Martingale....The Holy Grail or Are We Chasing A Myth? - page 2

 

Guys, have you ever thought about using Martingale on CFD of stocks? The down side of the move is limited by the value of the company, so the stock can not downtrend forever. SO, I believe that Martingale EA can be used quite well, when it is turned on for only buying after CFD has retraced, so that we do not enter at the peak. There are two problems: the stocks are not so valotile as currencies so we can wait several days or weeks for retracement, and the SWAPS while waiting...

Any suggestions?

 
aegis:
You're chasing a myth.

I agree.

However, myth-chasing is how a lot of very cool stuff gets invented.

I say, 'Proceed as you see fit, fellas!'

aegis:
Do you think professional money managers use martingale-type money management?

I'm not sure that "professional money managers" as a class have completely distinguished themselves as masters of nonlinear market dynamics. I seem to recall a test where a chimpanzee throwing darts at newspaper's stock section outperformed the money managers assembled for the test. "Your money managers milage may vary."

I was looking at the "Top Ten" money managers (CTAs) featured in this months Futuresmagazine and I remember thinking that some of the EAs on this Forum could outperform those guys, with a bit more development.

aegis:
Any system that relies on digging itself in a hole in order to make huge profits is far from the holy grail.

I think there is more than one way to look at this. Is it "digging itself in a hole" or is it "selective distribution of bets within the expected range of price action"?

I agree that there is a problem with these EAs: the big jumps of price outside of "the expected range of price action" are not being dealt with well, with the current techniques. And those jumps occur often enough and unexpectedly enough to pose a real and expensive challenge.

aegis:
...it cannot be changed or tweaked to work.

That may be so. It is definitely true if no effort is made to look at it...

However, I've been around hard problems long enough to know that all it takes is one guy looking at it in a different way ... one guy to have that one idea that changes the whole game.

SO FAR it DOES seem that the only way to end up slightly ahead in the long run is to start with a very large account with a very small inital position size that does not keep doubling it's size past a certain point.

...So far...

However, a lot of stuff hasn't been tried. For example, I've noted several people wondering what would happen if these EAs pyramided additional lots in the winning direction. AFAIK there are several untried possibilities in this area. What happens if orders are pyramided so that ordersize is denser in the direction of favorable excursions than the adverse excursion orders?

What happens if ... well it's getting pretty late here ... I'm gonna sign off now, but I think the "What happens if" - type questions are well worth asking w/ the martingale systems.

While the 10 Points 3s and mods thereof may not be ready for prime time, I think they have elements that have a lot of potential. Perhaps we can find that idea or combination of ideas that does the trick.

 
Pecunia non olet:

I say, 'Proceed as you see fit, fellas!'

Oh cool, you're still alive. That's a nice and warm statement. Btw, did you receive my email? Are you helping me to test my EA? Where is my statement?

Regards,

David

 
davidke20:
Oh cool, you're still alive. That's a nice and warm statement. Btw, did you receive my email? Are you helping me to test my EA? Where is my statement?

Regards,

David

Alive and testing! I'll post when I get to the other computer.

Thanks again!

 
Files:
 
 

Jeff,

That article was the basis of the Martingale systems that you see floating around here, courtesy of Alejandro Galindo. He is a very talented programmer who has contributed to a large number of EAs that we are using in some form or other.

By the way, boris and kathy lien have a very low credibility in my eyes. Didn't they start a site providing signals? Can anyone say how they are doing?

Aegis,

Pro money manager does not mean that he/she has the holy grail. If you search around, you will find that last year most of the big forex money managers lost money. I agree with your conclusion on keeping the sizes small compared to your capital for Martingale systems. Like you said everyone is looking for 1000000000% per annum and not a more realistic rate of return.

By the way, Aegis, I am sorry to bring it up here, but you haven't replied back the emails I have sent you. You owe me money for the EA I had worked up for you. By the way, I have been forward testing it with very conservative settings from the day I told you it was completed (about 2 weeks ago) and I think it is up about 5%.

 
Maji:
By the way, Aegis, I am sorry to bring it up here, but you haven't replied back the emails I have sent you. You owe me money for the EA I had worked up for you. By the way, I have been forward testing it with very conservative settings from the day I told you it was completed (about 2 weeks ago) and I think it is up about 5%.

Fascinating....5%....not bad. LMAO

 
davidke20:
Fascinating....5%....not bad. LMAO

Risking 1% per trade, and making 5% per 2 weeks is not bad in my eyes. If you can make 5 to 10% per month consistently, with about 20% drawdown, I think it is great. I am sure there are systems that make 100000% per year, but then I try not to be greedy .

Now if you give me such a system, LOL... you can then see me get real greedy

 
Maji:
Risking 1% per trade, and making 5% per 2 weeks is not bad in my eyes. If you can make 5 to 10% per month consistently, with about 20% drawdown, I think it is great. I am sure there are systems that make 100000% per year, but then I try not to be greedy . Now if you give me such a system, LOL... you can then see me get real greedy

Ok dude. We're programmer. I understand what you mean actually. Basically I'm doing hedge mode on my API. Monthly return about 5~25%, drawdown maintain at below 3%, is that consider good? I'm just too interested on Martingale system and cant take my eyes off it. Therefore, I'm still working with martingale system now. I wonder why MT4's broker like to hunt support and resistance. I've never face this problem on my API direct feed. Therefore, I'm trying to make an EA that can avoid incident created by widen spread(therefore the TP/SL did not honour by the broker). And also, MT4 sucks at certain stage, because the backtest data is not reliable.

Regards,

David

Below attached my screenshot of backtest. Have a good laught at it.