I talked about this with a few guys from work, and they were all quite convinced that I was completely, totally and utterly wrong. So I proposed a bet.

"OK," I said, "let's play the game a number of times. Say twenty times - the more often we play, the more random chance gives way to probability curves. The idea is simple. Each round you pay $25. I have three cards, the ace of hearts and two black kings. You have to find the ace. You pick one card of the three that I put down face down. I will then reveal one of the other cards that is not the ace. If you want to switch after that, you pay another $25 - but why would you want to, if you have a 50% chance? I then reveal the card that you finally picked and if it is the ace, I will pay you $60. That must be a neat way for you to earn money; if the breaks are 50/50, as you maintain, then on the average you earn $5 with every game."

In spite of their absolute conviction that I was wrong, nobody dared to take the bet. Cowards! :)

Whymme