Does the martingale system really work? | Page 17 | Crypto Craft

Options

Go to
Permalinks
Print Thread

Similar Threads

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

Does this simple system by Richard Donchian really work? 134 replies

Attachments: Does the martingale system really work?
Exit Attachments

Does the martingale system really work?

  • #321
  • Nov 27, 2013 5:13am Nov 27, 2013 5:13am
Quoting hanover
{quote} At the 14th level the total amount at risk is 1+2+4+8+16+32+64+128+256+512+1024+2048+4096+8192 = 16,383 times the original bet. {quote} Hence if 20% of the account needs to survive 14 levels, then the starting bet would need to be 20%/16383 = 0.00122%. That's why I asked you what your starting bet size was. I can't see how you can make a decent return risking only 0.001% of your account, per wager. What have I missed?
That is just it Hanover. I never lose more then 4 times in a row using the method which I have posted. Using the number "14" is a fact all in all, because the setup never experiences more then 4 losses all together. As far as the wagering is concerned my first orders are always small, but while using martingale it increases at times more then double the last wager. I am not sure if you were paying attention to the orders which I placed on AJ. Yet my first order in mt4 was .02 lots (4.00 usd) where as the next order in my binary options account would of been 25 usd. so a difference of more then 5 times. My reality is that I trade using martingale, and I have over all profit. I have even gone as far as posting a small part of my system which is used with the martingale. I don't trade in this bubble called "imaginary land". I trade using all the information given to me, then make inferences based on that I have.
  • #322
  • Nov 27, 2013 5:19am Nov 27, 2013 5:19am
Quoting PhoenixInc
Hello sir. I support your theory on martingale (Pro). As I currently use it and have had amazing results with it thus far. I don't scalp as you do, yet my win % is well over 92%. I'm currently the number PAMM leader on hotforex. You can simply click my name here on ff.com to see my stats. Master_kiwa you have my undivided attention. Please keep up the good work.
Your stats appear to be very good. At a quick glance I see you use martingale, and you have profited greatly from it. Excellent work.
  • #323
  • Nov 27, 2013 6:52am Nov 27, 2013 6:52am
With an unlimited bankroll the Martingale system will work... however infinite funds do not exist.

