Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Imagine an athletics race where there are three runners A, B and C. They all have an equal opportunity of winning. So the chances of A winning are 1 in 3.

How would I get to this answer if I only knew that the chances of A beating B is 0.5 and A beating C is 0.5 ? At first I thought that A beating B and A beating C were independent results and therefore the probability would be 0.5 * 0.5 = 0.25, but I know the answer should be 0.33

If the chances of A beating B was 0.4 and the chances of A beating C was 0.35. What is the chance of A winning the race? How do I go about working this out.

share|cite|improve this question
Why do you think that the probabilities you have are enough to work out the chance of $A$ winning in each case? – Chris Eagle Sep 20 '12 at 21:42
Ok, if I knew the probabilities of all permutations for all of the runners. Would I then be in a position to work out the probability of A winning? If all runners had .5 probability of beating each other. Could I work out the chance that A wins? Excuse my lack of knowledge in this area, I haven't done much probability for a while. I've read on other posts that I could perform a pairwise comparison if I know all individual probabilities. Is this correct? If I can understand how to worn out the answer j can then move on to scenarios where runners have different probabilities of beating each othe – Scott Sep 20 '12 at 22:05

If all you know is that $A$ beats $B$ with probability $1/2$ and $A$ beats $C$ with probability $1/2$, you can’t determine the probability that $A$ wins the race. That information is consistent with the possibility that each runner is equally likely to win, but it’s consistent with other scenarios as well. Suppose, for instance, that $B$ always beats $C$. Then the only possible outcomes are $ABC,BAC$, and $BCA$. Suppose that these have probabilities $p,q$, and $r$, respectively. Then the probability that $A$ beats $B$ is $p$, so $p=1/2$. The probability that $A$ beats $C$ is $p+q$, but it’s also $1/2$, so $q=0$. Thus, in reality the only possible outcomes are $ABC$ and $BCA$, and $A$ wins with probability $1/2$, $B$ also wins with probability $1/2$, and $C$ wins with probability $0$.

This is a rather implausible scenario, but it’s possible in principle, and it’s consistent with the given information.

share|cite|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.