Poll

1 vote (11.11%)
2 votes (22.22%)
2 votes (22.22%)
No votes (0%)
4 votes (44.44%)

9 members have voted

chrisr
chrisr
  • Threads: 13
  • Posts: 141
Joined: Dec 9, 2013
May 12th, 2014 at 3:30:29 PM permalink
To prove you are the master of baccarat you must play a series of hands and win at least 1 more hand than you lose betting on player. You get to decide how many resolving bets to place beforehand. However, you can only pick an even number. How many bets should you place to maximize the chance of winning this coveted title.

The probabilities of banker, player and tie are 0.458597, 0.446247 and 0.095156 respectively.

Note: ties do not count as resolved bet
AxiomOfChoice
AxiomOfChoice
  • Threads: 32
  • Posts: 5761
Joined: Sep 12, 2012
May 12th, 2014 at 3:37:09 PM permalink
I assume you mean, at least one more, not exactly one more?

Which mean, at least two more, since it must be an even number?
chrisr
chrisr
  • Threads: 13
  • Posts: 141
Joined: Dec 9, 2013
May 12th, 2014 at 3:38:41 PM permalink
Quote: AxiomOfChoice

I assume you mean, at least one more, not exactly one more?

Which mean, at least two more, since it must be an even number?



yes, that's right
AxiomOfChoice
AxiomOfChoice
  • Threads: 32
  • Posts: 5761
Joined: Sep 12, 2012
May 12th, 2014 at 3:40:11 PM permalink
Well, making it a multiple choice question makes it a lot easier. Just plug all the possible answers into a binomial calculator.
kubikulann
kubikulann
  • Threads: 27
  • Posts: 905
Joined: Jun 28, 2011
May 12th, 2014 at 4:26:00 PM permalink
I expect the minimum # of hands, since it is a negative exp game
But on the other hand it might be one of those un-intuitive questions like the "tennis pro" one?
Reperiet qui quaesiverit
AxiomOfChoice
AxiomOfChoice
  • Threads: 32
  • Posts: 5761
Joined: Sep 12, 2012
May 12th, 2014 at 4:32:13 PM permalink
Quote: kubikulann

I expect the minimum # of hands, since it is a negative exp game
But on the other hand it might be one of those un-intuitive questions like the "tennis pro" one?



You are not trying to maximize expectation.

The tennis pro one was pretty intuitive, I thought.

Over here you are just looking for n such that you maximize that chances of having at least n+1 successes in 2n chances with the probability of success = 0.458597 / (0.458597 + 0.446247).

I am not sure how to maximize this, but I do know how to put 4 different values of n into a binomial calculator and see which one gives the biggest number :)
Tomspur
Tomspur
  • Threads: 28
  • Posts: 2019
Joined: Jul 12, 2013
May 12th, 2014 at 4:49:36 PM permalink
There should be achoice for....."I do believe in math but I suck at it" :)
“There is something about the outside of a horse that is good for the inside of a man.” - Winston Churchill
chrisr
chrisr
  • Threads: 13
  • Posts: 141
Joined: Dec 9, 2013
June 13th, 2014 at 11:41:37 AM permalink
no one got the right answer! There are two opposing effects:

1) the bias in favor of banker
2) the probability of a tie as n increases

had we been allowed to play an odd number of games we could ignore (2) and play the minimum number of hands.

74


Quote: AxiomOfChoice

I am not sure how to maximize this, but I do know how to put 4 different values of n into a binomial calculator and see which one gives the biggest number :)



You could plug in all the values in excel and find the maximum.. but finding the general solution is difficult. So the next problem is to prove the general solution for any p<0.5 is:

1/(1-2p) (rounded to the nearest even int)
  • Jump to: