When fair standard -sided dice are thrown, the probability that the sum of the numbers on the top faces is can be written as where is a positive integer. What is ?
- A)
- B)
- C)
- D)
- E)
Answer
E
Key insight
Every die shows at least 1, so subtract 1 from each: distribute the 3 leftover pips among 7 dice, which is C(9,3) = 84 with no cap ever reached.
Solution
There are equally likely ordered outcomes, so is simply the number of ordered -tuples with each and .
Each die shows at least , so set . The condition becomes
The upper bound is automatic: no can exceed , so no die exceeds , well under . We are simply distributing identical pips among dice.
By stars and bars, stars and bars can be arranged in
ways.
The answer is .
Why this works
A sum of dice is a distribution problem once the mandatory on each die is removed. The shift converts "at least 1" into "at least 0," which is exactly the setting stars and bars handles. Always check whether the upper limit can actually be reached; when the leftover total is smaller than , it cannot, and no inclusion-exclusion correction is needed.
Alternative approach
Casework on the pattern of extra pips: three dice each show ( ways), one die shows and another (), or one die shows (). Total .
The trap
Forgetting that dice cannot show 0 and counting solutions to a sum of 10 over 7 nonnegative variables, or worrying about the upper bound 6 when it cannot bind.
Common mistakes
- Forgetting that dice cannot show 0 and counting solutions to a sum of 10 over 7 nonnegative variables, or worrying about the upper bound 6 when it cannot bind.
- Using or from a misremembered stars-and-bars formula; the count is with bars.
Techniques
Map the objects to something easier to count