I like to think of it like this.... The martingale strategy gives you a high chance to win a small amount and a low chance to lose a lot. However if you use it repeatedly (and you have to to make money) it will be inevitable that you will eventually make a loss equal to or greater than your profits to date.... or you will simply not have enough money to continue with the system.
  • #324
  • Edited 9:09am Nov 27, 2013 8:59am | Edited 9:09am
Quoting Master_kiwa
{quote} That is just it Hanover. I never lose more then 4 times in a row using the method which I have posted.
Thanks for your reply. OK, so the number 14 is essentially irrelevant; you've never had more than 4 consecutive losses.
How did you establish this, i.e. what was the nature of your testing? Did you run some kind of simulated test over historical data, or is this the result of live trading/wagering? Over how many trades/wagers has your testing been performed, and over how many years worth or market data? How long have you been using this system?
  • #325
  • Nov 27, 2013 9:51am Nov 27, 2013 9:51am
Quoting hanover
{quote} Thanks for your reply. OK, so the number 14 is essentially irrelevant; you've never had more than 4 consecutive losses. How did you establish this, i.e. what was the nature of your testing? Did you run some kind of simulated test over historical data, or is this the result of live trading/wagering? Over how many trades/wagers has your testing been performed, and over how many years worth or market data? How long have you been using this system?
I don´t run simulations on trading. I simple base my caluclations on actual trading experience. I have traded forex since 2006. In which I have added many different formations to my system. One can simply back test and keep track using a spreadsheet of different formations. I have a total of 14 formations, yet I only trade on 4 of them. I counter those 4 formations using binary options as a backup for a stop loss which will be hit. My strongest formation doesn´t have more then 2 consecutive losses. Although the odds are in my favor, I don´t over leverage incase some crazy market moving event was to happen. Running c++ programs is a good way to convience new traders that something works/ doesn´t work. Yet they NEVER translate into your live account. Only because a program or simulator can not factor in THE HUMAN ELEMENT. Which would rock any system´s world.
  • #326
  • Nov 27, 2013 9:53am Nov 27, 2013 9:53am
Quoting tradingboy
With an unlimited bankroll the Martingale system will work... however infinite funds do not exist. I like to think of it like this.... The martingale strategy gives you a high chance to win a small amount and a low chance to lose a lot. However if you use it repeatedly (and you have to to make money) it will be inevitable that you will eventually make a loss equal to or greater than your profits to date.... or you will simply not have enough money to continue with the system.
Very good way of putting it. Yet, as you mentioned FUNDS is the most important thing. Money management will curb emotions to the extent that you wont get cleaned out if your stick to your rules.
  • #327
  • Nov 27, 2013 10:10am Nov 27, 2013 10:10am
Quoting Master_kiwa
{quote}I am not sure if you were paying attention to the orders which I placed on AJ. Yet my first order in mt4 was .02 lots (4.00 usd) where as the next order in my binary options account would of been 25 usd. so a difference of more then 5 times. My reality is that I trade using martingale, and I have over all profit. I have even gone as far as posting a small part of my system which is used with the martingale. I don't trade in this bubble called "imaginary land". I trade using all the information given to me, then make inferences based on that I have.
You're such a bold faced liar. You've also hit your limit of 4 losers in a row several times in the mere 44 trades you've executed, and only didn't go further because your broker has the saving grace of a margin call to avoid your total bankrupcy.
Attached Image (click to enlarge)
Click to Enlarge Name: Untitled.png Size: 40 KB
  • #328
  • Nov 27, 2013 11:39am Nov 27, 2013 11:39am
Quoting Kanzler
{quote} You're such a bold faced liar. You've also hit your limit of 4 losers in a row several times in the mere 44 trades you've executed, and only didn't go further because your broker has the saving grace of a margin call to avoid your total bankrupcy. {image}
Do you just skim through what I write? I have several formations which I trade. Not all formations experience the same amount of losses. You would literally have to open each trade and compare the formations to what it was I was trading. Due remember that I use binary options to back up my mt4 trades. For example this week I have only traded two formations which has resulted in 8 consecutive wins. : ) FORMATIONS is the key to my system, not flipping the coin as you would suggest how I trade. You have hard on for me using martingale which is unstandable, because you don´t understand how to use it correctly. Just as your silly c++ program which was poorly created and in no way would translate to real life preformence.
  • #329
  • Nov 27, 2013 11:51am Nov 27, 2013 11:51am
Quoting Master_kiwa
{quote} Do you just skim through what I write? I have several formations which I trade. Not all formations experience the same amount of losses. You would literally have to open each trade and compare the formations to what it was I was trading. Due remember that I use binary options to back up my mt4 trades. For example this week I have only traded two formations which has resulted in 8 consecutive wins. : ) FORMATIONS is the key to my system, not flipping the coin as you would suggest how I trade. You have hard on for me using martingale which...
You are one of the most vile users I've seen here in quite some time. "Formations" are irrelevant, because there's no guarantee it's going to work and hand you a win. If there was, why would you bother trading any other "formation" you have? Each trade has a risk that it's not going to go the direction you need it to, and as a result you can't make a profit from it to recover your losses from before. If you really have a strategy that guarantees a win at a certain time, well, may I congratulate you on your victory - you're certain to rule the world in a few years as the richest man in existence. I'm still waiting on an explanation as to why you lost 98% of your account before your lucky streak today.
  • #330
  • Edited 10:28pm Nov 27, 2013 1:04pm | Edited 10:28pm
Quoting Master_kiwa
{quote} I don´t run simulations on trading. I simple base my caluclations on actual trading experience......
Suppose a trader tells you that, with a given system, he's never had more than 4 consecutive losses. It sounds impressive. But then you find out he's made only 20 trades in his entire career. Then it looks like nothing special.

But if the worst he's ever encountered was 4 consecutive losses in 2,000 trades over 10 years of changing markets. Then the statistics become more meaningful.

That was the reason for my question: over how many trades have you tested your system.

With uncapped martingale, it requires only one unexpectedly large losing sequence to cause ruin; hence I'd need to see a humungous history, to give me any kind of confidence. And even that's probably not enough. I'm wary of using analogies, but justifying martingale is kind of like saying "I've been driving for 20 years without ever having an accident, therefore I'll never be killed in a car crash". Driving a car likewise involves a human element; and just as the highways can throw an unexpected event at you at any time, so can the markets.

