Poll

1 vote (7.69%)
12 votes (92.3%)

13 members have voted

vegasrvp
vegasrvp
  • Threads: 11
  • Posts: 53
Joined: Jun 15, 2010
July 4th, 2010 at 8:30:07 AM permalink
I have posted this under BlackJack and am still looking for the math to help me disprove a friend. I know in theory it should fail but in reality it isn't.

System.

150 units needed
Goal per session to win 10 units

Process:

Session 1: Flat bet 1 unit playing basic strategy until up or down 10 units.
Result: Up 10 walk away and play again elsewhere - Down 10 increase flat bet to 2 units and repeat above.

Session 2: Flat bet 2 unit playing basic strategy until up or down 20 units.
Result: Up 20 walk away and play again elsewhere - Down 20 increase flat bet to 4 units and repeat above.


Session 3: Flat bet 4 unit playing basic strategy until up or down 40 units.
Result: Up 40 walk away and play again elsewhere - Down 40 increase flat bet to 8 units and repeat above.


Session 4: Flat bet 8 unit playing basic strategy until up or down 80 units.
Result: Up 80 walk away and play again elsewhere - Down 80 bust and session is complete

The goal again is to get up 10 units before losing 150. This obviously means you need to win 15 times per loss to break even.

I have played 23 times and won all 23 times only getting to the 4th session or 8 units one time.

Where this seems to have a little merit is when you get down 10 or 20 units through the first 1 or 2 sessions you increase in wager amount gets you back faster then you were losing.

If you lose 20 of the first 30 hands you have run through the first session and move to the second. If you then win 20 of 30 you are now at 50 wins vs 50 losses but ahead 10 units and walking away. This could even happen faster if you get opportunities to double, split or hit a blackjack.

PLEASE HELP ME WITH PROVEN MATH WHY THIS WILL NOT WORK.
boymimbo
boymimbo
  • Threads: 17
  • Posts: 5994
Joined: Nov 12, 2009
July 4th, 2010 at 9:12:43 AM permalink
vegasrvp.

You're asking us to prove with math why this doesn't work. How about you prove to us by math that it will?

Basically, you are risking 150 units to win 10. So, you need to prove that you will win on average 15 times for every time that you bust.

You did one session of 23 times. Try doing about 10,000 simulations and see how you do. Really, you should just look at the the probability of coming out +10 or -10 units. Then look at the probability of losing 10 units 4 times in a row. If it is greater than 1/15, then your strategy makes no sense. To simplify this, take a look at the following scenario.

Let's say that Blackjack is a fair game, a coin toss on an unbiased coin.

The odds of you winning 10 units or losing 10 units then is 50%.
The odds of you winning 10 units or losing 30 units then is 25%
The odds of you winning 10 units or losing 70 units then is 12.5%
The odds of you winning 10 units or losing 150 units then is 6.25%

Odds of winning 10 units = 50% + 25% + 12.5 % + 6.25% = 93.75%
Odds of losing 150 units = 6.25%

Net win = 93.75 * 10 - 6.25 * 150 = 0.

However, Blackjack is not a fair game.

Therefore the odds of winning 10 units is <50% + <25% + <12.5% + <6.25% = <93.75%

Odds of losing 150 units > 6.25%.

Net win < 0.

Therefore, in any negative expectation game, the odds of losing all of your bankroll outweighs those of winning a win gain of ANY unit. Therefore, NO betting strategey will result in a positive expectation in a negative expectation game.
----- You want the truth! You can't handle the truth!
AZDuffman
AZDuffman 
  • Threads: 240
  • Posts: 13886
Joined: Nov 2, 2009
July 4th, 2010 at 9:48:57 AM permalink
Quote: vegasrvp



PLEASE HELP ME WITH PROVEN MATH WHY THIS WILL NOT WORK.



The math is simple as the statement, "No matter what your system, you cannot overcome a negative expectation game."

Take a coin flip game. A win gives you $.99 for every $1 bet, a loss loses the whole $1. (You get your $1 bet back on a win.)

In 2,000 trials you will theoretically win 1,000 bets for $990 and lose 1,000 bets. You have left over $990 in wins plus the $1,000 returned winning bets or $1,990. Bet on a feeling, change your call every bet, over the long hall it does not matter. You might break some probability and win 5, 10, or 20 in a row. But that means a loss of just as many in a row can just as easily happen.
All animals are equal, but some are more equal than others
7winner
7winner
  • Threads: 9
  • Posts: 198
Joined: May 31, 2010
July 4th, 2010 at 10:54:12 AM permalink
Quote: AZDuffman

Take a coin flip game. A win gives you $.99 for every $1 bet, a loss loses the whole $1. (You get your $1 bet back on a win.)


This gives a .50% house advantage.
Quote: AZDuffman

In 2,000 trials you will theoretically win 1,000 bets for $990 and lose 1,000 bets.


Only a 1.784% chance of happening. Or 1 in 56.
Quote: AZDuffman

You have left over $990 in wins plus the $1,000 returned winning bets or $1,990. Bet on a fedeling, change your call every bet, over the long hall it does not matter. You might break some probability and win 5, 10, or 20 in a row. But that means a loss of just as many in a row can just as easily happen.


SD of 22.36.
So 68.57% range is: (very slightly more than 1 SD at 68.26%)
-53.78 with 978 wins (35.18% that wins are between 978 and 1000)
+33.78 with 1022 wins (35.18% that wins are between 1000 and 1022)

Quote: AZDuffman

Bet on a fedeling, change your call every bet, over the long hall it does not matter.
You might break some probability and win 5, 10, or 20 in a row.
But that means a loss of just as many in a row can just as easily happen.


