Does martingale really works? - page 91.

dooglus

legendary

Activity: 2940

Merit: 1333

Quote from: Wendigo on June 10, 2014, 07:21:16 PM

Can you use martingale on sportsbetting events?

You can use martingale on anything. If you've lost X so far, bet to win a little more than X on your next bet. It's also called "chasing your losses" and is generally considered a bad idea.

Here's the result of Deb using martingale on Bitzino's video poker. She started betting at 10 microbits (1000 satoshis). Each time she lost, she doubled up. Each time she hit jacks, which just returns your stake, she stayed the same, and each time she won, she reset to 10 microbets.

After doubling 10, 20, 40, 80, 160, 320, 640, 1280, 2560, 5120, 10240, 20480, 40960 she hit a royal flush. She couldn't afford to double again, and this was the first time she had ever bet as much as 40960 microbits (0.04096000 BTC) on a single hand. Incredibly lucky to hit a royal flush on your highest ever, and last ever bet. But it happened!

Then she sold the 32 BTC profit for about $32 each before the April 2013 bubble. Doh!

boumalo

legendary

Activity: 1918

Merit: 1018

Quote from: PacoMartin on June 12, 2014, 09:41:37 AM

Quote from: boumalo on June 12, 2014, 09:00:26 AM

It's about 40% but most people actually assume that a million trials of an event that has a probability of 1/1,000,000 of happening has 90% chance of seeing that event

Thank you for commenting. When guessing at his probability of winning, dooglus used P=1/50.5%^28 which I said is about 40%. A better estimate for 50% is P*ln(2)/49.5%.

Yes I read the article, it was interesting; did you write any other articles or can you tell us if you are a gambler or casino investor?

PacoMartin

member

Activity: 70

Merit: 10

Quote from: boumalo on June 12, 2014, 09:00:26 AM

It's about 40% but most people actually assume that a million trials of an event that has a probability of 1/1,000,000 of happening has 90% chance of seeing that event

Thank you for commenting. When guessing at his probability of winning, dooglus used P=1/50.5%^28 which I said is about 40%. A better estimate for 50% is P*ln(2)/49.5%.

boumalo

legendary

Activity: 1918

Merit: 1018

Quote from: PacoMartin on June 11, 2014, 06:16:10 PM

Quote from: boumalo on June 11, 2014, 05:41:19 PM

Check here as well : "Frank A. Martin, put together this one-page paper about the bet." :

Possibly, because long runs of events (good or bad) is alien to our primeval caveman ego, psychologists have shown that most people have no idea how to guess their probability of occurring. People almost always estimate wrongly and almost always think streaks are more rare than they really occur. The paper presents a simple recursion formula to tell you how to do this,

Another common mistake is to start with an event that has a one in a million chance of happening. People assume that if they do a million trials, that they will have a 50% chance of seeing that event. That is absolutely wrong. Huh

It's about 40% but most people actually assume that a million trials of an event that has a probability of 1/1,000,000 of happening has 90% chance of seeing that event

PacoMartin

member

Activity: 70

Merit: 10

Quote from: boumalo on June 11, 2014, 05:41:19 PM

Check here as well : "Frank A. Martin, put together this one-page paper about the bet." :

Possibly, because long runs of events (good or bad) is alien to our primeval caveman ego, psychologists have shown that most people have no idea how to guess their probability of occurring. People almost always estimate wrongly and almost always think streaks are more rare than they really occur. The paper presents a simple recursion formula to tell you how to do this,

Another common mistake is to start with an event that has a one in a million chance of happening. People assume that if they do a million trials, that they will have a 50% chance of seeing that event. That is absolutely wrong. Huh

boumalo

legendary

Activity: 1918

Merit: 1018

Quote from: dwma on June 11, 2014, 02:58:29 PM

Martingaling decreases the amount you can win and decreases the chances of you losing. You'll never win big though but may lose big.

For some people who get a big come up from the psychological edge of winning, then martingales are where it is. If winning a tiny amount does nothing for you, then they're kind of stupid. All martingales do is shift around variance. Whether you consider that "working" or not I can't say, but it sure doesn't make you a winning bettor.

It is a very interesting way to put it; it's all about changing variance that changes your experience

Quote from: PacoMartin on June 11, 2014, 02:40:46 PM

Quote from: dooglus on May 31, 2014, 05:17:41 PM

Quote from: f3tus on May 31, 2014, 04:38:14 PM

Holy fu*k! I'd be without a house and liver right now! I'm glad I only bet 0.002, lol!

We can go down in history for making a 30k bet on whether or not a losing streak of 28 happened, which is kinda cool. Not really, though, lol.

Also to the guy that lost 32 times... You'd have more chance getting struck by lightening 10 times in a row.

Interesting that there are 5 winning streaks of 27 or longer but only 2 losing streaks that long. At 49.5% you would expect more long losing streaks then winning ones.

I hope you enjoy the paper on the dooglus's blog. Please comment
http://just-dice.blogspot.ca/2014/06/incredible-winning-and-losing-streaks.html

Odds of dooglus winning as a function of the qualifying "even money" bets (out of 1.2 billion total)
38.6% : 200 million
42.2% : 225
45.6% : 250
48.8% : 275
51.9% : 300
54.7% : 325
57.4% : 350
59.9% : 375
62.3% : 400
64.5% : 425
66.6% : 450
68.6% : 475
70.4% : 500
72.2% : 525
73.8% : 550
75.4% : 575
76.8% : 600
78.2% : 625
79.5% : 650
80.7% : 675
81.8% : 700
82.9% : 725
83.9% : 750
84.9% : 775
85.8% : 800
86.6% : 825
87.4% : 850
88.1% : 875
88.8% : 900
89.5% : 925
90.1% : 950
90.7% : 975
91.3% : 1 billion

"I ended up winning the bet, but it turns out it was nowhere near as safe a bet as I originally thought!"

Check here as well : "Frank A. Martin, put together this one-page paper about the bet." :
https://just-dice.com/Just%20Dice%20Streak%20Bet.pdf

dwma

sr. member

Activity: 405

Merit: 250

Martingaling decreases the amount you can win and decreases the chances of you losing. You'll never win big though but may lose big.

For some people who get a big come up from the psychological edge of winning, then martingales are where it is. If winning a tiny amount does nothing for you, then they're kind of stupid. All martingales do is shift around variance. Whether you consider that "working" or not I can't say, but it sure doesn't make you a winning bettor.

Ashbite

full member

Activity: 168

Merit: 100

Well its all in your head tbh.
if you get lucky and you win you obv think it works.
but this shit is vice versa.

PacoMartin

member

Activity: 70

Merit: 10

Quote from: dooglus on May 31, 2014, 05:17:41 PM

Quote from: f3tus on May 31, 2014, 04:38:14 PM

Holy fu*k! I'd be without a house and liver right now! I'm glad I only bet 0.002, lol!

We can go down in history for making a 30k bet on whether or not a losing streak of 28 happened, which is kinda cool. Not really, though, lol.

Also to the guy that lost 32 times... You'd have more chance getting struck by lightening 10 times in a row.

Interesting that there are 5 winning streaks of 27 or longer but only 2 losing streaks that long. At 49.5% you would expect more long losing streaks then winning ones.

I hope you enjoy the paper on the dooglus's blog. Please comment
http://just-dice.blogspot.ca/2014/06/incredible-winning-and-losing-streaks.html

Odds of dooglus winning as a function of the qualifying "even money" bets (out of 1.2 billion total)
38.6% : 200 million
42.2% : 225
45.6% : 250
48.8% : 275
51.9% : 300
54.7% : 325
57.4% : 350
59.9% : 375
62.3% : 400
64.5% : 425
66.6% : 450
68.6% : 475
70.4% : 500
72.2% : 525
73.8% : 550
75.4% : 575
76.8% : 600
78.2% : 625
79.5% : 650
80.7% : 675
81.8% : 700
82.9% : 725
83.9% : 750
84.9% : 775
85.8% : 800
86.6% : 825
87.4% : 850
88.1% : 875
88.8% : 900
89.5% : 925
90.1% : 950
90.7% : 975
91.3% : 1 billion

Swordsoffreedom

legendary

Activity: 2884

Merit: 1115

Leading Crypto Sports Betting & Casino Platform

Quote from: Wendigo on June 10, 2014, 07:21:16 PM

Can you use martingale on sportsbetting events?

I guess you could but not on the same game, although in that case sportbooks would have odds so you could just try it with different results each time till you find the best scenario.
Martingaling the World Cup would be interesting all the way to the finals bracket unless you lost each time lol.

Wendigo

legendary

Activity: 2604

Merit: 1036

Can you use martingale on sportsbetting events?

kuverty

sr. member

Activity: 770

Merit: 250

Quote from: Zombie23 on June 10, 2014, 09:37:56 AM

I'm not talking about the probability of the next one. I'm saying with a extremely high number of bets, which is why i just chose 1000 that if you graph it out (depending on the house edge) 49.5% will be wins, and 50.5% will be loses. I understand every single event is independent of the previous ones, but the numbers at high values should always hover around the 49.5 and 50.5 mark.

