Inactive Tutor answered 01/04/22
We use C() for combinations and P() for permutations below.
It's interesting to start with an argument assuming independence, then see how closely you come.
If 3 people have equal chances of a selection and we make 3 selections, the probability of each getting 1 is
3!/(1!1!1!)(1/3)^3 = 2/9. We should expect our answer to be a little greater than this.
The number of ways that we can have 1 player with the 2, 1 with the 3, 1 with the 4, and 1 with the 5 is
4! C(48,12)C(36,12)C(24,12) = 4! * 48!/(12!)^4
The number of ways that we can have 1 player with the 5 of clubs and nothing lower is 4 * C(48,12) (any of the 4 can have this hand).
Then, the next player can have C(39,13) hands, and the next player C(26,13) hands. C(39,13)*C(26,13) = 39!/(13!)^3.
Thus, the total number of hands with one player having the 5 and nothing lower is
4 * C(48,12)*39!/(13!)^3
Thus, given this hand with the 5 and nothing lower, the chance of the other 3 having the 2, 3, and 4 in any order is
4! * 48!/(12!)^4/(4 * 48!/(36!12!)*39!/(13!)^3)) =
6*13^3/P(39,3) =
13^2/(19*37) =
169/703 =
.240398
As we hypothesized, this is a little bit greater than 2/9 = .222222, the probability assuming independence, as we expected.