No Loss Martingale Strategy | Page 6 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Big Loss Martingale Strategy! 12 replies

Martingale, Anti-martingale, and Compounding 40 replies

Martingale, Reverse Martingale, Modified Martingale, Maths 1 reply

How to change this Martingale to Reverse Martingale ? 3 replies

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

Attachments: No Loss Martingale Strategy
Exit Attachments

No Loss Martingale Strategy

  • #101
  • Sep 17, 2015 11:11am Sep 17, 2015 11:11am
Quoting fx4money
i have plenty of my own versions of martingale, need a real good broker this strategy to work. plenty of bad brokers around, they use plenty of ways to fail you with requote, offquote, server disconnection. no matter how disciplined you are, or how good is your entry, if the broker is not honest, they will eat you up.
have you been success? i use tradersway, what broker do you use? and do you know about exness and their cent account?
  • #102
  • Sep 17, 2015 11:16am Sep 17, 2015 11:16am
i use martingale to trade for small percentage of profit around 5-10% per month, use small margin so i can sleep well even with positions opened, my accounts managed to survive for about 2 years few years back.
LUCKY golden trades, LUCKY gold investor trading style
  • #103
  • Edited 1:41pm Sep 17, 2015 1:05pm | Edited 1:41pm
Quoting She1
{quote} All right, are you other guys agree with me and MathTrader7 that if i have a strategy that wins at least 1 trade in 5 trades, don't you think we are success to create "No Loss Martingale Strategy"?
I just have to mention that it works under the following conditions:
1) we have a system that wins at least one trade of every N consecutive trades;
2) the system has fixed stop loss and take profit for every trade (if the lot size multiplier is fixed).

For example, a system that guarantees at least one win in every 5 consecutive trades with a stop loss of 20 pips and a take profit of 20 pips that uses the following sequence of lot sizes proportional to L=(1,2,4,8,16) can be called "No Loss Martingale Strategy". However, I believe that there is no such system to work in long run.
Trading is the hardest way to make easy money...
  • #104
  • Sep 17, 2015 1:59pm Sep 17, 2015 1:59pm
Attached Image
  • #105
  • Sep 17, 2015 3:40pm Sep 17, 2015 3:40pm
i just want to post this to start a discussion about what is writen bellow:

" Mostafa Belkhayate proposes a position sizing rule and pretends that this rule will lead to marvellous gains.

This system works a follows:
- if last trade was a loser, position size for this trade is the same as for the previous trade (no doubling)
- if last trade was a winner (and the one before was a loser), position size for this trade is the same as the cumulated loss until now, plus one unit
- if two last trades were both winners, we have obtained a cumulated gain of +1, and we stop.

Mostafa Belkhayate takes the exemple of a coin toss (%win = 50% ; Win/Loss = 1). The expectancy of this system is 0. But Mostafa Belkhayate pretends that his position sizing system will turn it into a (very) profitable system.

He asserts that the probability of not having 2 consecutive winners in a sequence of 12 trades is negligible, so the trader is “guaranteed” to finish the sequence with a gain of +1."

MB is a successful Moroccan trader or at least this is what they say about him, he concluded that the market is unpredictable and even if you have a profitable system , you cannot win as a retail trader, the only way to avoid failure according to him is to use Math to define a position sizing that will allow a trader to overcome losing trades.

what you guys think about this statement and his mathematical expectancy theory... i have videos (in french) where he explains his method.--if someone intersted i ll post them
  • #106
  • Edited Sep 18, 2015 4:22pm Sep 17, 2015 5:51pm | Edited Sep 18, 2015 4:22pm
Quoting ollaya
i just want to post this to start a discussion about what is writen bellow: " Mostafa Belkhayate proposes a position sizing rule and pretends that this rule will lead to marvellous gains.
The XLS inside the attached ZIP file debunks his claims. It tests his staking (position sizing) system over 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11 and 12 trials.

 The first column, in each group of three columns (e.g. columns C,D,E), calculates the stake size, according to his formula, based on the previous trials.
 The second column is the outcome (Win or Loss). Rows 3 thru 258 cover every possible permutation of outcomes, for up to 12 trials.
 The third column is the accumulated profit/loss over the current and all prior trials, calculated from the stake size and the outcome.

Note that the total expectancy (row 259) is always zero, for every one of the 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11 and 12 trials. You can extend the XLS to cover 13 or more trials, the result is still the same. QED.

You can try this with ANY betting system: martingale, anti-martingale, d'Alembert, Guetting, Labouchere, Oscar's Grind, Paroli, One Half Up, Parlay, 1-3-2-6, ...... whatever (see the other attached XLSs). Providing that each trial is independent (uncorrelated), and hence that each outcome is equally probable, then no matter now many trials, the overall expectancy for every possible betting system is always the same as for "flat betting" (i.e. the betting sequence 1,1,1,1,1, ........)

Dr Mike Shackleford (the 'Wizard of Odds') reaches the same conclusion: "All betting systems are equal to flat betting when compared this way, as they should be. In other words, all betting systems are equally worthless. ........ I no longer respond to e-mails that suggest a player can beat a negative expectation game over the long run with a betting system. Such e-mail is deleted on sight." Very simply, providing that each trial is independent, position sizing systems can never provide an edge. The math is inviolable, as PipMeUp further demonstrates here.

Hence Mr Belkhayate is wrong about two things:
Quoting ollaya
(1) you cannot win as a retail trader,
(2) the only way to avoid failure according to him is to use Math to define a position sizing that will allow a trader to overcome losing trades.
(1) It is possible to win, and by simply using consistent position sizing (1,1,1,1,1,1,....).

(2) As we have demonstrated, position sizing systems have no effect on long term overall expectancy. Moreover, progressive sizing systems increase average position size, which accelerates risk of ruin.
Attached File(s)
File Type: xls Martingale comparison.xls   126 KB | 1,057 downloads
File Type: xls Martingale comparison testing.xls   157 KB | 973 downloads
File Type: zip Belkhayate expectancy debunked.zip   999 KB | 1,102 downloads
  • #107
  • Sep 17, 2015 7:44pm Sep 17, 2015 7:44pm
Quoting hanover
{quote} The first attached XLS debunks his claims. It tests his staking (position sizing) system over 1, 2, 3, 4, 5, 6, 7 and 8 trials.  The first column, in each group of three columns (e.g. columns C,D,E), calculates the stake size, according to his formula, based on the previous trials.  The second column is the outcome (Win or Loss). Rows 3 thru 258 cover every possible permutation of outcomes, for up to 8 trials.  The third column is the accumulated profit/loss over the current and all prior trials, calculated from the...
i think there is a slight misunderstanding of Belkhayte postion sizing theory ( or Alembex is he named). the file you attached does not reperesent the method MB uses . MB position sizing uses a sequence logic that end when you have to winner in a row and you trade the recent lot size when you have a loser . please see the attached file for an example of a sequence of trades that ended with WW.

let me know what do you think
Attached File(s)
File Type: xlsx belkhayte expactancy Alembex.xlsx   9 KB | 891 downloads
  • #108
  • Edited 8:14pm Sep 17, 2015 8:03pm | Edited 8:14pm
Quoting hanover
{quote} The first attached XLS debunks his claims.
Thanks for that dose of clarity David.

The old unicorn pops up from time to time when I see those seductive equity curves associated with forms of Martingale progression. I don't know how many times I have stood at the cliff's edge and thought I could fly.

Having come undone with Martingale's in the past, I still dabble from time to time with limited averaging down variants but only with overall positive expectancy....however they frequently still create drawdown drama's that I don't always have the cohonas to tolerate. It has forced me to just accept the fact that there are better more efficient ways using judicious position sizing with positive expectancy to avoid these unnecessary risk events that just makes me unpopular with the wife and kids and gives me an ulcer.

Each to their own of course....but they just give me heartburn despite their seductive allure. :-(
  • #109
  • Sep 17, 2015 8:56pm Sep 17, 2015 8:56pm
Quoting Copernicus
{quote} Thanks for that dose of clarity David. The old unicorn pops up from time to time when I see those seductive equity curves associated with forms of Martingale progression. I don't know how many times I have stood at the cliff's edge and thought I could fly. Having come undone with Martingale's in the past, I still dabble from time to time with limited averaging down variants but only with overall positive expectancy....however they frequently still create drawdown drama's that I don't always have the cohonas to tolerate. It has forced me...

i am not a big fan of martingale either and i had a bad experience adding to losers and hoping for price to range and come back to the starting point . i am just joining the discussion and trying to analyse this method that i think use different concept ; the creator claims that if you have a system that generate two wins in a row you should be able to close a sequence of trades not losing but wining +1.( which i think any trend following system can generate).

this is not a grid system ( random entries) . the intent is to use this position sizing with system that you already use ( ie MA crossing) and at the end you should have more chance to grow your account compared to using a fixed position sizing .