Uncapped martingale leaves your account open to unlimited loss. Mathematically, you'd need to have an infinitely long trade history to be able to say with 100% certainty that "it will never happen". I guess you could safely ignore the possibility of ruin if you could say with something like 99.9% certainty that it won't happen, but you'd need a very large trade history to deliver that level of statistical confidence.
  • #331
  • Nov 27, 2013 10:27pm Nov 27, 2013 10:27pm
Quoting hanover
{quote} Suppose a trader tells you that, with a given system, he's never had more than 4 consecutive losses. It sounds impressive. But then you find out he's made only 20 trades in his entire career. Then it looks like nothing special. But if the worst he's ever encountered was 4 consecutive losses in 2,000 trades over 10 years of changing markets. Then the statistics become more meaningful. That was the reason for my question: over how many trades have you tested your system. With martingale, it requires only one unexpectedly large losing sequence...
The problem is their is so much negativity with martingale. As if it is the sole reason as to why one will lose. The major portion of winning or losing would be the system itself. Martingale with a crappy system or a long losing streak will eat your account (very true). Yet, martingale with a system which is even 50% accurate will cause exponential growth (also true). A flat pancake has both sides. So saying how bad martingale is, will only make people wonder why it is actually good.

It is very wise to use martingale if you have a very decent system is all I contend. Yes you may end up running into a very long losing streak, etc. Yet with a strong system it isn't very likely.

As for wanting to see over 2000 trades. I would love to see anyone who has 2000 trades to reveal they have wasted their time, only to achieve non-impressive numbers. I am more impressed of someone who can turn 100-600% profit in a week, or even sustain triple digit group for several months. As you can analysis a person's trades to see what they have done. Which is why I focus on "Pip draw down"
The lower the pip draw down then the more efficient your trade is. Efficient because you are exposed to market conditions for less time, compared to being in the red and trying to make adjustments. If I have placed 10 trades, and turned over 200% profit with a avg pip draw down on 6 pips, then it means my entry/system is dead on! That to me would be a much better conversation to build on rather then this martingale crap which is by far as bias as religion.
  • #332
  • Nov 28, 2013 12:51am Nov 28, 2013 12:51am
Quoting Master_kiwa
{quote} That is just it Hanover. I never lose more then 4 times in a row using the method which I have posted. Using the number "14" is a fact all in all, because the setup never experiences more then 4 losses all together. As far as the wagering is concerned my first orders are always small, but while using martingale it increases at times more then double the last wager. I am not sure if you were paying attention to the orders which I placed on AJ. Yet my first order in mt4 was .02 lots (4.00 usd) where as the next order in my binary options account...

Kiwa, I am always open to new ideas, but I have to ask how long have you been trading? I think your method may be market specific. Martingale is an exceedingly harsh way to trade.....have you tried other methods?
If you are good at something, never do it for free--Joker
  • #333
  • Edited 2:49am Nov 28, 2013 2:31am | Edited 2:49am
Quoting Master_kiwa
{quote} The problem is their is so much negativity with martingale.........
At the end of the day MM must be a personal decision, and I expect that each trader's perspective is influenced by his own financial situation.

For example, in my own case, virtually all of my income comes from investments (both active and passive) in other markets. I don’t trade forex myself, but my friend (mim2005) and I are slowly building and forward testing an arsenal of EAs which will, if they prove profitable, become a small part of the portfolio (see this thread for more info). Since it’s part of my life savings that’s involved, risk management is of paramount importance to me, much more so than attempting to maximize income.

However, if for example, I was simply throwing a trivial amount (e.g. $500) into an account, then I’d be more willing experiment with MM ideas, possibly even some martingale-like variants, and then withdraw the original stake after I’d doubled it. And if I happened to encounter a fatal losing sequence before that time, well, it was money that I could afford to lose. But I see that as a completely different ballgame: it would be having some fun gambling, like a night out in Vegas, rather than serious trading.

