Rock-paper-scissors

First, let me say that I know probably just enough mathematical game theory to be dangerous, having watched the first 9 lectures or so from a Yale course on the subject. Lecture 8 actually discusses rock-paper-scissors. The Nash equilibrium is to play every move with equal probability. But the Wikipedia article on the game says "unlike truly random selections, it can be played with skill if the game extends over many sessions, as a player can often recognize and exploit the non-random behavior of an opponent." This is true because the Nash equilibrium is not the best strategy, it is the situation where both players are using, and know that they are, the best strategy against the strategy of their opponent. Neither has anything to gain by changing his strategy. Larps like to assume that all players are playing the Nash equilibrium, so the result will be random.

I contend, as I have here, that people aren't good at generating random numbers. Most simply, this means that players cannot play the Nash equilibrium, as they can't choose from rock, paper, and scissors randomly, with no correlation to previous plays. They may produce an even distribution of the three moves, but that is not random. Once one player is not playing randomly and his opponent has some information about this, it is strategically better for his opponent to not select randomly with equal probability. So even if you could pick randomly well, your opponent isn't going to be, so you are best off not doing it.

Now, rock-paper-scissors is a game so this is all fine and good. Except that larps think that it's not a game, as my understanding goes, they think it is a random number generator, a way to replace dice. Because of this, peoples in larps who are good at rock-paper-scissors will get what they want more. Or maybe cause moves the other way, so players in larps are better than normal people at rock-paper-scissors.