Why martingale doesn't work in trading | Page 4 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Martingale, Anti-martingale, and Compounding 40 replies

Martingale, Reverse Martingale, Modified Martingale, Maths 1 reply

Martingale Calculation EA (also reverse martingale) 3 replies

How to change this Martingale to Reverse Martingale ? 3 replies

Martingale vs. Non Martingale (Simplified RoR vs Profit) 1 reply

Attachments: Why martingale doesn't work in trading
Exit Attachments

Why martingale doesn't work in trading

  • #61
  • Aug 27, 2010 4:08pm Aug 27, 2010 4:08pm
It's gambling. If you trade martingale you are gambling not trading. Full stop!
  • #62
  • Aug 27, 2010 4:24pm Aug 27, 2010 4:24pm
It's enticing because you can get get quite the string of winners and start to believe you've found a way to beat the universe. But disaster is sure to come.
Open to new approaches.
  • #63
  • Aug 27, 2010 8:08pm Aug 27, 2010 8:08pm
Quoting redcaesar
I am no fan of martingale, but I think there is a chance how he can work for some people (sort of)
......
So if someone is betting this strategy with martingale, he will see increased performance, because he is risking an amount far from optimal in first place.

Conclusion: if martingale works for you double your lots and forget the martingale
Yes, but this is a thread about martingale and how some hope to use it to benefit themselves. In your example, it is the increased turnover, and not martingale, that is responsible for the increased profit. As you say yourself, why not just double the bet size in the first place? However, I'm not actually arguing against martingale, as like any other staking strategy it just reshapes the distribution of profit and loss. If someone understands this and prefers consistent gains with the occasional big loss, I'm not going to advise them against it.
  • #64
  • Sep 1, 2010 4:04pm Sep 1, 2010 4:04pm
Hello Everyone,

Well in my mind the problem of the Martingale is a very significant one in trading. I think that a very good argument that can be put forward here is that there has yet not been a single long term profitable martingale which has been shown to work in trading.

Let me explain myself better, there have been thousands of martingale systems that have been tried and failed with time (some even showing profits for a few years) but there has been no martingale that has ever been able to trade without total loss of funds through a significant period of change in long term volatility (more than 10 years) with profitable results, at least to the best of my knowledge this system doesn't exist.

So the facts here are quite simple to interpret :

 

  1. Countless martingales have lost accounts and large amounts of funds
  2. There is no martingale that has been profitable for at least ten years (or such a system is not made public or published anywhere)
  3. For a martingale to succeed a system is needed where the probability to have a favorable outcome on the next trade increases with the number of losses.

Now the facts here tell us that we would need to have a system which would have an implicit increase in its statistical probability to win after every negative outcome, something which cannot be proved to not exist (which means it might be possible).

However there are countless systems that have been proved to work for more than 10 years that are used by traders all over the world which have shown to resist changes in market conditions and use sound money management principles. Donchian channel systems are a good example of this.

So the question becomes : would you rather pursue an extremely risky venture that no one has publicly been able to succeed at or would you rather use trusted systems that have shown to work through a myriad of market conditions ? The answer for most professional traders here is obvious.

As always remember the saying :

"there are old and bold traders but there are no old bold traders"

I hope this helps the discussion

Best Regards,

Daniel

  • #65
  • Feb 13, 2012 11:36am Feb 13, 2012 11:36am
I see everybody here is telling martingale never work coming from mathematics and probabilistic theory. If you say according to mathematics this will never work you must accept first market is truly random which means what happened no way effects the future. If so and market is truly random it means not only martingale will not work also none of the technical analysis will also not work.

So if you say according to probability or laymans terms gambling martingale is a suicide then you must accept market is truly random which means we should stop trading because technical analysis should not work in a random market.
  • #66
  • Edited 6:33pm Feb 13, 2012 5:50pm | Edited 6:33pm
Quoting iwontsmoke
I see everybody here is telling martingale never work coming from mathematics and probabilistic theory. If you say according to mathematics this will never work you must accept first market is truly random which means what happened no way effects the future. If so and market is truly random it means not only martingale will not work also none of the technical analysis will also not work.

So if you say according to probability or laymans terms gambling martingale is a suicide then you must accept market is truly random which means we should stop...
Money management (position sizing) and market timing (entries and exits) are two completely different things.

The market is not entirely random. It is possible to be long-term, on-balance profitable using TA to determine entries and exits.

Money management in itself can not provide an edge, whether trading or playing casino games. This can be proved mathematically. To benefit from varying your position sizes, you need to know in advance which trades are more likely to be winners, and size them commensurately higher, in which case you must already have an edge. Otherwise you are merely playing a lucky game of hit or miss.
  • #67
  • Feb 13, 2012 6:20pm Feb 13, 2012 6:20pm
Depends actually. Unlike random coin toss in the market probability of decline of price increases the longer the price increases. But this does not give any info on how much the market decline or reverse. I have seen 400 pips moves in couple of hours but never seen 1500 pips move continuously so far.

Does that mean it cant be? No ofcourse not. we can see that kind of move but it is highly unlikely. Because market is not pure mathematical equation. There is an emotional side of it.

We all know what happened with Long-Term Capital Management L.P.

