Is Martingale really dangerous? | 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: Is Martingale really dangerous?
Exit Attachments

Is Martingale really dangerous?

  • #1
  • First Post: Jan 24, 2008 8:34am Jan 24, 2008 8:34am
I think that the most of you know the Martingale priciple (doubling up when you are loosing). Certainly most of you have done it already (I know I have!) and know it is no good.

When using Martingale, you open to much lots in the wrong position, the more the position goes wrong, the more lots you open. This way you are overexposed on a single pair that probably will keep going in the wrong direction (mini-trend, correction on the trend) and when it finally turns around, you will already got a margin call. So I think that it is already established and proven that Martingale doesn't work!

But here is my view on the subject: If you are a trader who trades 3-4 pairs on different timeframes and achieve a winning percentage of more than 51%. What if you double up after a lost trade, assuming you won't add positions to a loosing position. You should make the lost pips on one trade back or not?

Maybe a little example will help:

Trader: 60% winning average on trades
Stoploss is 25 TP is 50
For example reasons 1 pip=1USD

1 MiniLot LONG @ EURUSD ----> Lost 25 pips
2 MiniLot SHORT@ GBPUSD----> Won 50 pips per lot (=100 pips)
1 MiniLot SHORT@ USDCHF----> Won 50 pips
1 MiniLot SHORT@ USDCHF----> Lost 25 pips
2 MiniLot LONG @ NZDJPY-----> Lost 25 pips per lot (=50 pips)
4 MiniLot LONG @ EURUSD-----> Won 50 pips per lot (=200 pips)

This works out quite well for the trader! So can we assume that when a trader has an above XX% winning average and always sets stoplosses AND NEVER TRADE THE SAME PAIR OR A CORRELATED PAIR after a previous loss, we can use Martingale as a moneymanagement system?

Just I tought! I'm really curious what you guys think....
  • #2
  • Jan 24, 2008 8:49am Jan 24, 2008 8:49am
you know something, this method works for me really well, but i have some extra rules.

lets say, uj was on 114. probably that was an extreme, but then the next days fall to 109, open a position, if the market goes against me, lets say 200 points, open a bigger one who lets close the upper position fast. with just a few pips in profit. but also not risking much of the account.
then close the position and open another little one.

lets say the market fall another 200 points, i open a bigger one.

just remember yesterday, opened the big guns on 105.10 with 5lots.
tp 106. made some nice bucks in the explosion.

of course not opening a position just for making a trade, just when yesterday could saw the dow jones reacted to support very strongly after the 2nd selloff in the morning, just knew was time to enter.

so it can be martingale + another thing.

sometimes, even the little entry i make for testing using martingale method reach the tp fast, so cant apply the method, but its like having insurance.

other important aspect i think is not not doing it on eu, gu, cause those currencies could never see again the lands they are now. at least i am going with the strongest currency in a really oversold area.

my conclusion after trading this for months, yen pairs work better. avoiding gj can be a good idea or the account is not going to have much life.
---------------------------------------------------
could add : for this method probably is needed steel nerves or big balls. be patient and have sharp hedging skills when something really bad happens, but here is the interesting, its like a video game, you dominate the market, or the market kills you.
if you win, the profits can be big.
---------------------------------------------------
  • #3
  • Jan 24, 2008 8:50am Jan 24, 2008 8:50am
Martingale position sizing does work when it is applied to the correct method as you have already hinted.

When you come from Futures trading environment where most traders have a negative reward-to-risk ratio combined with high percentage win/lose ratio (70% or higher) based on a multi opportunity method, losing trades can often be compensated by doubling the lot size on the next trade.

It does not have to be in different currency and it doesn't have much to do with the prvailling trend.

Most futures traders don't look much at charts and if they do, its usually in a different way that most TA related traders do.

regards
daytrading
Enter Signature
  • #4
  • Jan 24, 2008 8:55am Jan 24, 2008 8:55am
IN MY opinion, suicide.
Make a Research about "kelly criteria" , is similar, but less agressive


Best Regards,
Marcelo Marques
  • #5
  • Jan 24, 2008 8:58am Jan 24, 2008 8:58am
Quoting daytrading
Martingale position sizing does work when it is applied to the correct method as you have already hinted.

When you come from Futures trading environment where most traders have a negative reward-to-risk ratio combined with high percentage win/lose ratio (70% or higher) based on a multi opportunity method, losing trades can often be compensated by doubling the lot size on the next trade.

It does not have to be in different currency and it doesn't have much to do with the prvailling trend.

