> The conflict you describe is that some people want more of a fair fight, and
> others put more weight on wanting "my team to win". Of course the second
> group doesn't want to win via too easy or obvious an advantage. They may want
> the rough appearance of fairness, but in fact want enough unfairness for them
> to win.
> Can we model this behavior as resulting from rational agents, or is some
> irrationality required to make such a story work?
It doesn't seem to require any irrationality. Say insisting on an unfair
game brings you benefits but has the cost that people may complain. It
seems natural to assume that the costs created by complaint will increase
as the unfairness of the game increases. If the "complaint-unfairness"
curve crosses the "unfairness-advantage" curve, then people will be more
more fair. Dictators, for example, have pushed the "complain-unfairness"
curve down by ruthlessly hurting dissidents. In democratic societies, the
costs imposed by complaints can be high enough to force people back to the
crossing point of the curve. Now that you put it this way, I'd say it's a
nice econ 101 problem.