Derren Brown recently had a show in which he flipped ten heads in a row. He just flipped coins all day and waited for it to happen eventually. If I flip a fair coin, I should believe there's a 50% chance it will come up heads. If I flip it three times, I should believe there's a 12.5% chance it will come up heads three times. If I have eight goes at flipping it three times, it seems I should believe there's a 100% chance of flipping three heads. If that's right, what's wrong with being increasingly confident at the beginning of each set of flips that this will be the one in which I flip three heads? It's obviously a bad argument: every time I fip the coin, there's a 50% chance it will turn up heads. But how could it be rational for me to bet that during the course of a day of coin flipping I'll flip three heads eventually but not be rational for me to be increasingly confident that the next set of three flips will be of three heads as the day progresses?
Read another response by Peter Smith
Read another response about Probability