Are my calculations of almost guaranteed long-term profit correct?

#1
I was thinking that maybe you can use Probability to get an accurate estimation of how much money and how many tries you need in order to get 99% success rate if you always keep betting twice your previous bet when you are losing in order to get everything you've currently lost back again (although this seems too good to be true).

Anyway, you can use Probability to calculate the success rate after a certain number of tries:

P = 1 - (x / [x+y])^t

P = probability for winning after t number of tries
x = unfavorable outcomes
y = favorable outcomes
t = number of tries

If we assume that the chances of winning are 40% each game then this gives the following result:

t = log(1-0,99) ÷ log(6 ÷ [6+4]) = 9.

So if we lose one dollar and then bet two dollars etc until we win all lost money back we get this:

2^0 + 2^1 ... + 2^8 = 511

...which means we would need $511 as a starting bet in order to have a 99% long-term theoretical success rate.

Is this correct?
 

Nynefingers

Well-Known Member
#2
I just skimmed the math, but it appears correct. The problem is that for the 1% of the time that you lose, you've lost $511. The 99% of the time that you succeed, you only make $1. So you are correct, but it certainly isn't a means for beating a -EV game. Do a search for "Martingale".
 

fwb

Well-Known Member
#3
Nynefingers said:
I just skimmed the math, but it appears correct. The problem is that for the 1% of the time that you lose, you've lost $511. The 99% of the time that you succeed, you only make $1. So you are correct, but it certainly isn't a means for beating a -EV game. Do a search for "Martingale".
In other words, if you were to clone 100 of yourself to play with your money, 99 of them would have $1 of profit, and one of them would have lost $511. You still lose in the end.
 

johnnyb

Well-Known Member
#5
Laurelindo said:
I was thinking that maybe you can use Probability to get an accurate estimation of how much money and how many tries you need in order to get 99% success rate if you always keep betting twice your previous bet when you are losing in order to get everything you've currently lost back again (although this seems too good to be true).

Anyway, you can use Probability to calculate the success rate after a certain number of tries:

P = 1 - (x / [x+y])^t

P = probability for winning after t number of tries
x = unfavorable outcomes
y = favorable outcomes
t = number of tries

If we assume that the chances of winning are 40% each game then this gives the following result:

t = log(1-0,99) ÷ log(6 ÷ [6+4]) = 9.

So if we lose one dollar and then bet two dollars etc until we win all lost money back we get this:

2^0 + 2^1 ... + 2^8 = 511

...which means we would need $511 as a starting bet in order to have a 99% long-term theoretical success rate.

Is this correct?
Your system is a progressive betting system. Progressive betting systems are LOSING systems, no matter which way you look at it. There are too many threads about this already, and there is math to prove it.

BTW, where did you get this math? 40% chance of winning? Where did that come from? Not correct at all.
 

21gunsalute

Well-Known Member
#6
Not sure about the math, but many assumptions are incorrect. For instance, $1 tables are difficult, if not impossible, to find. And if you are lucky enough to find one, the table max will probably be $100 or less. So you're bankroll will have to be many times higher than $511, but table limits are still going to wipe out any chance of a martingale succeeding. It's not that difficult to lose consecutive hands that reach into double digits, so even with a huge bankroll and unheard of table limits such a system is still doomed to failure. And a "successful" progression will only net 1 unit. Just not worth the risk.
 

Daggers

Well-Known Member
#7
i used to think that progressive systems work but now i don't. find a free blackjack game online for play money and try it out for a couple of weeks. you will find that if it was real money, you would be broke. you will reach the table max quickly by using that system. if you use it with counting and only double when the count is high, then you might have a better chance. But you would need a counting strategy with high PE. http://www.qfit.com/book/ModernBlackjackPage172.htm has a list of strategies.
 

aslan

Well-Known Member
#8
We have a padded cell prepared for all Forum members who persist in believing that a progression of any kind is a winning system. It is right next to my cell, which is reserved for blackjack players who insist on throwing away some of their blackjack winnings on high negative EV games and slots. A bad bet is a bad bet is a bad bet. No matter that sometimes you win, or even that you're ahead at this time. A bad bet is a bad bet .... I hope I get out on good behavior soon. I wouldn't want to end up with Big Bubba for a cell mate. :cry: :whip:
 

aslan

Well-Known Member
#10
b jay cobbson said:
...Can't wait to try it on my next visit to a casino.

Thanks,
Cobbson :)
Take a friend. That way you can play two different seats and double your chances of winning. You don't owe me anything for this advice, but you could slip me 10 or 15% of your winnings, sort of like a finder's or consultant's fee. :grin:
 

Coach R

Well-Known Member
#11
"Martingale" Look it up, will it work? sure till you lose, and you will. You can make it look like some kind of an earth shattering breakthrough with a lot of quotions and formulas, but it is still the Martingale system that will ALWAYS get you in the end. If it worked, Vegas wouldn't be able to pay the light bills. I think they make alot more than that from people who think they can bet their way out of a losing streak by betting more, and more. Learn to count,it's the ONLY time you can adjust bets to put yourself at an advantage
 

Coach R

Well-Known Member
#12
aslan said:
Take a friend. That way you can play two different seats and double your chances of winning. You don't owe me anything for this advice, but you could slip me 10 or 15% of your winnings, sort of like a finder's or consultant's fee. :grin:
Better take a friend alright, he loan you bus fare and money to get breakfast with.
 
Top