So you agree that after 1000 tails, heads is no more probable? Then you must understand that this does not help you with betting at all. It is of no use predicting the outcome of any single bet, they are all the same. I mean, if you for some reason try to get a long streak of losses, you will probably have had a similar amount of wins and winning streaks before that. The statistics do go like that. But that will not affect anything you do in the future.

It is the same as if you never had the previous results. It is analogous to flipping a coin. The history just does not matter.

Or are you saying that the history somehow matters? Because trying to get losing streaks before betting big suggests that, and only that. Am I correct? If not, what is the reason for trying to get a losing streak if it did not (supposedly) matter for future bets?

Zombie23

newbie

Activity: 5

Merit: 0

Quote from: kuverty on June 09, 2014, 09:23:14 PM

Quote from: Zombie23 on June 09, 2014, 08:53:12 PM

Quote from: kuverty on June 09, 2014, 08:42:22 PM

Quote from: Zombie23 on June 09, 2014, 08:28:47 PM

In my personal experience no. The best way I have found is to gamble in alt-coins. Preferably one that is only worth less than 100 Satoshis. Then find a site where you can bet .00000001 of that coin. Play until you have lost like 5 in a row then start martingale at 1 or 2. Sure the profit will be slow, but this gives you more protection. The only real way to make money on this would be to actually use a program to run it for you, but this is just my idea.

It does not offer any protection. If the game works like it should, and it will or it would be bankrupt by now, the events are distinct. You lose 5 times in a row before betting big; so what? The next events do not depend on those, you still have the same chance to lose.

Let's imagine you throw a coin 10 times and get 10 heads; on the 11th, is it more probable that you get heads or that you will get tails?

It is just there to give you more protection. With the .00000001 bet you aren't really losing any value, and if the game truly is fair the distribution should bring it to the point where you would win. It is all about getting more layers of protection. Because lets be honest, you aren't going to lose 1000 games in a row, the odds are so small I'm confident it wouldn't happen because the statistics will balance it out, so you are right, you still have the same odds (an independent event) where if you bet >30 times than a normal distribution curve should apply. (This is all coming from a highschool stat class, but this is pretty basic stat)

While I am waiting you to answer - I'd like you to, I have to point out that you seem to be thinking of "statistics" as some kind of a mystical power that will "balance things out". Statistics have no will. Statistics will behave in a certain manner though - you will see that 1000 losses is not very probable. But it is a 50-50 chance, that streak of 1000 losses, be it "0000000...." is just as probable as "10110011100011..." or any other sequence of wins/losses. Of course those cases where you both lose AND win some of the bets are more numerous, there are almost 2^1000 that kind of sequences, and only one sequence where you lose all bets. None of this means, that after a streak of 999 losses, the final, 1000th one will be more probably a win. Both are of equal probability. I would like to hear you simply answer yes or no to the question I posed, disregarding additional arguments, if I may ask you to do that.

I'm not talking about the probability of the next one. I'm saying with a extremely high number of bets, which is why i just chose 1000 that if you graph it out (depending on the house edge) 49.5% will be wins, and 50.5% will be loses. I understand every single event is independent of the previous ones, but the numbers at high values should always hover around the 49.5 and 50.5 mark.

Umer

member

Activity: 98

Merit: 10

Quote from: leex1528 on June 10, 2014, 07:37:07 AM

I like how everyone tries methods, but fails at math so bad. No matter how much you increase your odds(Which Martingale does) your odds are not favorable, if they were, Casinos wouldn't be in existence. All the money would go to the players and we would all be rich.

Well i try winning dice like this ...

First i bet about the mount of fees 0.0001 2 or 3 bets or maybe 1 ...

If the both two or three bets below 50 i immediately bet 0.1 or the required amount..

I win but some time i lose..

The dice is processing by a pattern first up 2nd donw like..

leex1528

hero member

Activity: 784

Merit: 1000

I like how everyone tries methods, but fails at math so bad. No matter how much you increase your odds(Which Martingale does) your odds are not favorable, if they were, Casinos wouldn't be in existence. All the money would go to the players and we would all be rich.

Umer

member

Activity: 98

Merit: 10

Quote from: boumalo on June 10, 2014, 05:55:46 AM

Quote from: Umer on June 10, 2014, 05:52:08 AM

Quote from: hensi on May 15, 2014, 01:08:09 PM

Everyone who has been a part of any gambling site must have tried martingale. so i just wanted to know your experience that have u ever gained profit from martingale? or its name should be changed to martinfail?

Well in my opinion martingale does not work with me dont know ..
Maybe the personal matters with the martingale :d lol

And some of my friends says tht it really works..