i did not verify this method to the fullest and it`s up for discussion
  • #110
  • Sep 17, 2015 9:17pm Sep 17, 2015 9:17pm
softer version of martingale is grid trading where we just put another trade with the same size or just weee bit bigger.
it takes lot longer time to blow our account.
i guess in real world martingale works better, especially with non perishable goods.
just because there is inflation. thus the selling price will always be higher.
  • #111
  • Sep 17, 2015 9:46pm Sep 17, 2015 9:46pm
Quoting ollaya
{quote} i am not a big fan of martingale either and i had a bad experience adding to losers and hoping for price to range and come back to the starting point . i am just joining the discussion and trying to analyse this method that i think use different concept ; the creator claims that if you have a system that generate two wins in a row you should be able to close a sequence of trades not losing but wining +1.( which i think any trend following system can generate). this is not a grid system ( random entries) . the intent is to use this position...
Yep, provided there is a positive expectancy there are ways to use Martingales and it's average cost variations....however...in my opinion they are not necessarily as efficient as other methods of money management with positive expectancy. The same less than optimal efficiency can be said about pyramiding into trades or a random scaling into trades. Each of these techniques despite the appearance of being very powerful forms of money management actually can dilute overall performance and increase drawdowns as you are spreading your risk exposure over a price action timeseries which has a dilutive effect as opposed to simply focussing on single trade entries that are specifically tailored for a particular price action event. IMHO there is a trade-off going on between trade efficiency and risk exposure which I think only makes itself apparent over large sample sizes.
  • #112
  • Edited Sep 18, 2015 2:03am Sep 17, 2015 10:38pm | Edited Sep 18, 2015 2:03am
Quoting ollaya
{quote} i think there is a slight misunderstanding of Belkhayte postion sizing theory ( or Alembex is he named). the file you attached does not reperesent the method MB uses . MB position sizing uses a sequence logic that end when you have to winner in a row and you trade the recent lot size when you have a loser . please see the attached file for an example of a sequence of trades that ended with WW. let me know what do you think {file}
Sorry, there was a bug in my staking calculation. I've fixed it, and also extended the original XLS to a max of 12 trials. The end result is still the same: a total expectancy of 0, no matter how many trials. The updated XLS is too big to be attached to a FF post, hence I've zipped it (and attached the ZIP file to my original post #106).
__________________________________________
[EDIT]
Quoting Copernicus
.....the fact that there are better more efficient ways using judicious position sizing with positive expectancy to avoid these unnecessary risk events.......
Um, but that's exactly the myth that I was (looking to) debunk. Unless you can get the larger size positions to coincide with higher probability trades, there's no such thing as "judicious" position sizing, because -- given the proofs and examples in post #106 -- every staking system delivers exactly the same long term expectancy as "flat betting". Staking systems merely affect risk of ruin; to improve expectancy, you need more accurate entries and/or exits.

Here's another way of looking at it: the P/L of each trade is the product of pos size and the number of pips between entry and exit, summed across all of the trades in your Terminal. To improve performance using pos sizing, then, you would need to get the larger pos sizes to coincide with the biggest pip-based winners, and vice versa. To know (at the time you place each trade) which trades are more likely to win the most pips, you must already have some kind of methodological edge; otherwise your position sizing amounts to little more than a lucky game of hit or miss. There is no mathematically justifiable reason for increasing pos size to recover recent losses: firstly, market behavior and probabilities don't change because of one insignificant trader's P/L; and secondly, recent losses are no more significant to overall bottom line than any other losses.

Even though no martingale variant can improve long term expectancy, (uncapped) martingale is arguably a special case in that all losses are quickly recovered, while the negative component of the expectancy is all gathered up in one outcome: the 'death sequence'. Hence, as the OP suggests, eliminate the DS and you have a holy grail: guaranteed continuous recovery with zero risk. The problem is that (1) the longer you trade, and (2) the greater the randomness in price movement, the higher the probability (I would say, inevitability) that you will eventually encounter the DS. Hence martingale becomes an exercise in being lucky enough to withdraw your winnings before the DS occurs, which realistically makes it suitable only for fooling around with nickel-and-dime size accounts. Otherwise, the risk of serious loss is too great.

Please feel welcome to reply, but I really don't want to take this further as I'm merely repeating many of my earlier posts (e.g. here), and also basic probability principles that are easily found on the internet. At the end of the day, people either believe in the validity of math, or they let their trading hopes and dreams convince them otherwise.
  • #113
  • Edited 3:12am Sep 18, 2015 2:33am | Edited 3:12am
Thanks Ollaya, have you ever used that system, or is there any EA for it. and can you link us to the strategy owner?
  • #114
  • Sep 18, 2015 3:24am Sep 18, 2015 3:24am
Quoting hanover
{quote} At the end of the day, people either believe in the validity of math, or they let their trading hopes and dreams convince them otherwise.
Yes, why the let the math(s) get in the way of a theory.

