How Long Is the Survival Rate of Martingle EA?

blackmore
434
blackmore  

Does anyone here use martingale ea, or martingle-based strategy in their ea? What is the survival rate like, or how long have you use them?

It seems like it is mathematically NOT impossible to implement this successfully -- hope to hear your take on this.

This idea is derived from the St.Petersburg Paradox --> https://en.wikipedia.org/wiki/St._Petersburg_paradox

which concludes that the traders (or players) always wins the house (market)... what do you think?

Ubzen
5300
Ubzen  

My response as always, it depends on the System and the Trader.

If you're playing a game of "flip-coin" where you won whatever-you-bet for head; and lost whatever-you-bet for tails. How much should you risk/bet? Ans=Nothing.

If you're playing a game of get "blue gum-ball out the machine of 2-gum-balls (red & blue). How much should you risk on the first try? Ans=The Minimum.<--(I'm not sure about the math on that :)). How much should you risk on the second try? Ans=Everything.

If you're playing a game of "flip-coin" where you won 1/2-of-whatever-you-bet for head; and lost whatever-you-bet for tails. How much should you risk/bet? Ans=Run Away. Money-Management Cannot help you here. If the situation was flipped, then you should bet 50% of Everything you have. Again- (I'm not sure about the math on that).

In the Gum-ball example Martingale is Under-Betting.

In the Run-Away Example, Martingale will Kill you faster than Flat-Betting the Minimum.

In the Advantage Coin Example, Martingale or some form of Progression would be Just about Right.

Using the Gum-Ball Betting in the Advantageous Coin Example will Ruin you before you get to Profit.

And Finally, In the First Coin Flip Example 50:50 Martingale could Kill or Make you depending on how Big the Bank is.

So the answer to if Martingale is good or bad Depends on your Edge. IMO.

-Ps- whats with the interest in Martingale lately? Edges to justify martingale don't exist in Forex AFAIC. I don't even Know if Forex really have an Edge that can be Mathematically Guaranteed.

blackmore
434
blackmore  
ubzen:

My response as always, it depends on the System and the Trader.

If you're playing a game of "flip-coin" where you won whatever-you-bet for head; and lost whatever-you-bet for tails. How much should you risk/bet? Ans=Nothing.

If you're playing a game of get "blue gum-ball out the machine of 2-gum-balls (red & blue). How much should you risk on the first try? Ans=The Minimum.<--(I'm not sure about the math on that :)). How much should you risk on the second try? Ans=Everything.

If you're playing a game of "flip-coin" where you won 1/2-of-whatever-you-bet for head; and lost whatever-you-bet for tails. How much should you risk/bet? Ans=Run Away. Money-Management Cannot help you here. If the situation was flipped, then you should bet 50% of Everything you have. Again- (I'm not sure about the math on that).

In the Gum-ball example Martingale is Under-Betting.

In the Run-Away Example, Martingale will Kill you faster than Flat-Betting the Minimum.

In the Advantage Coin Example, Martingale or some form of Progression would be Just about Right.

Using the Gum-Ball Betting in the Advantageous Coin Example with Ruin you before you get to Profit.

And Finally, In the First Coin Flip Example 50:50 Martingale could Kill or Make you depending on how Big the Bank is.

So the answer to if Martingale is good or bad Depends on your Edge. IMO.

-Ps- whats with the interest in Martingale lately? Edges to justify martingale don't exist in Forex AFAIC. I don't even Know if Forex really have an Edge that can be Mathematically Guaranteed.


So I think what you mean is the first entry should always be min. lot. Then subsequent entries should keep increasing.

Of course, as you mention, with unlimited resources, no problem. But does that mean, with limited resource, it will fail in matter of time? And how much initial deposit you think require if ever this to work out?

Can it ever be possible for Martingle to last more than 1,2,3,5 or even 10 years, given small deposits, limited resource?

Ubzen
5300
Ubzen  

So I think roughly what you mean is from your answer the first entry should always be min. No....But if you're going to martingale then you better not start at the Maximum ;).

How much initial deposit you think require? How many losses can the system have in a roll... I mean would it be possible to have 1 more loss after that?

And also, is it possible for Martingle to last more than 1,2,3,5 or even 10 years??? Yeah, if the broker Ask is < the Bid. Within the maximum number of tries you calculated of course. If this patten ever changes, you're screwed.

In the real world, the house usually sets up a Maximum (Lot). One of the reasons its done is to combat some Billionaire from going martingale Happy on em. Tho he'd outta be out of his mind to even try.

Correction..... its broker Ask<Bid+Spreads. Or Vice-Versa.

blackmore
434
blackmore  
ubzen:

So I think roughly what you mean is from your answer the first entry should always be min. No....But if you're going to martingale then you better not start at the Maximum ;).

How much initial deposit you think require? How many losses can the system have in a roll... I mean would it be possible to have 1 more loss after that?

And also, is it possible for Martingle to last more than 1,2,3,5 or even 10 years??? Yeah, if the broker Ask is < the Bid. Within the maximum number of tries you calculated of course. If this patten ever changes, you're screwed.

In the real world, the house usually sets up a Maximum (Lot). One of the reasons its done is to combat some Billionaire from going martingale Happy on em. Tho he'd outta be out of his mind to even try.

Correction..... its broker Ask<Bid+Spreads. Or Vice-Versa.

I would have the same answers as you, if I were to consider the practicalities, just as anyone would.

Mathematically however, St.Peter's is saying the market has a Shadow Guarantee, that is unbeknowned yet existing, in order for it to operate not for the current, but for the future. This is where the paradox lies - neither true but also not false, or both at the same time. And the math behind that paradox goes, "if traders are losing, its good for traders, and bad for the brokers, because if the traders keep on making same calls, the market will eventually crash to the calls. But because the brokers don't want the market to crash to those calls, (as it will put them out of business), the traders will have to win."

So, I think martingale can be efficient, along this line. But math aside, I was hoping for some brilliant, ingenius coders, who has actually implemented this successfully into EAs.

blackmore
434
blackmore  
Anyone else??