Raashan, Sylvia, and Ted play the following game. Each starts with . A bell rings every seconds, at which time each of the players who currently have money simultaneously chooses one of the other two players independently and at random and gives to that player. What is the probability that after the bell has rung times, each player will have ? (For example, Raashan and Ted may each decide to give to Sylvia, and Sylvia may decide to give her dollar to Ted, at which point Raashan will have , Sylvia will have , and Ted will have , and that is the end of the first round of play. In the second round Raashan has no money to give, but Sylvia and Ted might choose each other to give their to, and the holdings will be the same at the end of the second round.)
- A)
- B)
- C)
- D)
- E)
Answer
B
Key insight
Only two kinds of holdings occur, (1,1,1) and a permutation of (2,1,0), and from either one the next state is (1,1,1) with probability exactly 1/4.
Solution
The total is always . A player can gain at most in a round and must give away if they have any, so nobody ever reaches . The only possible holdings are or some arrangement of .
From : each of the three players picks one of two recipients, equally likely outcomes. Everyone ends with exactly when each player receives one dollar, i.e. the gifts form a cycle: or . That is of , probability .
From , say Raashan , Sylvia , Ted : only Raashan and Sylvia give, equally likely outcomes. For everyone to end with , Ted must receive exactly one dollar and Raashan must receive none: Raashan gives to Sylvia and Sylvia gives to Ted. That is of , probability .
So no matter what the holdings are before the final bell, the probability that the th ring produces is .
The answer is .
Why this works
A process with many steps looks intimidating until you list its states; here there are effectively two. Because the transition probability into the target state is identical from every state, the distribution before the last step is irrelevant and the huge number is a red herring. Always compute one-step transition probabilities first and look for such a coincidence.
The trap
Trying to track 2019 rounds or assuming the answer depends on the parity of 2019; the transition probability into (1,1,1) is the same from every state.
Common mistakes
- Trying to track 2019 rounds or assuming the answer depends on the parity of 2019; the transition probability into (1,1,1) is the same from every state.
- Miscounting the case as or by treating the three choices as a permutation instead of independent choices.
Techniques
Split into exhaustive cases and handle each · Define states/recurrence and iterate