AZDuffman is exactly correct as to how equal it is to win or lose a streak. as applied to coin tossing.
The HA of .5% shows that when one wins you win less than what you lose with the same probability.
7 winner chicken dinner!
AZDuffman
AZDuffman 
  • Threads: 240
  • Posts: 13886
Joined: Nov 2, 2009
July 4th, 2010 at 12:23:39 PM permalink
Quote: 7winner

This gives a .50% house advantage.



Picked it because that is close to most BJ games.

Quote:

Only a 1.784% chance of happening. Or 1 in 56.



I didn't mean it will be exact, hence my use of theoretically. The point wasn't to be exact, it was to show you cannot overcome a negative expectation game no matter how you change the bets.

Quote:

AZDuffman is exactly correct as to how equal it is to win or lose a streak. as applied to coin tossing.
The HA of .5% shows that when one wins you win less than what you lose with the same probability.




As simple as that math is, so many people don't realize it at first. I think the problem is in BJ the math is fairly complex, whereas a coin flip is something anyone can understand. So when you make it that simple and say, "the math is the same" people get it more.
All animals are equal, but some are more equal than others
ahiromu
ahiromu
  • Threads: 112
  • Posts: 2107
Joined: Jan 15, 2010
July 4th, 2010 at 6:22:44 PM permalink
I do something like this but not nearly as regimented. What you're basically doing is after a given set of losses you're increasing your bet size to increase your variance in order to make up those losses. All it takes is one bad session to ruin your past 15 wins. Like the Martingale, if you had an infinite bankroll and there were no table minimums you would eventually win. I think that your answer is "no" to both, Mimbo did a good job at explaining the math.
Its - Possessive; It's - "It is" / "It has"; There - Location; Their - Possessive; They're - "They are"
rtpud
rtpud
  • Threads: 6
  • Posts: 26
Joined: Sep 2, 2010
September 7th, 2010 at 6:53:15 AM permalink
Isn't this a wading through a deck to attain a favorable count? obviously with martingale you aren't moving far enough if the count is even or against you as you lose consecutive hands, but in this style you are allowing the count to adjust and POTENTIALLY move into your favor during hte larger bets.

Follow this:
You sit down to a fresh shuffle - if you win 10 straight, its likely the deck was stacked in a way that the count is now very against you...doesn't matter you get up and leave. if you lose 10 straight, it is likely the count is for you or even, you keep playing...I would argue if you lose 30 hands in the way described during a single shuffle, the count is likely to be in your favor, giving a better EV on your martingale-esque increase toward the back end of the deck.

what does a simulation of 6 or 8 deck blackjack look like for back counting? e.g. how often do you get a -15 or +15 during some part of the sessions deal?
Ramond
Ramond
  • Threads: 10
  • Posts: 34
Joined: Nov 8, 2010
March 7th, 2011 at 9:15:55 AM permalink
I tried some kind of system like this a few times, because I like to go home with a win.
I Start betting €40,- flat. After 10 wins I go home with €400,-. After 20 losses I start betting €100,- flat. And again I go home when I have €400,-. When I lose another 10 units I'm €1.800,- behind. After that I do just 1 more bet, €2.000 on Banker. When I win, I go home, when I lose, I also go home.

There were a lot of times I had a great evening and go home with full pockets, even 2 times with €1.800 behind the Banker bet saved me.
But in the end I lost... and €3.800 is a lot. In the end everybody will lose with some kind of system like this, unless you're a good card counter (but thats not possible in Holland because of the shuffle machines).

In the end the house advantage will take your winnings, unless youre very very lucky.

But I understand why you use a system like this, because most of the times you will win (altough it is small).

But don't see this like an investment. I compare my losses with sports betting, which can be won!
SOOPOO
SOOPOO
  • Threads: 121
  • Posts: 10942
Joined: Aug 8, 2010
March 7th, 2011 at 11:09:49 AM permalink
I recommend you read the Wiz's 'betting system' section on wizardofodds.com If after reading it you still don't understand then I can assure you the subsequent hundreds of posts with questions and answers won't help, either. The easiest way to know that it does not work is that you KNOW if it did work you would not be the first human in history to figure it out.
thecesspit
thecesspit
  • Threads: 53
  • Posts: 5936
Joined: Apr 19, 2010
March 7th, 2011 at 11:27:21 AM permalink
Quote: Ramond

I tried some kind of system like this a few times, because I like to go home with a win.
I Start betting €40,- flat. After 10 wins I go home with €400,-. After 20 losses I start betting €100,- flat. And again I go home when I have €400,-. When I lose another 10 units I'm €1.800,- behind. After that I do just 1 more bet, €2.000 on Banker. When I win, I go home, when I lose, I also go home.

There were a lot of times I had a great evening and go home with full pockets, even 2 times with €1.800 behind the Banker bet saved me.
But in the end I lost... and €3.800 is a lot. In the end everybody will lose with some kind of system like this, unless you're a good card counter (but thats not possible in Holland because of the shuffle machines).

In the end the house advantage will take your winnings, unless youre very very lucky.

But I understand why you use a system like this, because most of the times you will win (altough it is small).

But don't see this like an investment. I compare my losses with sports betting, which can be won!



This reminds me a little bit of a certain Video Poker system that was highly discussed elsewhere on the board :)
"Then you can admire the real gambler, who has neither eaten, slept, thought nor lived, he has so smarted under the scourge of his martingale, so suffered on the rack of his desire for a coup at trente-et-quarante" - Honore de Balzac, 1829
  • Jump to: