An urn contains one red ball and one blue ball. A box of extra red and blue balls lies nearby. George performs the following operation four times: he draws a ball from the urn at random and then takes a ball of the same color from the box and returns those two matching balls to the urn. After the four iterations the urn contains six balls. What is the probability that the urn contains three balls of each color?
- A)
- B)
- C)
- D)
- E)
Answer
B
Key insight
Every sequence of two red and two blue draws has the same probability 1/30, because the denominators are always 2,3,4,5 and the numerators are 1,2 for each color.
Solution
Ending with three of each color means exactly two red draws and two blue draws in some order. Compute the probability of one specific order, say red, red, blue, blue:
- Draw 1: red of balls, probability ; urn becomes .
- Draw 2: ; urn becomes .
- Draw 3: blue, ; urn becomes .
- Draw 4: blue, .
Product: .
Any other order gives the same product: the denominators are always (the urn grows by one each time), and the numerators are the current counts of the drawn color, which are then for red and then for blue regardless of interleaving. So each of the orders has probability .
Total probability: .
The answer is .
Why this works
This is a Polya urn. Its hidden symmetry is that the probability of a sequence depends only on how many of each color were drawn, not on the order: the numerators for each color run in turn and the denominators run . So probability of a final composition (number of orderings) (probability of one ordering). Whenever sequential draws change the urn, check whether the products are order-independent before drawing a full tree.
Alternative approach
Track states with a tree. From : or each with . Continue to the four-draw level and add the probabilities of paths ending at ; by symmetry compute the paths through and double. Paths through : RRBB, RBRB, RBBR each , sum ; doubled gives .
The trap
Multiplying probabilities as if the draws were independent with fixed probability 1/2, or forgetting to count all 6 orderings.
Common mistakes
- Multiplying probabilities as if the draws were independent with fixed probability 1/2, or forgetting to count all 6 orderings.
- Computing the probability of a single ordering () or of the three orderings starting with red () and stopping there.
Techniques
Define states/recurrence and iterate · Exploit symmetry to reduce work or pair up objects