They must be filthy rich then Grin

LOl the have only 1 btc or more in thier account Cheesy

boumalo

legendary

Activity: 1918

Merit: 1018

Quote from: Umer on June 10, 2014, 05:52:08 AM

Quote from: hensi on May 15, 2014, 01:08:09 PM

Everyone who has been a part of any gambling site must have tried martingale. so i just wanted to know your experience that have u ever gained profit from martingale? or its name should be changed to martinfail?

Well in my opinion martingale does not work with me dont know ..
Maybe the personal matters with the martingale :d lol

And some of my friends says tht it really works..

They must be filthy rich then Grin

Umer

member

Activity: 98

Merit: 10

Quote from: hensi on May 15, 2014, 01:08:09 PM

Everyone who has been a part of any gambling site must have tried martingale. so i just wanted to know your experience that have u ever gained profit from martingale? or its name should be changed to martinfail?

Well in my opinion martingale does not work with me dont know ..
Maybe the personal matters with the martingale :d lol

And some of my friends says tht it really works..

boumalo

legendary

Activity: 1918

Merit: 1018

Quote from: kuverty on June 10, 2014, 04:41:41 AM

Quote from: boumalo on June 10, 2014, 02:26:11 AM

Quote from: dooglus on June 10, 2014, 12:23:40 AM

Quote from: kuverty on June 09, 2014, 09:23:14 PM

While I am waiting you to answer - I'd like you to, [...]

Everything you say is true, but I have little hope that you will ever convince him of that.

I see people using the "bet small until you see N the same in a row, then bet high that the streak will end" technique a lot on Just-Dice. Usually they wait for the streak and then start Martingale betting.

What's funny is that it often works very well.

The reason for this is that Martingale *does* work very well until you hit a streak that's too long, and if you play enough times you will hit that streak.

But when you wait for a streak before you even start, it has the side effect of really reducing the number of real bets you can make per hour, which really increases the time until you reach a streak that busts you.

So the described strategy "works" in a sense, in that Martingale is a horrible strategy, and waiting for a certain condition to occur before starting each Martingale progression means you play less.

People see this, and think that the magic of waiting for a streak in the dust bets is what it working, whereas smoking a cigarette or reciting the Lord's Prayer between each bet would "work" just as well, to the extent that both would slow you down.

Yes; it is amazing to see how many players think they can change the outcome by the way they bet or when they bet

If you are going to use 100BTC to play a martingale 1BTC 2BTC ect. and you decide you will stop after 50BTC have been won you have better odds betting 100BTC @ 1.5 once

I was thinking about that yesterday looking at a player bet strange amounts in a strange order then say he played badly on the chat after he lost; you said it doesn't make sense to allow autobeting but it does because probably more than 50% of the players don't think in terms of probability but in other terms

Yeah, the first time I saw these autobetting options on a dice site, I thought the operators were dumb or something... but apparently it makes sense.

You have to adapt to customers, I used the auto-betting function on PD and I understand why you can have fun with it to try some martingale fantasy

kuverty

sr. member

Activity: 770

Merit: 250

Quote from: boumalo on June 10, 2014, 02:26:11 AM

Quote from: dooglus on June 10, 2014, 12:23:40 AM

Quote from: kuverty on June 09, 2014, 09:23:14 PM

While I am waiting you to answer - I'd like you to, [...]

Everything you say is true, but I have little hope that you will ever convince him of that.

I see people using the "bet small until you see N the same in a row, then bet high that the streak will end" technique a lot on Just-Dice. Usually they wait for the streak and then start Martingale betting.

What's funny is that it often works very well.

The reason for this is that Martingale *does* work very well until you hit a streak that's too long, and if you play enough times you will hit that streak.

But when you wait for a streak before you even start, it has the side effect of really reducing the number of real bets you can make per hour, which really increases the time until you reach a streak that busts you.

So the described strategy "works" in a sense, in that Martingale is a horrible strategy, and waiting for a certain condition to occur before starting each Martingale progression means you play less.

People see this, and think that the magic of waiting for a streak in the dust bets is what it working, whereas smoking a cigarette or reciting the Lord's Prayer between each bet would "work" just as well, to the extent that both would slow you down.

Yes; it is amazing to see how many players think they can change the outcome by the way they bet or when they bet

If you are going to use 100BTC to play a martingale 1BTC 2BTC ect. and you decide you will stop after 50BTC have been won you have better odds betting 100BTC @ 1.5 once

I was thinking about that yesterday looking at a player bet strange amounts in a strange order then say he played badly on the chat after he lost; you said it doesn't make sense to allow autobeting but it does because probably more than 50% of the players don't think in terms of probability but in other terms

Yeah, the first time I saw these autobetting options on a dice site, I thought the operators were dumb or something... but apparently it makes sense.

Topic: Does martingale really works? - page 91. (Read 123304 times)