It is you who decides when to enter that trade. And this brings chaos to the equation.

Thats why i may succeed using martingale based on my entry (which is chaos) or I may fail horribly.

Not everything is deterministic. Those kind of newtonian thinking can solve daily life problems but economics is alot more complex than that.
  • #68
  • Feb 13, 2012 6:41pm Feb 13, 2012 6:41pm
Quoting iwontsmoke
Depends actually. Unlike random coin toss in the market probability of decline of price increases the longer the price increases. But this does not give any info on how much the market decline or reverse. I have seen 400 pips moves in couple of hours but never seen 1500 pips move continuously so far.

Does that mean it cant be? No ofcourse not. we can see that kind of move but it is highly unlikely. Because market is not pure mathematical equation. There is an emotional side of it.

We all know what happened with Long-Term Capital Management L.P.

It...
Agree that market is not pure mathematical equation, due to a wide number of factors. But these factors (e.g. supply and demand, macroeconomics, self-fulfilling prophecy, triangular equilbrium, etc) create trends and patterns that are exploitable.

MM strategies like martingale makes no sense, for the reason that I gave earlier. Increasing position size simply to recover from recent losses represents an emotional, rather than a mathematically justifiable response: market behavior and probabilities aren't going to change simply because one insignificant retail trader has had a recent losing sequence. All martingale does is increase risk of ruin, without improving trade expectancy.
  • #69
  • Feb 13, 2012 6:53pm Feb 13, 2012 6:53pm
I understand what you are telling.

first of all i am not suggesting a pure MM.

You may have a winning system if you can enter the patern at the correct time. But sometimes market does a tail and gets you out and then makes the expected move.

So market moved according to mathematics, pattern etc. And your prediction was right. But timing was a little off..

There enters the martingale. If you can master it and if your entries are high probability entries martingale gives you a second, third ... chance.

By martingale I am not exactly saying doubling up. What i mean is increasing the position size to cover off timed but correct entries. The amount of increase should be more than your system's historic loosing streak.

Not every system can be used with martingale. And it needs a little luck that a historic event will not take place. Also your calculated risk should be correct.

Am i using it? No. but there can be a way to use it with the right set of tools.

Again it is a possibility.
  • #70
  • Feb 13, 2012 7:27pm Feb 13, 2012 7:27pm
Quoting iwontsmoke
There enters the martingale. If you can master it and if your entries are high probability entries martingale gives you a second, third ... chance.
IWS,

You still get a second or third chance, no matter how you size your positions.

However, if, as you say, the probability of reversion to a mean increases the further price moves away from the mean, then an increase in position size is sound. But, to be mathematically justifiable, it would need to be commensurate with the improvement in the probability being offered by the market, as opposed being based on the size needed to recover the trader's open losses. And you would need to prove the mean reversion statistically, though; I suspect the "don't add to losses" mantra that's bandied around by veteran traders is the result of hard experience.

Increasing position size merely to recover recent losses is nothing more than an emotional response, and will inevitably end in disaster, because emotions will increase the deeper the account spirals into drawdown.

Most martingales are sized around what the worst seen adverse move. Like you say, just because 1500 pips hasn't occurred, doesn't mean that it won't in the future, even if it is highly unlikely. Nonetheless, IMO that is gambling rather than trading; because just one 'black swan' event is all that's needed to cause ruin. If I simply size at (for example) 1% per trade, then I guess I'm likewise gambling that my losses will never exceed my wins by 100, which would likewise cause ruin. But if my entries/exits provide a decent edge, that's unlikely to happen; and even if it does, I have plenty of time to suspend my trading and re-evaluate (and if necessary, overhaul) my strategy before the drawdown gets seriously out of hand. For me, that's the important difference.

To each their own; everybody must proceed with whatever they feel most comfortable. I am conservative by nature, hence risk management is my #1 priority. But it's not for me to tell anybody else what they should or shouldn't do.

Thanks for the discussion. I wish you every success.

David
  • #71
  • Feb 13, 2012 7:36pm Feb 13, 2012 7:36pm
As I said I am not using it either but I am not also saying 100% it will fail. I guess it will always be in my mind when constructing new ways of trade.

What we discuss here is mainly philosophy and it never ends with a certain decision. Aim of discussion is only to open new horizons to our thinking

I certainly enjoyed our discussion.

Thank you.
  • #72
  • Oct 20, 2017 8:35pm Oct 20, 2017 8:35pm
I have recently started the matriangle strategy with smaller lot,but I use it after studying the market and when I know I'm following the trend and only 2 bets,first 0.01lot and another is 0.03.if it doesn't work,i just quit this pair and don't keep on increasing the lots.i think it is a good strategy if used with precaution and without greed.
  • #73
  • Last Post: Feb 15, 2019 4:11pm Feb 15, 2019 4:11pm
Quoting T4Trade
I have recently started the matriangle strategy with smaller lot,but I use it after studying the market and when I know I'm following the trend and only 2 bets,first 0.01lot and another is 0.03.if it doesn't work,i just quit this pair and don't keep on increasing the lots.i think it is a good strategy if used with precaution and without greed.
Are you still using your martingale system today?
1 trader viewing now