Also interesting how the behavior of the repeated prisoner's dilemma differs from the repeated red/blue game. The repeated prisoner's dilemma converges to an optimal strategy of "tit for tat" - you signal your conditional cooperation, but also punish defections. The repeated red/blue game converges to an optimal strategy of always choosing red. The blue-pressers will most likely be wiped out in the first round, and if they are not, they will be wiped out in some round in the future, leaving only red-pressers left in the population.