>Strategies like Tit-for-Tat (TFT) are context-sensitive and not universally "optimal."
Of course. Hence why I didn't claim it to be universally optimal. All I'm saying is that your willingness to defect against defectors should always be non-zero, just to keep the players in line who start out with a higher predisposition towards defecting.
The noisiness of the real world should probably bias us more in the direction of cooperation, to avoid a cascade of defection as you mention, but a player who only cooperates will get taken advantage of regardless of the precise details of the game. Some amount of this dynamic can be seen currently in the relationship between western companies and the Chinese state, a relationship that is currently very different from the relationship between Chinese companies and the rest of the world.
It is also generally true that the longer the game, the more defectors suffer.
Of course. Hence why I didn't claim it to be universally optimal. All I'm saying is that your willingness to defect against defectors should always be non-zero, just to keep the players in line who start out with a higher predisposition towards defecting.
The noisiness of the real world should probably bias us more in the direction of cooperation, to avoid a cascade of defection as you mention, but a player who only cooperates will get taken advantage of regardless of the precise details of the game. Some amount of this dynamic can be seen currently in the relationship between western companies and the Chinese state, a relationship that is currently very different from the relationship between Chinese companies and the rest of the world.
It is also generally true that the longer the game, the more defectors suffer.