EvolutionX robot review

[PipCop]

From http://www.fxstreet.com/forum/showthread.php?t=3211 comes the next EA, called EvolutionX. It's designed to run EURUSD 15M chart.

UPDATE: The authors site is here http://www.binoo.biz/

It also has a version to trade on Mini/Micro accounts, both of which are attached here.

UPDATE - 2008/12/16 - I manually closed the "hanging" trades from 11/17 and 11/19, and then the robot closed ALL trades.

UPDATE - 2009/03/17 - I manually closed the "hanging" trades as this robot seems to get stuck.

UPDATE - 2009/07/22 - I manually closed the "hanging" trades again.

UPDATE - 2009/08/05 - I have changed the lot size from 0.1 to 1.0, as this robot seems to be able to handle it at this point . Let's see if it takes off.

FireFox users warning:
Statements are sometimes cached by your browser!
Click the blue statement link (above this window) to update it!

Should I buy/use this EA on a live account? - Government Required disclaimer - Earnings disclaimer

evolutionx

X11D

[johngh]

Just a quick question regarding settings, their web site says:

Instruction:
1) Don't change anything in the setting.
2) It's may take time to open a trade patient is good.
3) But if the you can see a down trend you can set the
TradeSell=1 otherwise set the TradeBuy=1. Or if you don't know
the trend just don't change anything. Just choose either one.
4) Don't interrupt the robot just let the robot do all the trading.
5) Minimum Deposit USD3000 with 0.1 Lots.

I have had this EA running for a day without a trade, however during this exact time I see that your Demo Acc has had some 12 trades...i know it says "if you don't know
the trend just don't change anything", but as both the TradeSell=0 and the TradeBuy=0,
does this mean no trades will be placed?
You help please.
Thanks

[geekguy]

The trades on this EA are interesting, it seems that some of the StopLoss are set to 0.0002 and 0.0003 for some reason (assuming a code error).

[PipCop]

John,

Did you check the MetaTrader Journal and Experts tabs to be sure there are no errors? What brokerage are you demoing this on? Is it Alpari as well?

[garr8107]

James,

The figures shown in the "Live Quick Stats" box are current balance, which is great because it's a real-time snapshot. But wouldn't it make sense to include the equity figure as well? Just now Evolution X shows floating P/L of -$545 which is about one-third of the amount gained in closed trades. This skews the equity considerably. This might lead to an unfair comparison between rabbits like Evolution and turtles like RoboMiner, which usually maintains a far smaller floating P/L figure.

What do you think?

Thanks,

Bill

[PipCop]

Originally Posted by garr8107

James,

The figures shown in the "Live Quick Stats" box are current balance, which is great because it's a real-time snapshot. But wouldn't it make sense to include the equity figure as well? Just now Evolution X shows floating P/L of -$545 which is about one-third of the amount gained in closed trades. This skews the equity considerably. This might lead to an unfair comparison between rabbits like Evolution and turtles like RoboMiner, which usually maintains a far smaller floating P/L figure.

What do you think?

Thanks,

Bill

I fully agree. I have already planned a setup where you will have a little "I" icon next to each one, and you can mouse over it and a little window will appear with all the detailed stats. Of course, the floating P/L really doesn't always matter, I've seen several times where it has turned out fine, so in some regards since it's not final it shouldn't matter.

Of course, visitors would be remiss to not look at the detailed statements, especially since all the data is provided just one click away!

Ultimately, no one should be trading these live based off of a few days or weeks of activity. It defeats the purpose of a complete review, so I would hope no one is looking at the balance and saying "Hey, big balance, I'm going to trade live with that!" I've seen several EAs (such as the 10 points 3) build up a balance very quickly, and then wipe out the account in a matter of hours.

[geekguy]

Equity drawdown matters, but depending on the strategy, it may be anticipated and normal operation of the EA, specifically when using Grid or Oscillation strategies.

[garr8107]

James,

In this case it looks like Evolution lucked out. Today EURUSD retraced dramatically starting about 20:00 GMT which made good all those sell orders that were hanging out there. That won't always happen.

Your idea about the stats mouse-over is better than I had imagined. I'd say you must have done this before.

And I do look at the statements, for just the reasons you named. I also keep my own stats, basically a time-sequenced record of margin , equity, and profit, so I can trend them in Excel and see what the peaks were. It probably doesn't help control the actual risk, but it makes me feel better knowing what to expect.

Bill

[1ka]

I have some concerns over this EA. It could just be my environment, so I'd be interested to hear how other people get on.

Firstly, if I do a backtest starting from the 10th Nov. Then I get a bit of activity but nothing much. If I start it on the 11th, then there is consistent activity all the time. It does appear that this EA's behaviour is determined by the conditions at the point you start it. It could be that if you start it at the 'wrong' time, it won't do anything. In my forward test this EA has done nothing.

Secondly, I see many 4051 errors where it appears to be closing an invalid order. Since it does close many orders at a loss, it makes me wonder if these are errors or part of the algorithm. There's also an Order 131 error, incorrect volume right at the start.

I'm assuming that I've got the latest version, it's the one off his website, but without any version number in the code, then I'm not sure...

[PipCop]

Originally Posted by garr8107

James,

In this case it looks like Evolution lucked out. Today EURUSD retraced dramatically starting about 20:00 GMT which made good all those sell orders that were hanging out there. That won't always happen.

I totally agree. An EA which operates on luck isn't really the best strategy, in my opinion.

Originally Posted by garr8107

Your idea about the stats mouse-over is better than I had imagined. I'd say you must have done this before.

Thanks. I have a few other features planned as well that will make your spreadsheet obsolete.

While I've not done Forex to this extent, I'm a web developer by trade and the last site I built became the worlds largest for it's specific topic, so I learned a trick or two.

Originally Posted by 1ka

Firstly, if I do a backtest starting from the 10th Nov. Then I get a bit of activity but nothing much. If I start it on the 11th, then there is consistent activity all the time. It does appear that this EA's behaviour is determined by the conditions at the point you start it. It could be that if you start it at the 'wrong' time, it won't do anything. In my forward test this EA has done nothing.

I'm no expert, but I have seen the same behavior with several other EAs as well. Start it one day and it tanks almost immediately. Start it 1 day before or 1 day after and it is a winner. I am sure someone else has a better explanation for this phenomenon.

Originally Posted by garr8107

Secondly, I see many 4051 errors where it appears to be closing an invalid order. Since it does close many orders at a loss, it makes me wonder if these are errors or part of the algorithm. There's also an Order 131 error, incorrect volume right at the start.

I think that may be caused by a bad data feed and/or bad historical data?

Originally Posted by garr8107

I'm assuming that I've got the latest version, it's the one off his website, but without any version number in the code, then I'm not sure...

Good point - authors should include version numbers in their code, I know I do.