Does the martingale system really work? | Page 15 | Metals Mine

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?

  • #281
  • Edited 5:22pm Nov 26, 2013 4:41pm | Edited 5:22pm
Quoting Master_kiwa
Good day to all. I've attached two indicators. .....
Very interesting; many thanks for sharing.

Some questions, if I may:

1. How many wagers would you place, on average, during the course of a week?
2. How big is your starting wager (before the martingale kicks in), relative to the size of your capital base (i.e. account)?
3. How many times does the martingale reach each level (1,2,3, ... 14) per 1,000 trades?
4. Across how many wagers, and years worth of market data, have you tested your system?

Quoting Master_kiwa
After the red candle. You can simply wager that after 5 mins the price will not close above the high of the candle which closed at the pivot (binary options wager).
5. What odds/payoff does the bookmaker/br0ker offer, on average, for this type of wager?

Quoting Master_kiwa
The reality is if a person is trading well with martingale, he or she should be withdrawing the profit as that is the best money management technique of them all.
6. How often do you withdraw money, how much do you withdraw (relative to your account size), and what is the reason for the withdrawals (given that it's 100% certain that you'll never encounter a fatal losing sequence)?
  • #282
  • Nov 26, 2013 5:06pm Nov 26, 2013 5:06pm
you posted a nice example, thanks.

Quoting Master_kiwa
Now as I mentioned before, it is unfair to say martingale doesn't work, without looking at the system which is actually being used. Martingale is simply a form of wagering (entering your lot sizes), but your entry or exit is based on your system (another ideology all together) .
  • #283
  • Nov 26, 2013 6:21pm Nov 26, 2013 6:21pm
Quoting focusonmoney
{quote} Nice post Master_kiwa.. Funny thing people will still bash you and tell you it won't work lol..
It doesn't. If he's cracked the code then why are his account explorers at -99% return since year start? While I give him credit for presenting a good post, it's really just muddying the issue by throwing a separate system in with it. The profitability of a system here is of great importance; a system with a positive expectancy would naturally increase yields for a martingale run and one with a negative expectancy would get you bankrupt faster. So what we really want to look at is martingale by itself with a 50/50 coin flip trade generator over many runs. And if you had a system with positive expectancy, why would you risk the possible outcome of bankruptcy using something that could risk your entire balance eventually? Greed is the only thing I can think of.
  • #284
  • Nov 26, 2013 7:01pm Nov 26, 2013 7:01pm
Quoting charvo
you posted a nice example, thanks. {quote}
You are more then welcome. I will continue to post different examples. I would love to continue working with the example I gave as it will allow us to identify all the counters which the market can give us. Everything has a counter where is martingale helps you. I use both MT4 and binary options using the Martingale form of wagering. The account which I have posted with this account is meerly a test account. Only wagering with small amount of funds. Currently I have over 25k in buying power. I simply can not post with that account as it has been labled a COMMERICAL account here on ff.com I have been awaiting the assistance of mods from this site, but to no avail. Should they upgrade my account you will see all of my trades. Account name MECHIE
  • #285
  • Nov 26, 2013 7:24pm Nov 26, 2013 7:24pm
I'm working on something to absolutely kill any discussion over this redundant, done to death martingale topic again. Expect a thread soon.
  • #286
  • Nov 26, 2013 7:38pm Nov 26, 2013 7:38pm
Quoting Kanzler
I'm working on something to absolutely kill any discussion over this redundant, done to death martingale topic again. Expect a thread soon.
Just like religion, this topic has been killed a long time ago. Some people still cling to the dead horse.