Most futures traders don't look much at charts and if they do, its usually in a different way that most TA related traders do.

regards
daytrading
Interesting, I didn't knew that. Thanks...
  • #6
  • Jan 24, 2008 8:59am Jan 24, 2008 8:59am
Quoting marcelom
IN MY opinion, suicide.
Make a Research about "kelly criteria" , is similar, but less agressive


Best Regards,
Marcelo Marques
I'll look it up, thanks...
  • #7
  • Jan 24, 2008 9:01am Jan 24, 2008 9:01am
Quoting acidguy
you know something, this method works for me really well, but i have some extra rules.

lets say, uj was on 114. probably that was an extreme, but then the next days fall to 109, open a position, if the market goes against me, lets say 200 points, open a bigger one who lets close the upper position fast. with just a few pips in profit. but also not risking much of the account.
then close the position and open another little one.

lets say the market fall another 200 points, i open a bigger one.

just remember yesterday, opened the big guns on 105.10 with 5lots.
tp 106. made some nice bucks in the explosion.

of course not opening a position just for making a trade, just when yesterday could saw the dow jones reacted to support very strongly after the 2nd selloff in the morning, just knew was time to enter.

so it can be martingale + another thing.

sometimes, even the little entry i make for testing using martingale method reach the tp fast, so cant apply the method, but its like having insurance.

other important aspect i think is not not doing it on eu, gu, cause those currencies could never see again the lands they are now. at least i am going with the strongest currency in a really oversold area.

my conclusion after trading this for months, yen pairs work better. avoiding gj can be a good idea or the account is not going to have much life.
---------------------------------------------------
could add : for this method probably is needed steel nerves or big balls. be patient and have sharp hedging skills when something really bad happens, but here is the interesting, its like a video game, you dominate the market, or the market kills you.
if you win, the profits can be big.
---------------------------------------------------

Ive often do it your way, but I try to evade it! It's dangerous, but is sure is fun!!!
  • #8
  • Jan 24, 2008 9:02am Jan 24, 2008 9:02am
Quoting acidguy
you know something, this method works for me really well, but i have some extra rules.

lets say, uj was on 114. probably that was an extreme, but then the next days fall to 109, open a position, if the market goes against me, lets say 200 points, open a bigger one who lets close the upper position fast. with just a few pips in profit. but also not risking much of the account.
then close the position and open another little one.

lets say the market fall another 200 points, i open a bigger one.

just remember yesterday, opened the big guns on 105.10 with 5lots.
tp 106. made some nice bucks in the explosion.

of course not opening a position just for making a trade, just when yesterday could saw the dow jones reacted to support very strongly after the 2nd selloff in the morning, just knew was time to enter.

so it can be martingale + another thing.

sometimes, even the little entry i make for testing using martingale method reach the tp fast, so cant apply the method, but its like having insurance.

other important aspect i think is not not doing it on eu, gu, cause those currencies could never see again the lands they are now. at least i am going with the strongest currency in a really oversold area.

my conclusion after trading this for months, yen pairs work better. avoiding gj can be a good idea or the account is not going to have much life.
---------------------------------------------------
could add : for this method probably is needed steel nerves or big balls. be patient and have sharp hedging skills when something really bad happens, but here is the interesting, its like a video game, you dominate the market, or the market kills you.
if you win, the profits can be big.
---------------------------------------------------

Ive often do it your way, but I try to evade it! It's dangerous, but sure, it is a lot of fun!!!
  • #9
  • Jan 24, 2008 9:08am Jan 24, 2008 9:08am
I don't know if this is considered martingale

Let's say you decide to enter a trade, you already determined that you will only risk 2% total.

So intial entry you only put up 0.5%. When price goes down, you enter another 0.5%..keep going until you get 2% (about 4 positions) and you just wait and see how the trade plays out

I think this method gives a better chance of minimizing your losses from regular martingale but at the same time maximizing profits when the trade is right.
  • #10
  • Edited 10:46am Jan 24, 2008 9:17am | Edited 10:46am
Quoting hilmy83
I don't know if this is considered martingale

Let's say you decide to enter a trade, you already determined that you will only risk 2% total.

So intial entry you only put up 0.5%. When price goes down, you enter another 0.5%..keep going until you get 2% (about 4 positions) and you just wait and see how the trade plays out

I think this method gives a better chance of minimizing your losses from regular martingale but at the same time maximizing profits when the trade is right.
No, this is averaging into a position and IMO a very good idea. if you are trying to pick a range. I use the approach myself.. But it is not Martingale, because it is not a negative progression approach.

