[algogeeks] Re: probability of winning with two cards

2012-01-23 Thread Lucifer
@Don, I think i misunderstood the question again.. :).. One of major things that i went wrong with was that for me the deck consisted of 13*3=39 cards..( basically an assumption made based on the way i understood the question) Thanks for the explanation.. On Jan 23, 9:17 pm, Don wrote: > Close.

[algogeeks] Re: probability of winning with two cards

2012-01-23 Thread Don
Close. You actually have to be sure that all 6 cards dealt to the players are unique. For instance, if I get 3 points, if you don't require that all the cards dealth in the game are unique, you would conclude that there is a very small, but positive probability that I will win. In reality, 3 points

[algogeeks] Re: probability of winning with two cards

2012-01-23 Thread Lucifer
@Don and Sundi.. As Don pointed out, all we are looking for is: sum of a1 > sum of a2 sum of a1 > sum of a3 Assumption: 1) The 2 cards picked for a particular player are unique. 2) Cards are numbered : 1,..., 12, 13. Hence, the following code should give the answer for the a1's probability to

[algogeeks] Re: probability of winning with two cards

2012-01-23 Thread Lucifer
@Don.. Yup, it seems I misread it ... :) .. Thanks On Jan 23, 9:17 am, Don wrote: > I think that you are misreading the problem. A1 wins if his sum is > larger than A2's sum and larger than A3's sum. A1's sum doesn't have > to be larger than A2+A3. > Don > > On Jan 22, 5:18 pm, Lucifer wrote: >

[algogeeks] Re: probability of winning with two cards

2012-01-22 Thread Don
I think that you are misreading the problem. A1 wins if his sum is larger than A2's sum and larger than A3's sum. A1's sum doesn't have to be larger than A2+A3. Don On Jan 22, 5:18 pm, Lucifer wrote: > @sundi.. > > Lets put is this way.. > > Probability of (a1 wins + a1 draws + a1 losses) = 1, >

[algogeeks] Re: probability of winning with two cards

2012-01-22 Thread Lucifer
@sundi.. Lets put is this way.. Probability of (a1 wins + a1 draws + a1 losses) = 1, Now, sample count a1 wins = 46298 ( using the above given code) Hence, the probability (win) = 46298/474552 = .097561 [ @ Don - as i mentioned in my previous post that i had initially missed a f

[algogeeks] Re: probability of winning with two cards

2012-01-22 Thread Sundi
Hi Lucifer, Have you checked the sum of probability of (a winning + b winning + c winning + draw)==1 ? On Jan 22, 2:38 pm, Lucifer wrote: > @above > > editing mistake.. (btw the working code covers it) > /* > int j =*1*; > for(int i = 0; i < 12 ; i+=2) > { >     A[i] = A[i+1] = A[22-i]

[algogeeks] Re: probability of winning with two cards

2012-01-22 Thread Lucifer
@above editing mistake.. (btw the working code covers it) /* int j =*1*; for(int i = 0; i < 12 ; i+=2) { A[i] = A[i+1] = A[22-i] = A[21-i] = j; ++j; } */ On Jan 22, 6:53 pm, Lucifer wrote: > @Don.. > > Well i will explain the approach that i took to arrive at the > probability.. > Well ye

[algogeeks] Re: probability of winning with two cards

2012-01-22 Thread Lucifer
@Don.. Well i will explain the approach that i took to arrive at the probability.. Well yes u are correct in saying that it doesn't make a lot of sense but then the no. of wins by a1 keeping in mind that a1 > a2 + a3 is much less than a1 <= a2 + a3.. Or may be I have gone wrong in calculating the

[algogeeks] Re: probability of winning with two cards

2012-01-19 Thread Don
You are saying that a1 wins roughly 1 in 20 times? How does that make any sence? Don On Jan 19, 2:35 pm, Lucifer wrote: > @correction: > > Probalilty (a1 wins) = 24575/474552 = .051786 > > On Jan 20, 1:30 am, Lucifer wrote: > > > > > hoping that the cards are numbered 1,2,3,,13.. > > > Proba

[algogeeks] Re: probability of winning with two cards

2012-01-19 Thread Lucifer
@correction: Probalilty (a1 wins) = 24575/474552 = .051786 On Jan 20, 1:30 am, Lucifer wrote: > hoping that the cards are numbered 1,2,3,,13.. > > Probalilty (a1 wins) = 21723/474552 = .045776 > > On Jan 20, 12:47 am, Don wrote: > > > > > > > > > P= 8800/28561 ~= 0.308112461... > > > On Ja

[algogeeks] Re: probability of winning with two cards

2012-01-19 Thread Lucifer
hoping that the cards are numbered 1,2,3,,13.. Probalilty (a1 wins) = 21723/474552 = .045776 On Jan 20, 12:47 am, Don wrote: > P= 8800/28561 ~= 0.308112461... > > On Jan 18, 7:40 pm, Sundi wrote: > > > > > > > > > there are 52 cards.. there are 3 players a1,a2,a3 each player is given > > 2

[algogeeks] Re: probability of winning with two cards

2012-01-19 Thread Don
P= 8800/28561 ~= 0.308112461... On Jan 18, 7:40 pm, Sundi wrote: > there are 52 cards.. there are 3 players a1,a2,a3 each player is given > 2 cards each one of A=2...J=11,Q=12,K=13..a user wins if his sum of > cards is greater then the other two players sum. > > find the probability of a1 being t

[algogeeks] Re: probability of winning with two cards

2012-01-19 Thread Dave
@Sunny: The probability of a1 being the winner is not 1/3 because of ties. I.e., if a1 = a2 > a3, then a1 and a2 are tied and there is no winner. What we can say with no calculations is that P(a1 winning) = (1 - P(no winner)) / 3. Dave On Jan 18, 10:52 pm, sunny agrawal wrote: > isn't the answer

[algogeeks] Re: probability of winning with two cards

2012-01-19 Thread Sundi
:)... Lets say you have two players a and b one card is distrbuted to each player if the card with a is higher then a wins else a loses. probability of a winning: total num of cards with combos where a wins a is the first player: 2-1,3-1...13-1 3-2...13-2 .. 13-12 this sum equals = (13+12+...