Asking people to see the stupidity in martingale is like asking a left-handed person to use their right hand, or a fish to fly. Good luck with that endeavor.
Congratulations ScalaFX for coming out of the closet!
  • #287
  • Nov 26, 2013 8:47pm Nov 26, 2013 8:47pm
Quoting nubcake
{quote} Just like religion, this topic has been killed a long time ago. Some people still cling to the dead horse. Asking people to see the stupidity in martingale is like asking a left-handed person to use their right hand, or a fish to fly. Good luck with that endeavor.
I'm creating a primitive simulator in C++. The goal is for people to use it and to discover the results themselves. Even without spread/fees eating into the balance, it should convince people in a concrete way why martingale doesn't work.
  • #288
  • Nov 26, 2013 9:06pm Nov 26, 2013 9:06pm
Quoting Kanzler
{quote} I'm creating a primitive simulator in C++. The goal is for people to use it and to discover the results themselves. Even without spread/fees eating into the balance, it should convince people in a concrete way why martingale doesn't work.
Why bother?

It all comes down to having positive expectancy and an edge. If someone has an edge and positive expectancy then adding martingale is only increasing risk of ruin. If someone does not have an edge then they won't have positive expectancy, and adding martingale is only increasing risk of ruin.

To put it simply, for the simpletons, if your shit is a loser then martingale will bust you out at a rate that is different to just trading a set amount each time. That rate might be faster, or slower. Martingale simply muddies the waters of what is a failing strategy.

It is tied-into stupidity such as trading without stops and adding to losers. It's all just different angle of the same shitty failing idea, just turned around and upside down so that at first glance it looks like a brand-new grail to stupid people. It's all just the same shit from the same bucket, and that bucket sits on the losing side of the fence.

Being on the winning side of the fence means protecting capital, not needlessly risking it. Protecting capital doesn't automatically make you a winner (you still need positive expectancy), but needlessly risking capital with or without a positive expectancy DOES automatically make you a loser.

It is all just obfuscating the risk of ruin in various forms. You take the same shitty system or strategy, add in a dash or martingale, or no stops, or both, and you still have the same shitty system. It might look different, but to some of us we can smell it and it smells identical. All it takes is taking a moment to breathe in and it is obvious.

Positive expectancy? Yes or no? Martingale on top of either is pure stupidity.

It can all come back to a coin-flip. Is it weighted towards losing, or winning? Martingale on top just sucks-in the risk of ruin. There are spreadsheets out there that people can play with and see it for themselves already. But, people choose to try and make the situation more confusing for themselves and add-in averaging down, and up, and sideways, and stops here but not there... and blah blah blah... so that they can't even tell what their base metrics are and can hide the patent truth of their failing methods under layers of obscurity.
Congratulations ScalaFX for coming out of the closet!
  • #289
  • Nov 26, 2013 9:13pm Nov 26, 2013 9:13pm
no no kanzler, let's say me, i don't want to look at 50/50 coin flip, and i know you said it right, it'll die faster

i want to see a system, good (> 50%) or very good (70%), that can be applied with martingale so that bigger/faster profit may be accumulated.

in fact, my own manual statistics is about 60% on RR1:1, but 1. i cannot trade during the day since i have to work; 2. even i manage to trade, the return doesn't justify the time/energy spent, since i don't have a large capital. this is the fundamental reason i look into martingale. If I have a system with 70% edge, I will not look at martingale.

i have known "martingale fails" since 2008, just like hanover and nubcake showed. I completely understand and agree the proving procedure why it fails. however, i've gained a different mindset recently. I start to look for any setup that do not appear to fail for say, 10 consecutive times, I think that's not a non-existent thing in the market.

I think there's a difference (if not flaw) from the usual comparison of coin-flip or "independent random incident" examples. Coin-flips are separately independent from each other; while a trading sequence in a market is not totally independent from each other trade, is it???

a virtual example, let's say everytime we see a daily inside bar of gbpusd, then we go to H1 timeframe, after london open, buy up; if losing, X2 sell down; if losing, X2, buy again, you get the point......, right? will that easily lead to death sequence or not? I don't know, but I've done some similar tests, and find some promising setups.

the other flaw of famous coin-flip showed that everytime the winning sequence end up winning merely 1 unit, but risk is unlimited. well, here, the winning sequence doesn't need to end up winning merely 1 unit. the winning size may be increased with the raised position size. Still the london open range example above, for same TP pips, the 4th level winning may be 5 times of 1st level winning, if X2 is used. (take a paper and calculate yourself, if interested). in this way, the original loser (if SL at 1st level) turns into a 5X bigger winner and you take 5X risk (again, you can calculate). what a huge difference compared to coin-flip example!