BTW Kelly has nothing to do with Martingale, for the same reason. It is a positive progression, not negative
  • #11
  • Jan 24, 2008 9:19am Jan 24, 2008 9:19am
Quoting smjones
No, this is averaging into a position and IMO a very good idea. if you are trying to pick a range. I use the approach myself.. But it is not Martingale, because it is not a negative expectancy approach.
Indeed.

SMJones, would you consider the Martingale-esque Money Management in the opening post as a system with a negative expectancy?
  • #12
  • Jan 24, 2008 9:19am Jan 24, 2008 9:19am
Quoting hilmy83
I don't know if this is considered martingale

Let's say you decide to enter a trade, you already determined that you will only risk 2% total.

So intial entry you only put up 0.5%. When price goes down, you enter another 0.5%..keep going until you get 2% (about 4 positions) and you just wait and see how the trade plays out

I think this method gives a better chance of minimizing your losses from regular martingale but at the same time maximizing profits when the trade is right.
Hi hilmy83,

Do you employ this kind of method in your fund? Just curious.

Regards,

Ed.
  • #13
  • Edited 10:47am Jan 24, 2008 9:22am | Edited 10:47am
Quoting Sidus
Indeed.

SMJones, would you consider the Martingale-esque Money Management in the opening post as a system with a negative expectancy?
I corrected my post to say negative progression, not expectancy. expectancy has nothing to do with MM. Just like the casino's edge on their games. No MM will defeat negative expectancy in the long run.
  • #14
  • Jan 24, 2008 9:24am Jan 24, 2008 9:24am
Quoting smjones
I corrected my post to say negative regression, not expectancy. expectancy has nothing to do with MM. Just like the casino's edge on their games. No MM will defeat negative expectancy in the long run.
No prob, mate!
  • #15
  • Jan 24, 2008 9:26am Jan 24, 2008 9:26am
Quoting Ted1983
Hi hilmy83,

Do you employ this kind of method in your fund? Just curious.

Regards,

Ed.

Yes I do actually . But I caution that people who use this type of trading needs to trade on higher timeframs (preferably 1hr>) since the spread might cut significantly into your profits if you trade on lower timeframes.
  • #16
  • Jan 24, 2008 9:27am Jan 24, 2008 9:27am
Quoting hilmy83
I don't know if this is considered martingale

Let's say you decide to enter a trade, you already determined that you will only risk 2% total.

So intial entry you only put up 0.5%. When price goes down, you enter another 0.5%..keep going until you get 2% (about 4 positions) and you just wait and see how the trade plays out

I think this method gives a better chance of minimizing your losses from regular martingale but at the same time maximizing profits when the trade is right.
Hilmy,

What you describe is not martingale but adding to a losing trade.

Martingale is originally devised for a straight forward probability game with fixed odds in respect of the law of large numbers. (like roulette)

Which means each throw counts as one trade. You have odd/even or red/black (forget the gree zero for a minute)

The greater the amount of throws you can do the better because the law of large numbers dictates that the 50/50 chance is more certain with the growing number of attempts made. (i.e. if you play roulette and you go for red, but you only play 10 times, it could come up black ten times despite the chance of 50/50 -- if you play 1000 times, you are almost guaranteed that you have apprximately 50% or 500 times red and 50% black)

With respect to Martingale, if you size your inital position small enough that you can have as many try's as possible to play red for example, and you double your position for as long as it takes to get your first red, than you come out on top every time.

Most people's trading method won't allow that because most trading methods are based on a win/lose ratio with a high reward/risk ratio, where there is no need for Martingale.

regards
daytrading
Enter Signature
  • #17
  • Jan 24, 2008 9:31am Jan 24, 2008 9:31am
Sorry folks, I have to correct my last sentence.

it should say....low win/lose ratio with high reward/risk ratio.

regards
daytrading
Enter Signature
  • #18
  • Jan 24, 2008 9:43am Jan 24, 2008 9:43am
I would like to hear some more reactions to my approach? I cannot find any disadvantages, I would love to be proven wrong!
  • #19
  • Jan 24, 2008 9:57am Jan 24, 2008 9:57am
Problem is bad streak of loss.

I have a pretty high success rate (83.6% if i look my last statement) but if i look in my december statement i had 10 successive loss in a row. Using a martingale strategy i dont even know if i would have had enough margin for the 10th order.
  • #20
  • Jan 24, 2008 9:59am Jan 24, 2008 9:59am
Quoting melpheos
Problem is bad streak of loss.

I have a pretty high success rate (83.6% if i look my last statement) but if i look in my december statement i had 10 successive loss in a row. Using a martingale strategy i dont even know if i would have had enough margin for the 10th order.
Indeed, enough balance and small lot size would be also a very important factor...
0 traders viewing now