This may help the OP - impossible.
  • #115
  • Sep 18, 2015 4:33am Sep 18, 2015 4:33am
Quoting She1
Thanks Ollaya, have you ever used that system, or is there any EA for it. and can you link us to the strategy owner?
No, did not use it . Mostafa Belkhayte is a Moroccan trader and if you google it you will find only results in french.
SO if you understand french i can post the links here.


thanks
  • #116
  • Sep 18, 2015 5:40am Sep 18, 2015 5:40am
Quoting hanover
{quote} Sorry, there was a bug in my staking calculation. I've fixed it, and also extended the original XLS to a max of 12 trials. The end result is still the same: a total expectancy of 0, no matter how many trials. The updated XLS is too big to be attached to a FF post, hence I've zipped it (and attached the ZIP file to my original post #106). __________________________________________ [EDIT] {quote} Um, but that's exactly the myth that I was (looking to) debunk. Unless you can get the larger size positions to coincide with higher probability...
thank Hanover for your reply.
Just to clarify again and generate more discussion; Belkhayte Alembex has 2 important rules:

- We end a sequence after 2 winners in a row
- If we have losses in the past trade and then a win, the next trade will be the Cum + 1.

Just for the sake of discussion, if we consider that a trader decided to take Max 12 trades per day and he will stop trading for the day either after 2 winners in a row or after hitting the Max 12 trades.

Most of the days either he will gain +2 or +1 , depends on the location of the 2 winners in a row and rest is mixed.

Belkhayte claims that the probability of hitting 2 winners in a row in 12 trades is almost 100%. I Did not have the chance to verify this claim.

What do you think, will this change the expectancy?
  • #117
  • Sep 18, 2015 7:10am Sep 18, 2015 7:10am
Quoting ollaya
thank Hanover for your reply. ........ Belkhayte claims that the probability of hitting 2 winners in a row in 12 trades is almost 100%. I Did not have the chance to verify this claim. What do you think, will this change the expectancy?
It's not about what I think; the actual result can easily be proved by math. I have added columns AO thru BC at the right of the XLS. Unless I've misunderstood, the XLS follows the exact rules that you've specified, i.e. we stop after 2 consecutive wins, or 12 trials (i.e. trades), whichever happens first.

The result again highlights the fallacies in Mr Belkhayate's claims:

Firstly, the claim that "the probability of hitting 2 winners in a row in 12 trades is almost 100%". Column BC totals the number of winning permutations. This is 3719 out of all possible 4096 outcomes, or 90.8%

Secondly, the notion that the 'Alembex' betting system delivers positive expectancy. As we can see in cell BB4099, the total expectancy, once again, is zero.
In other words, no better or worse than flat betting. Sure, 3719 of the 4096 outcomes are winners (net P/L is +1 or +2), but the 377 losing outcomes include much bigger losses (the biggest is -47). This type of distribution (many small wins; a few huge losses) is typical of most martingale variants.

Again, this illustrates that position sizing can not improve expectancy. Like every other trading system, the expectancy (or profit factor) of Mr Belkhayate's system is determined solely from the quality of the entries and exits.

Here's a challenge to anybody who's interested:
Devise a position sizing system that, in itself, delivers positive expectancy; and then create an XLS like I have done, to prove it mathematically. I believe that it's impossible, but I'd love to be proved wrong, because it means that it would be possible to write an profitable EA that (assuming the +EV is sufficient to overcome the transaction costs) uses ANY entries and exits. In other words, we could all potentially become forex millionaires without the need to perform any kind of analysis.
Attached File(s)
File Type: zip Belkhayate expectancy debunked.zip   1.3 MB | 877 downloads
  • #118
  • Sep 18, 2015 7:54am Sep 18, 2015 7:54am
Quoting hanover
{quote} It's not about what I think; the actual result can easily be proved by math. I have added columns AO thru BC at the right of the XLS. Unless I've misunderstood, the XLS follows the exact rules that you've specified, i.e. we stop after 2 consecutive wins, or 12 trials (i.e. trades), whichever happens first. The result again highlights the fallacies in Mr Belkhayate's claims: Firstly, the claim that "the probability of hitting 2 winners in a row in 12 trades is almost 100%". Column BC totals the number of winning permutations. This is 3719...
thanks again Hanover .

The expectancy will be always 0 in the long run for 50/50 system (coin tossing) . the advantage that I see in Alembex is that , we are not adding to losers and the drawdown is much less than other (anti-martingale )systems.

And I believe that coupling this betting method with system that can generate 2 winners in a row in less trades will make the system more profitable .


As a I said I did not use this system and cannot prove anything. We would love to see and EA using this system and compare the result of the same EA with fixed betting ( % or fixed lot/$).
  • #119
  • Sep 18, 2015 10:14am Sep 18, 2015 10:14am
Attached Image
  • #120
  • Sep 18, 2015 10:43am Sep 18, 2015 10:43am
Quoting hanover
Here's a challenge to anybody who's interested: Devise a position sizing system that, in itself, delivers positive expectancy;
That's just being wickedly nasty ....the key words here are "in itself". It is like placing a $1M bet for any takers to devise a system to disprove the Second Law of Thermodynamics.
0 traders viewing now