I actually believe that someone who reads this can derive an awesome system like me, lol.

the above is actually not essential. the essential part is we recognize it and choose it.

there's a chinese saying "to get a baby tiger(precious), you need to get into tiger cave(dangerous)", it describes many people's life dilemma. in this tragic, boring, stupid human life, many ppl would risk/sacrifice massively to gain a shocking reward. many times, we need the 1st bucket of gold to carry our dream or desire. if only we know the potential risk, we could just give a try. all thru human history, we've seen so many feats achieved by various heroes taking gambling-like actions. Do we advise them not to since it is too risky? sometimes I feel it is in our gene deciding if you take the chance, lol......

anyway, my stupid babbling, maybe it looks disturbing to Nub again, lol, i guess that i can accept the risk as i know it beforehand.

Quoting Kanzler
{quote} So what we really want to look at is martingale by itself with a 50/50 coin flip trade generator over many runs. And if you had a system with positive expectancy, why would you risk the possible outcome of bankruptcy using something that could risk your entire balance eventually? Greed is the only thing I can think of.
  • #290
  • Edited 9:42pm Nov 26, 2013 9:25pm | Edited 9:42pm
Quoting charvo
maybe it looks disturbing to Nub again, lol
Yup. That was a safe-bet, though.

You've ventured into the land of "but a market isn't a coin-flip". Ok, that's correct. It's not. BUT, for most people it is because they have no edge and are just trading what can be considered to be randomly.

If you now want to discuss martingale under the premise of having an actual edge, then it begs the question "why would you risk your capital under such a silly and dangerous betting scheme if you already have a profit-generating method?".

If you have an edge and want to martingale in some form then you start really small to protect your capital once your bets begin to gain size, or you start at some larger level and are just a few steps closer to risk of ruin.

That's the crux of martingale. It just brings the risk of ruin closer, and you either mitigate that by making your initial betting pointless, or you don't.

Instead, if you have an edge you should be looking to work-out your own risk of ruin based on your metrics that are bringing you that positive expectancy. Then you can tailor your trading size to fit. Then grind it out.

In the end the 1% risk guideline is there for a reason. It's small enough to keep your account safe, but not so small that the risk and reward is absurdly pointless. People can simply go on autopilot and not have to concern themselves with calculating anything but what 1% of their account is and calculate their size based on their stop from that. It's a simple betting scheme that keeps the noise within a positive (or negative) expectancy system to a comfortable level.

After all is said and done, the numbers can be simplified to a weighted coin. It is either weighted one way or another, and to some degree based on your edge or lack thereof.

With a fair coin, if I paid you $1.10 for every heads, and you paid me $1 for every tails... would you just grind-out this game and profit? Or, would you risk some silly betting scheme that only serves to obfuscate your metrics on the game and also damages your capital that you have to spend on the game?

With an unfair coin that was further biased in your favor you would go away, work the numbers, and figure-out the numbers attached to the risk of ruin, and then you could increase how much you bet accordingly while still keeping your capital safe. But, I'm sure you still wouldn't be stupid-enough to engineer silly betting schemes.

You would... grind it out.

edit: and furthermore, in regards to you maybe skipping some trades and waiting for a more "guaranteed win" then (in a system with an edge) all you are doing is missing-out on all of the times you would have won anyway in the meantime, all for the sake of getting more strings of winners. But, to that I also add, you are probably still going to get the exact same metrics as if you had taken all trades, it's just taken you longer to get the same sample size because you have been skipping potential trades.

Unless you can actually quantify an individual trade as having a better probability of being a winner than another, then you are just introducing noise into your metrics, but in the end they will be the same. IF you can pick a higher-probability trade, NOT because you have had losers but because of some actual information pertaining to the trade itself, THEN you can increase your metrics in a positive manner.

If you are just skipping trades for the sake of waiting on a string of losers then you are still trading within the same probabilities as if you had just traded all trades. It's only when you can individually rank a trade on its own merits that you can accordingly affect your metrics.

I.e. If you can sharpen your edge, then great. Otherwise, skipping trades is just randomly skipping trades.
Congratulations ScalaFX for coming out of the closet!
  • #291
  • Nov 26, 2013 9:49pm Nov 26, 2013 9:49pm
Quoting charvo
I completely understand and agree the proving procedure why it fails. however, i've gained a different mindset recently. I start to look for any setup that do not appear to fail for say, 10 consecutive times, I think that's not a non-existent thing in the market. I think there's a difference (if not flaw) from the usual comparison of coin-flip or "independent random incident" examples. Coin-flips are separately independent from each other; while a trading sequence in a market is not totally independent from each other trade, is it??? {quote}
Hello charvo. What happens with noobcake and the other non-believers is they trade with the "hEAD and tail mentality" which is why martingale appears to be a waste of time. Guys like yourself who have the ability to identify different setups or structures which will result in a 1:1 R:R ratio will see martingale as the smart form of wagering, only because your formation ends up hitting your tp say 70% of the times. At a 70% win rate it will be very unlikely that you lose 4 consecutive times. So why wouldn't you use martingale? The +10'ers have yet to even mention a system or post a system which HASN'T worked with martingale. When clearly "heads/tails" works with martingale. It works because you can't lose 100% of the times. Instead of noobcake loading the indicator i POSTED and disproving what I've mentioned he insist on using his bias opinion to disprove martingale. Seems like he has something personal against martingale.
  • #292
  • Nov 26, 2013 9:52pm Nov 26, 2013 9:52pm
Quoting Master_kiwa
{quote} Hello charvo. What happens with noobcake and the other non-believers is they trade with the "hEAD and tail mentality" which is why martingale appears to be a waste of time. Guys like yourself who have the ability to identify different setups or structures which will result in a 1:1 R:R ratio will see martingale as the smart form of wagering, only because your formation ends up hitting your tp say 70% of the times. At a 70% win rate it will be very unlikely that you lose 4 consecutive times. So why wouldn't you use martingale? The +10'ers...
Still whining about V's, and still defying simple facts of life. You must be a hoot to be around. I'm sure you have lots of opinions on all manner of topics you haven't the slightest grip on.
Congratulations ScalaFX for coming out of the closet!
  • #293
  • Nov 26, 2013 9:53pm Nov 26, 2013 9:53pm
Back to my system. Look what has happened the last 6 times that the price has touched the pivot, and closed with a candle indicating reversal. Looks like a 100% win rate. I have about 17 different formations which I wager on. The very same formation which I posted several hours ago, has just appeared with a very good win %'age. In that case no need to use martingale as the system has yet to fail. Had it failed, with every failure my chances increase of actually closing 5 mins later in the correct direction.
Attached Image (click to enlarge)
Click to Enlarge Name: current.PNG Size: 65 KB
  • #294
  • Nov 26, 2013 9:58pm Nov 26, 2013 9:58pm
Thanks, Nub! This is a good lesson for me. I'll look it over.

Quoting nubcake
{quote} You would... grind it out.
edit: If you are just skipping trades for the sake of waiting on a string of losers then you are still trading within the same probabilities as if you had just traded all trades. It's only when you can individually rank a trade on its own merits that you can accordingly affect your metrics. I.e. If you can sharpen your edge, then great. Otherwise, skipping trades is just randomly skipping trades.
  • #295
  • Nov 26, 2013 10:10pm Nov 26, 2013 10:10pm
Here is where the martingale comes into to play. I have a .02 sell on AJ open. If AJ closes under 92.56 on the 1m I will take the loss, open a sell for .04 and enter a binary options sell for 25 usd once the price retraces to the middle band and closes with a red candle. I have more then enough margin to lose 20 times in a row and not blow more then 20% of my account.
Attached Image (click to enlarge)
Click to Enlarge Name: current1.PNG Size: 35 KB
  • #296
  • Nov 26, 2013 10:13pm Nov 26, 2013 10:13pm
What I want is to lose the very small orders so that as the larger orders come in I will have a much higher probability of earning my 1 to 1 ratio. Weeks may go buy before I ever place an order higher then 200usd in my binary account or 1.00 lot size in my mt4. Yet, if the small lot sizes add up all together.
  • #297
  • Nov 26, 2013 10:21pm Nov 26, 2013 10:21pm
You can see that the price has yet to close under my stop loss price. The .02 buy is still open, and I have yet to place a binary trade. Based on the current formation, AJ is appearing to be heading into an accumulation formation. How can one tell the price is heading into an accumulation. Once the reversal has occured (Color change red to blue dot or visa versa, and any particular price after the first color change is not been closed across more then 4 times (two to the buy side, and two to the sell) without a 15+ pip breakout.
Attached Image (click to enlarge)
Click to Enlarge Name: yettoclose.PNG Size: 43 KB
  • #298
  • Nov 26, 2013 10:30pm Nov 26, 2013 10:30pm
Quoting hanover
{quote} Very interesting; many thanks for sharing. Some questions, if I may: 1. How many wagers would you place, on average, during the course of a week? 2. How big is your starting wager (before the martingale kicks in), relative to the size of your capital base (i.e. account)? 3. How many times does the martingale reach each level (1,2,3, ... 14) per 1,000 trades? 4. Across how many wagers, and years worth of market data, have you tested your system? {quote} 5. What odds/payoff does the bookmaker/br0ker offer, on average, for this type of wager?...
Hello hangover. I will be uploading my accounts so you can analyze the very questions you've asked. As for the reasons of withdrawing I would have to say it is to reap my rewards of excellent trading. It also a form of good money management to withdraw profit especially since I never risk more then 20% of my account anyway. Many people risk 2% of their account and use the other 98% to back it up. Which to me is asinine as it would suggest that 98% of your account will be used mostly to recover the 2% wager which failed.
Also my binary options broker offers anywhere from 81-89% profit on the 5 min wager. Within the binary account is where I do most of the heavy martingale trading.
  • #299
  • Nov 26, 2013 10:42pm Nov 26, 2013 10:42pm
My simulator is uploaded in the other thread...take a look. The data is pretty clear; while the martingale strategy often makes some ridiculous returns, it always goes bankrupt, often within a couple thousand trades if not much faster at 1 percent risk and a 60% win rate. With lower values it's much much sooner. Standard risk management is much slower but I've yet to see a simulation where you lose your money with a real edge.
  • #300
  • Nov 26, 2013 10:52pm Nov 26, 2013 10:52pm
Quoting Kanzler
My simulator is uploaded in the other thread...take a look. The data is pretty clear; while the martingale strategy often makes some ridiculous returns, it always goes bankrupt, often within a couple thousand trades if not much faster at 1 percent risk and a 60% win rate. With lower values it's much much sooner. Standard risk management is much slower but I've yet to see a simulation where you lose your money with a real edge.
Once again what good is your simulator if your entries aren't calculated. You're simply entering positions on heads or tails. Most of us don't take wagers on that method. At least I don't. Which means your c++ is a "waste of time". As what is simulated almost never occurs as frequently in the real world.
1 trader viewing now