Economic and Game Theory
|
"Inside every small problem is a large problem struggling to get out." | |||||
Topics | Thread and Full Text View
I'm developing a game where a small group of players will rate each other's performance and the best rated player will win. The problem is: players will have a twisted incentive to give out low ratings, because this increases the chances that they're the best player. To clarify the problem, here's a simple example with 3 people and a 0 to 10 rating range: player A (fair) - gives a 9 to B and a 6 to C player B (fair) - gives a 10 to A and a 6 to C player C (cheater) - gives a 0 to B and a 0 to C If we simply do the average, we'll notice that A ends up with a final score of 5 B ends up with a final score of 4,5 C ends up with a final score of 6 So C will win the round because he cheated and didn't give fair ratings to his peers, and A will be second place, while he should be first. So the final rank would be: C A B While the correct rank should be A B C Is there a theoretically optimal and preferably simple solution for the final score calculation for such a game, instead of just the simple average? I must be able to rank the players in the end. Does this kind of rating system have an official name? Thank you. [Manage messages] |