Simple, profitable EA

 

Hi All,

Here's an EA I have written, run against EURUSD on the D1 timeframe.

This EA looks to trade once a month.

If you make any improvements to the EA please post them.

Cheers!

lowtrader1.mq4

strategytester_1.htm

lowtrader1.ex4

Symbol EURUSD (Euro vs US Dollar)

Period Daily (D1) 1999.05.12 00:00 - 2011.05.13 00:00

Model Every tick (the most precise method based on all available least timeframes)

Parameters maxRiskPercent=1; gearingLevel=100; pipsStop=200;

Bars in test 3309 Ticks modelled 48940390 Modelling quality 87.28%

Initial deposit 1000.00

Total net profit 18762.82 Gross profit 37262.40 Gross loss -18499.58

Profit factor 2.01 Expected payoff 136.95

Absolute drawdown 65.16 Maximal drawdown 9116.20 (43.83%) Relative drawdown 43.83% (9116.20)

 

looks like martingale, and starting order # 115 it is not good

Try to run the test with a fixed lot and show us results.

 

No, its not a martingale!

I really wish you knew what you were talking about mate.

There is nothing even remotely martingale about the EA.

The lot sizes are determined by the strength of the entry criteria.

 

The good thing about this EA is that it's based off the D1 candle, so this means backtesting will be more accurate.

 

Cheers!

D1 bars should be the same across brokers. The frequency of trading too (once per month) should mean the strategy could even be utilised by those with shonky internet connections.

I haven't forward tested the bot yet, but the fact that the system is VERY simple, and it survives a 10 year backtest on the EURUSD seems quite promising.

The only downside of the bot I can see is that it won't make anyone rich overnight.

I was really hoping for some fellow programmers that use this board (I am a professional ERP programmer with 10 years commercial experience) to have a play with the bot to see if they can squeeze more profit out of the algorithm.

I've been writing robots for the last 3 years in my spare time, probably a thousand or more and it is very rare to stumble on any robot configuration that will survive a 10 year back test with a strong profit factor and a smooth equity curve.

Lets just hope I haven't fitted the algorithm to the data set.

 

130 total trades means 1 trade per month.

starting 115 -- 135 the EA was loosing meaning that more than two years it was not ok.

For 1D EA it is IMHO not good result.

 

Land of cash, admittedly it is far from perfect and yes what you have pointed out is correct.

When the market has all kinds of crazy fundamental forces pushing it to reverse its trend direction on a dime, or to switch from trending to ranging and then back again in a short space of time, it is imho unrealistic to think that you can ever make an algorithm that will anticipate such things and never have a down time or lull. The best you can do is win on average.

The only place you have guaranteed returns are in the bank... and those returns aren't big.

If you have any ideas as to what I might be able to do to make the EA better please share them.