Karsus

If you can stop it from reversing position too much. A system that has only one loss in quite a while could be decent.

Well its not all bad. Some of it is pretty decent... Some of it is appallingly bad. I'll admit one of the systems does seem like plain treachery to me -- chance of 90% drawdown for minimal chance of profit.

So is anyone really running this EA, with settings that seem profitable?

Re: EXMQ1 Expert Advisor - 50% Profit Monthly

 

The Live statement shows a massive dive in profits.

Re: Forex Spectrum

 

So what do the backtests look like on 4 digit?

Re: Is it super-gral or Dont trust optmize results for the past

 

Normally how long parameters continue to work well depends on the period of optimization. If lets say you optimized for all of 2009, the system should continue to work profitably for at least the first couple of months of 2010.

 

The problem isn't so much with optimization as with the fact that the markets and people's behaviors are changing over time. Looking at historical data, I've found trades that worked well, lets say in most of 2009, but the same ideas don't work as well in 2008. The average profit targets vary, or sometimes the idea almost seems to have no application.

 

Also, apart from some super-general trading setups on higher timeframes, there is variation in how well different trade setups will work across different pairs.

 

Personally, I'd look at the equity curve of the EA, whether its fairly straight, and whether its consistently profitable in the last several months.

 

For something like that EA, you probably don't want to change the ATRPeriod... one additional factor I suppose, is to make sure that there are a fair number of trades that form your equity curve on the backtest. If you've filtered down to it having very few trades on the backtest, then the backtest isn't going to be very useful.

Re: Fusion V 1.2 Nedd Educated

 

I assume you're using a high risk setting. At higher risk settings you make money fast, but you can lose money just as fast.

 

Also, you should backtest EAs with the amount of money you intend to invest, to see how they are likely to work out... and do the backtest over a reasonable period of time. For Fusion, that means doing a whole bunch of backtests to make sure that the various systems within it all work well on your broker, and so you can disable the ones that don't.

Hi guys,

 

Does anyone know of a backtesting software that will test several MetaTrader 4 EAs at the same time on different timeframes and pairs? I know that they exist but I can't seem to find one.

Re: Fusion v1.1a

 

Reverse-engineering a dll and converting it into something like mql4 is no minor feat. It also won't help any. The reason HiRider makes its profits is because of the huge TP/SL ratio, and because it often takes much larger risks than it should. Of course, that means that when it does hit its SL... it tends to pretty much kill the account.

 

You could change the risk setting and the number of trades open at a time, and see how much that helps you - backtest your settings.

Re: Fusion v1.1a

 

Is anyone seeing profit on LMD Multi-currency?

Re: Fusion v1.1a

 

Hi Rider is a risky EA that periodically blows the account. Maybe it should be treated like a martingale... take out profit periodically.

Hi guys,

 

Just saw an EA

 

http://forexpipzen DOT com/

 

Seems like a decent EA. 146% Profit, for a 16% drawdown on the backtest for Jan 2008 - Dec 2009 (2 years)

 

Alright, so its not suggesting ultimate awesomeness, but that's not bad.

 

The EA comes with a 60 day free trial, with you being charged if you choose to use it, after 60 days. Payment made through a subscription on Paypal... Which sounds pretty reasonable to me.

 

It is a little under-marketed, and doesn't even directly show up in a google search for it :P

Re: Fusion v1.1a

 

Well they did convert it into a C library, add their own protections and modify the code so that it could work from a C library instead of MQL4.

 

The flaw in HiRider is probably part of the original design. It was likely never intended to be commercial quality if its a private EA.

Re: Elite Currency Trader EA

 

Well that really depends. Lets say you have a setup with an expectation of 1:1 risk/reward. That may or may not end up succeeding on both the pairs.

 

In any case, a 3% max drawdown means that if you risked 1% per trade, you can't have more than 3 consecutive bad trades in a year. To additionally have that constraint on more than a few pairs simultaneously, is a rather heavy constraint. Unless you risked less than 1% per trade, which is what you'd have to do to meet your requirement since almost no system guarantees that you can't have 3 consecutive bad trades.

 

Since this is what you consider a good backtest, and not a very impressive backtest. Are you implying that you know of such an EA?

Re: Elite Currency Trader EA

 

On a single EA with no difference in settings across pairs, for more than a few pairs (meaning lets say, not just the USD pairs)?

Re: Elite Currency Trader EA

 

A backtest with a decent ratio between overall profit and maximum drawdown that has more than a few trades, a reasonably straight equity curve, and a not particularly extreme drawdown... Although I care more about the ratio of Overall profit to maximum drawdown, since the actual drawdown can be managed by modifying the risk.

 

Personally, I like the average profit trade to be greater than the average loss trade, and more winners than losers - although if you had a system with a 1 to 10 average profit to loss that had an average winning trades to losing trades ratio of 100 to 1, I wouldn't really object too much, and the drawdown to actual profit would be fairly decent.

 

Generally the backtest should also be over a decent period of time. But really, I think it is more than just a case of good strategy and bad strategy - the markets do change, and because of that, you want to use trade setups that are more profitable now when you're going to need them and use them, even if they weren't always profitable. Just as there were trading strategies in the past that don't work quite as well today as they once did because too many people knew them.

 

 

So what is your idea of a good backtest?

Re: Fusion v1.1a

 

It is, although I couldn't get Birt's data to work right with my MetaTrader. Birt also says the EAs in Fusion stink.

Re: Fusion v1.1a

 

@OneMore: Where did you find good M5 data from several different brokers for 1999 - 2010?

When you use History Center data, you pull data from MetaQuotes which is not the same as the feed of data from your own broker... and its pretty bad quality.

Re: Elite Currency Trader EA

 

Well I think the :P made it obvious that I was joking :)

 

And what would make me different would be the fact that I would do research to produce a reasonable system that isn't based on some variant of scalping. Seriously, I think its impossible that you can't make decent EAs yet we're seeing a whole lot of questionable EAs. Its like no one is willing to put in the effort to produce a good product.

 

Most EAs seem to be implementations of simple ideas, instead of well thought out systems. Its not that I'm shocked at the scams, I'm shocked that as far as I can tell, nobody in the market has developed a truly respectable EA, when there's so obviously a decent market for it. Maybe I should go banging at the doors of a few VCs :P

I have an EA with at least decent backtests - that should give me some credibility to ask for funding to make a better system, at least. LOL.

Re: Elite Currency Trader EA

 

Dang, hype everywhere and no promising EAs :P

 

People, fund me and I'll do the research and make you a decent EA :P

Re: Fusion v1.1a

 

Oh Thank God, I didn't spend money on this thing. I was seriously considering it. If I had, I really would've risked running HiRider to recoup the losses :P

 

@Simoniex: Did you backtest it on a GMT +1 broker and get a decent equity curve? I saw a comment here somewhere that it didn't do too well on a GMT +0 broker, even if the overall year ended in profit, because by the end of the year it was just giving profit away.

Re: Fusion v1.1a

 

I like Birt's reviews.

 

@Birt: Could you understand the actual entry logic in the systems btw? Was it something interesting? I admit that for something complex, trying to understand the ASM would've been a considerable pain.

Re: Forex Robot World Cup

 

It has a complex dll protection, even if they share it, you can't test it.

Re: Forex Robot World Cup

 

10 stoplosses sounds like a major ouch on that thing. That should wipe out like 200 good trades or something. What do the backtests for the first place one look like?

Re: Forex Robot World Cup

 

Can someone who has bought these robots confirm that these backtests are accurate?