rationalskepticism.org

GrahamH wrote:

iamthereforeithink wrote:jerome, what's your opinion of Paul the Octopus?

A 6% chance:

wikipedia wrote:Assuming Paul's predictions were no better than fair coin flips, the probability of 12 or more successful predictions in 14 attempts is p = 0.0056 (~0.56%), as given by the binomial distribution.[36] And the probability of 8 successful predictions out of 8 attempts is p = 0.0039 (~0.39%).[37] However, the first three matches were in the group stage where the outcome could have been a win, loss, or tie, resulting in a 1/3 probability of getting the result correct. As a result, the true probability of Paul's feat is [(.3^3)*(.5^5)] or .12%. Of course there were many public predictions about World Cup results using quirky methods that were wrong and hence did not get international attention. It is an xample of the Prosecutor's fallacy to use the same data that brought Paul to fame to suggest that he has precognition. Paul started to receive international recognition after he correctly predicted Germany's win over England, and after that he only made four correct predictions. The chance of those final four predictions being correct is 0.5^4 = 0.0625 = 6.25% (odds 15/1).