cannot make the test faster.
  • #2
  • May 10, 2011 4:25am May 10, 2011 4:25am
Quoting akukaya
What is best EA should be?

I've developed many ideas into EA and truely some are craps & some are profitable but not last long. they're not adapt to market changes.

For some EAs, they are profitable during the correct trend and worst at ranging or MC at reverse trend
For some EAs, profitable only at the ranging only, worst when trend occurs.

I do believe other developers also facing similar experiences.

So recently I had a new thought:
Forget about other trade secrets, one need only a good system to deal with 3 things(TREND, RETRACE,REVERSE...
Did you test that out with metatrader? Cause if you did, you should be aware that the backtests generated with that program are complete crap and don't reflect reality.
  • #3
  • May 10, 2011 5:18am May 10, 2011 5:18am
Quoting akukaya
results so far:
with depo amount of $1000, start lot size 0.1 (10 cents/pip on account leverage 1:500)
Jan gain 50%, with maxDD 50%
Feb gain 60% with maxDD 48%
March still testing...results so far 28% 10 days trades...
I agree with Custos and also, would you have the nerve to stomach a 50% drawdown for a 50% profit? Most investor begin bitching around the 15% mark and by the time you reach 25 they move on to the next newest and greatest.
  • #4
  • May 10, 2011 6:19am May 10, 2011 6:19am
Quoting Custos
Did you test that out with metatrader? Cause if you did, you should be aware that the backtests generated with that program are complete crap and don't reflect reality.
Yes, I've gone thru this topic many times.
I dont just listen to people, I make test myself.

Dont mean I trust backtest 100 per se. But it depends on your objective to make use of what backtest can offer with it limitation.

I did with my previous strategy: backtest, live demo and live account.
I understand what is the different.

never use optimisation, to mold results.

Just test the strategy and for analyse, that's sometime why my backtest takes time too
  • #5
  • May 10, 2011 6:58am May 10, 2011 6:58am
Quoting traderjai_inhi
I agree with Custos and also, would you have the nerve to stomach a 50% drawdown for a 50% profit? Most investor begin bitching around the 15% mark and by the time you reach 25 they move on to the next newest and greatest.

Traderjai, Understand that, but you can make the number look small too, it just mathematical.

It's all depends on traders risk appetite.

if account size $10,000 with similar calculation strategy, the gain will be 5% per month and maxDD also 5%.(or by using similar $1000 with 0.01 lotsize)
Now it's sound good on risk percentage?
You want it to be 10%, just refer to account size $5,000 instead of $1000.

BTW, what I want to stress is best EA with consistent profitable results when running 24/5. what ever market condition is.

for MM part, I just make reference point with $1000 with 0.1 lot/pip as an example.
  • #6
  • May 10, 2011 7:56am May 10, 2011 7:56am
Quoting akukaya
Yes, I've gone thru this topic many times.
I dont just listen to people, I make test myself.

Dont mean I trust backtest 100 per se. But it depends on your objective to make use of what backtest can offer with it limitation.

I did with my previous strategy: backtest, live demo and live account.
I understand what is the different.

never use optimisation, to mold results.

Just test the strategy and for analyse, that's sometime why my backtest takes time too
I have nothing against backtests. If you use a proper backtesting program, then a backtest can reflect reality, of course those backtesting programs cost anything from +1,000$.
I just have everything against a metatrader backtest, cause there almost any strategy seems to work out.
  • #7
  • May 10, 2011 8:40am May 10, 2011 8:40am
Quoting Custos
I have nothing against backtests. If you use a proper backtesting program, then a backtest can reflect reality, of course those backtesting programs cost anything from +1,000$.
I just have everything against a metatrader backtest, cause there almost any strategy seems to work out.
Ok, Custos. I take note on your comment.

BTW, your post are among posts that I like to read. Most of the point are relevant and useful. Among others, I like to read Hanover/Aja/Craigs comments too. To name a few.
  • #8
  • May 18, 2011 11:52am May 18, 2011 11:52am
backtested results Daily Jan2011 till now
total 1379 trades
Attached Image(s) (click to enlarge)
Click to Enlarge Name: DailyProfit.JPG Size: 146 KB
Click to Enlarge Name: DailyProfitEA.JPG Size: 62 KB
  • #9
  • May 18, 2011 11:55am May 18, 2011 11:55am
Quoting akukaya
backtested results Daily Jan2011 till now
total 1379 trades
uh, that looks like martingale technique. Not sure if I would wanna trade that. Why you don't try another money management style?

Martingaling will surely one day lead to the depletion of your account. Just a warning.
  • #10
  • May 18, 2011 7:10pm May 18, 2011 7:10pm
Quoting Custos
uh, that looks like martingale technique. Not sure if I would wanna trade that. Why you don't try another money management style?

Martingaling will surely one day lead to the depletion of your account. Just a warning.

Hi Custos,

yes, sure it can also design without martingale, kind of lot size positioning,etc.

Main thing here is I am testing the concept mention in post #1.
When you can have system that define 3 things : TREND, RETRACE, REVERSE...that's it. You can design a profitable EA.

Hey, see the graph attached..not a standard martingale there..

With standard martingale, you have death trade during the month of Feb itself.

best rgds,
  • #11
  • Jul 6, 2011 12:40pm Jul 6, 2011 12:40pm
so, let us test your creation
  • #12
  • Jul 6, 2011 1:00pm Jul 6, 2011 1:00pm
Quoting Sgt.forex
so, let us test your creation
knew it was only a matter of time...
  • #13
  • Jul 6, 2011 1:05pm Jul 6, 2011 1:05pm
Quoting traderjai_inhi
knew it was only a matter of time...
and whats your problem with this?
  • #14
  • Jul 6, 2011 2:29pm Jul 6, 2011 2:29pm
As long as you have sound entry logic it's possible to successfully use the martingale method to achieve reliable results.
Attached Image (click to enlarge)
Click to Enlarge Name: result.jpg Size: 97 KB
  • #15
  • Jul 6, 2011 2:57pm Jul 6, 2011 2:57pm
A coin toss entry with Martingale sizing will work over time as long as you can withstand 15 heads (or tails) in a row without getting a margin call. The 15 in a row number is arbitrary. The real number would be based on your initial position size, your initial account balance, your leverage, etc.

The probability of getting n heads in a row over so many coin tosses is elementary and well documented.
You are in a maze of twisty little passages, all alike.
  • #16
  • Jul 6, 2011 2:57pm Jul 6, 2011 2:57pm
Quoting Purger
As long as you have sound entry logic it's possible to successfully use the martingale method to achieve reliable results.
Funny how everyone saying this puts up backtests...
  • #17
  • Jul 6, 2011 3:23pm Jul 6, 2011 3:23pm
I have a buddy who has traded that same EA with real cash for a while and has attained reliable results in real life. I'm also forward testing it right now myself.
  • #18
  • Jul 6, 2011 3:48pm Jul 6, 2011 3:48pm
uh huh... ok

So why not put up your "forward tests"... hmm
  • #19
  • Aug 5, 2011 5:27am Aug 5, 2011 5:27am
Quoting akukaya
backtested results Daily Jan2011 till now
total 1379 trades
Hi,

Can you share your EA, please
  • #20
  • Last Post: Aug 10, 2011 3:16am Aug 10, 2011 3:16am
http://www.forexfactory.com/attachme...1&d=1305737315
appreciated

BISE Sargodha
0 traders viewing now