But I guess everyone sees it differently. Some people have a greater appetite for risk than I do. Whereas I’m happy to be conservative; I see investment as a marathon rather than a sprint. Each to their own.
  • #334
  • Nov 28, 2013 2:45am Nov 28, 2013 2:45am
master kiwa's account is around 50% loss now, his pivot martingale trading Technics not working as well
  • #335
  • Nov 28, 2013 2:47am Nov 28, 2013 2:47am
Quoting FX_James
master kiwa's account is around 50% loss now, his pivot martingale trading Technics not working as well
-96% all time.
  • #336
  • Nov 28, 2013 11:35am Nov 28, 2013 11:35am
Quoting Master_kiwa
Yet, martingale with a system which is even 50% accurate will cause exponential growth (also true).
No it isn't true, it is the opposite. Martingale is a formula for slow straight line growth with exponential risk. I don't know where this idea that martingale increases returns comes from.
  • #337
  • Edited 12:10pm Nov 28, 2013 11:55am | Edited 12:10pm
Ok i won't quote all of you guys since a good time passed since i posted here but all of you who responded to this tread since my last reply i tell this post to you guys.

Lets simplify math.

Suppose we code the wins and losses: 1 for win and 0 for losses.

Now if the market is totally random which it isn't (according to my latest research,although it has some random elements), we have a 50% chance of winning on each trade.

And for simplicity we only go to martingale level 7 (so that we dont lose much). We start with 0.01 lot so at the end we end up with 0.64 lot which is still good for a 200 $ account.It doesnt blow it up atleast.

And we roll the dices....trade... we have a 0.5^8 chance of losing and 50% chance of winning, so we only lose at 8 times a row losing streak which has a probability of 0.390625% and a win chance of 50%

That being said, but we dont know the winrate of our system, or atleast not at the beginning (and i say that 51% is required) we may need additional tools to find it out.

And i also emphasized the 1:1 RR, to standardize the system since the martingale won work, so if we have a 0.01:1 RR right at the winner trade which recompensates our 6 losing streak we still end up in negative and didnt recuperated out losses.So we must standardize the system and 1:1 RR is good for that.Now the only problem is the winrate, so with 50% winrate ,since the expected value will converge to infinity, we only take limited trades in a lifetime, we either end up with more winners or with more losers.

But we need an edge to tip that balance so atleast 51% is good.Also with that winrate you would be a winner even without martingale, but martingale is a quick get rich method for its best.

If we know that our trades have atleast 51% winrate always and 1:1 RR we can draw a nice little probability distribution curve


Take the following example:
Here we got 66,66% win rate (which is bigger than 50%), and the RR being the same, we lose always the same amount as we win, we can ignore that and focus on the probability.And we only "really" lose if we get to losing streak of 8, anything below 8 is acceptable.But we win if we only win once, so at this 66% win rate, we have:
66% of win and 33% of lose, but on the martingales viewpoint we can improve this chance to:

  1. 99.98593591381759% of winning 1x times the sum (1 WINNING TRADE IS ENOUGH)
  2. 0.01406408618241 % of losing ^7x times the sum (AT 8 LOSING TRADES IN A ROW)

(By using martingale level 7 aka LOT ^ 7)


This is its probability distribution based on a 1000 randomly generated sample (with 66% winrate in total)

Attached Image (click to enlarge)
Click to Enlarge Name: asdasd.png Size: 64 KB


You can clearly see that the 66% winrate has a smaller spread out on the extremities, calculated in sigma, hence it has a smaller chance of losing rows, and the majority of the outcomes tend to accumulate on 1 tail which clearly means that it has an edge.So its logical to assume, and the math also suggests, that the higher winrate, the less chance of losing rows, which is btw logical just put 66% on a 8 row lose situation = (1-0.66)^8 (there is only .33 % of loss so we calculate based on that)

0.01406408618241% which is way less than 0.390625% on the 50-50 side.So yes guys go learn some basic statistics first then complain.

And so that being said that higher winrate does mean smaller chance of losing strikes = martingale works on this situation.Now i dont say that it will work forever, nor that it is a sustainable thing, but if the requirements are met, then it can be a good gambling system with a nice edge.
I linked the spreadsheed for those who would like to analyze my experiment.
Attached File(s)
File Type: xlsx MARTIN.xlsx   36 KB | 753 downloads
"There's a sucker born every minute" - P.T. Barnum
  • #338
  • Edited Nov 29, 2013 10:37am Nov 28, 2013 1:42pm | Edited Nov 29, 2013 10:37am
Quoting Proximus
Lets simplify math. .............
Proof that martingale (or any staking system) does not alter overall expectancy

Attached is a XLS that I’ve posted many times before. It allows you to test different staking strategies over 6 trials, by changing the values in the yellow cells, and then the XLS sums all 2^6=64 combos to calculate expectancy. You can see in the rightmost column of the table how the wins and losses are distributed.

