Negative Progression roulette

[EliteNinja]

lmao
5 ppl voted yes.
So sad is the state of 2+2.

[popniklas]

Just to make it clear if anyone still thinks that Martingale would be a winning system under certain circumstances (such as a very big bankroll, or an infinite bankroll):

If, and only if, you assume a) an infinite bankroll b) possibilities to bet an infinite sum but NOT c) the possibility of an infinite losing streak, the Martingale is a safe system.

In the real world a) and b) are practically impossible conditions. Also, I don't think you ever could assume that an infinite losing streak is impossible, but someone better than me at maths might convince me otherwise, I don't know.

However, even if you had an infinite bankroll, could bet any amount you like and never hit an infinite losing streak, this would not be a winning system. Why? Because infinity + 1 = infinity. If you have an infinite bankroll, you can not add to your bankroll. It is already infinite.

So... back to the real world. If you want a probable small win, and can accept a small possibility of a huge loss, then you can play the Martingale for a while. But it will still be a -EV proposition, and if you do it often enough, you'll end up losing your bankroll often enough more often than your many small winnings would compensate for. Nuff said.

[TStoneMBD]

[ QUOTE ]
would i be correct in saying that if you bet $1 every time, and made a profit of $2048 after an extended period of time, that the law of averages would dictate that you would in turn lose $2048 consecutively by using the martingale system, cancelling out all profits and totally discreditting the martingale system?

[/ QUOTE ]

i think everybody ignored my post because 1), you dont know the answer and 2), you thought i actually believed the martingale system works.

i believe that it doesnt work simply because of all the talk on these boards in the past that it doesnt. i believe you guys. now i want to know why it doesnt work, and how to explain that it doesnt work to a person who is not familiar with gambling.

[popniklas]

why it doesn't work: (in this example, you do the martingale betting black at roulette)

1. sooner or later, red WILL turn up ten times in a row and you will lose all your money. that DOES actually happen and there is nothing strange about that.

2. the odds at casino games like roulette and craps are set up against you. on each bet you make, you will lose on average. if you bet on red a millions times, you will lose more than half a million of those times. as long as the odds remain the same, no betting system can change that.

if you do the martingale for just a short while, you are more likely to win than lose. however, if you do it over again a million times, your many small wins will not cover up for your unfrequent, but big, losses.

3. size of the bankroll does not matter, any finite bankroll could be lost, because red could theoretically come up any number of finite times in a rew.

4. even if you have an infinite bankroll, it still does not work (for reasons stated in my previous post). and by the way, if you have an infinite bankroll, why bother to try to make more money?

[popniklas]

...and if they still don't believe you, tell them that the casinos would be out of business by now if this system worked. and give them the url to www.wizardofodds.com

[The Goober]

[ QUOTE ]
[ QUOTE ]
would i be correct in saying that if you bet $1 every time, and made a profit of $2048 after an extended period of time, that the law of averages would dictate that you would in turn lose $2048 consecutively by using the martingale system, cancelling out all profits and totally discreditting the martingale system?

[/ QUOTE ]

i think everybody ignored my post because 1), you dont know the answer and 2), you thought i actually believed the martingale system works.

i believe that it doesnt work simply because of all the talk on these boards in the past that it doesnt. i believe you guys. now i want to know why it doesnt work, and how to explain that it doesnt work to a person who is not familiar with gambling.

[/ QUOTE ]

I think part of the reason you didn't get any replies is that you used the dubious term "law of averages". I know you don't mean it this way, but lots of fallacious thinkers use it to justify the thinking that previous random events have an impact on future indepedant events (i.e. "its hit black 10 times in a row, by the law of averages this next spin is more likely to be red!").

Here's my take on the easiest way to explain why martingale doesn't work:

Suppose we play a simple game - you bet 1 unit and I shuffle and turn over a card. If the card is 9-A, you win a bet, if it's 2-8 you lose your bet. Clearly your expectation is -1/13 bet per play (over 13 tries, you lose 8 times and win 7 times).

Now we modify the game. Now, you have to bet 13 units to play. If the card is anything but a deuce, you win 1 unit. If its a deuce, you lose your whole bet (all 13 units). Sounds good right? Now, for each time you play you have a 12/13 chance of winning 1 unit.

It's clear, though, that you haven't changed your expectation at all. Over 13 tries, you win 1 unit 12 times, and lose 13 units once. This is exactly what you are doing by playing a martingale system - you are trading off a greater chance of a small win for smaller chance or a large loss. The more times you can afford to double up (and the not hit the table limit), the more likely you are to make a small win, but the amount that you stand to lose goes up proportionally and you keep your expectation the same.

Also note that as your bankroll goes to infinity your chance of winning a small amount approaches 1, your chance of losing approaches 0, and the amount that you can lose approaches infinity. Your infinitesimal chance of losing is cancelled out by your infinite loss, and the whole mess is still going to have the exact same EV as you started with.

[cardcounter0]

Why the martingale doesn't work:

Using the martingale, double your bet on the next toss.
Flip a coin, heads you win, tails you lose. Here are the possible outcomes:

H - win $1
T - lose $1
net results - $0

flip it twice. Here are the possible outcomes:
HH - win $2
HT - win $0
TH - win $1
TT - lose $3
net results - $0 (notice you have two winning sessions, but your one losing session wipes out your wins)

flip it three times. Here are the possible outcomes:
HHH - win $3
HHT - win $1
HTH - win $2
THH - win $2
TTH - win $1
THT - win $0
HTT - lose $2
TTT - lose $7
net result - $0 (notice you have 5 winning sessions, but two losing sessions wipe out your wins)

Continue flipping and using a martingale as many times as you want. You will find that eventually you hit the long losing streak that wipes out all your previous wins.

Now a coin flip is a strict 50/50 proposition, with no house edge. That is why the ending result comes out to $0. If you are playing a game with an unfair payout (ie. the house edge) then your results will come out to losing your AVG BET X HOUSE EDGE.

So just take the total amount you wager, divide by the number of bets you make, this gives you your average bet, multiple by the HOUSE EDGE for the game - that is how much a martingale will lose you in the long run. In the short run, you will have many winning sessions, until you eventually hit the wall on the long losing streak, and wipe out your profits.

[bholdr]

it's a joke. if you take a look at a wheel, you'll see that his system eliminates ALL numbers. (thus, no HA!) [img]/images/graemlins/grin.gif[/img]

[47outs]

I voted 'good idea' cause I want you to learn the hard way.


outs

[youtalkfunny]

Question for all you guys who say it doesn't work:

When you were younger, did you go broke trying it?

(I did. Now I'm one of you, telling all the youngsters, "It doesn't work.")