Screenshot #1: flat betting with a 50% win rate and 1:1 RR.
Result: Total overall expectancy = 0; 50% win rate; average bet size = 1.

Screenshot #2: martingale betting with a 50% win rate and 1:1 RR.
Result: Total overall expectancy = 0; 68% win rate; average bet size = 1.

Screenshot #3: martingale betting with a 66% win rate and 1:1 RR.
Result: Total overall expectancy = 212.80; 83% win rate; ; average bet size = 3.4635 (average bet size is > 1 because bets are 1,2,4,8, etc)

Screenshot #4: back to flat betting with a 66% win rate and 1:1 RR. Set average bet size for both winning and losing trades to 3.4635
Result: Total overall expectancy = 212.80; 50% win rate; average bet size = 3.4635.


Conclusions:

Martingale does not change expectancy (in fact you can try any MM/staking system, and you’ll reach the same conclusion). All martingale does is increase the percentage of positive outcomes, but the losses that do occur are commensurately bigger.

The positive expectancy in screenshot #3 comes from the 66% win rate at 1:1 RR, NOT the martingale. Screenshot #4 shows how the same result can be achieved with flat betting, by using the same average bet size.

You can extend the XLS to 10 trials, 20 trials, 100 trials, any number of trials, the result remains exactly the same.

QED
____________________________

[EDIT] You can use the XLS to try different staking systems. I also wrote a MQL4 script to test the Guetting progression, and some XLSs to test different variants of Oscar's Grind. The conclusion: no MM system can alter expectancy. Or if you prefer a neat mathematical proof, PipMeUp has provided one here.
Attached Image(s) (click to enlarge)
Click to Enlarge Name: Image 173.png Size: 14 KB
Click to Enlarge Name: Image 174.png Size: 15 KB
Click to Enlarge Name: Image 175.png Size: 15 KB
Click to Enlarge Name: Image 176.png Size: 15 KB
Attached File(s)
File Type: xls Martingale comparison.xls   125 KB | 605 downloads
  • #339
  • Nov 28, 2013 1:57pm Nov 28, 2013 1:57pm
Quoting hanover
{quote} Proof that martingale (or any staking system) does not alter overall expectancy Attached is a XLS that I’ve posted many times before. It allows you to test different staking strategies over 6 trials, by changing the values in the yellow cells, and then the XLS sums all 2^6=64 combos to calculate expectancy. You can see in the rightmost column of the table how the wins and losses are distributed. Screenshot #1: flat betting with a 50% win rate and 1:1 RR. Result: Total overall expectancy = 0; 50% win rate; average bet size = 1. Screenshot...

Martingale isn't used to change expectancy. It is only used to recover your string of losses much quicker. Which is why talking about win %, or anything regarding % is pointless. As it is mearly an avg. It NEVER translates into real life situations. I could have a 86 % win rate, but the amount of trades actually placed will determine if my account margin calls or not. If I place 1000 trades a 86% win rate means I closed 860 trades for a win. Yet what if my first 140 trades were all losers. That would mean martingale won't work. So what should really be focusing on is the system. Let us try to find systems where a person has placed over 2000 trades, and let us take apart the system. That is the only way we will be able to tell if martingale will work. You have to compare it against something. Which of course would be the system itself.
  • #340
  • Nov 28, 2013 2:00pm Nov 28, 2013 2:00pm
Quoting Master_kiwa
It is very wise to use martingale if you have a very decent system is all I contend. Yes you may end up running into a very long losing streak, etc. Yet with a strong system it isn't very likely
If p is your winrate, NOT getting a single string of k losers (you say k=4) out of n trades (say n=200, approx one trade per session in a year) is given by
Attached Image

Certainly not the most sexy formula I agree...
No greed. No fear. Just maths.
0 traders viewing now