Couldn’t parse what FAWS’s comment was saying, but the following may be similar in spirit:
Assume we have a two-player game with a unique Nash equilibrium. Now allow player 1 to make a credible threat (“if you do X then I’ll do Y”), and then player 2 must react to it “optimally” without making any prior precommitments of his own. If the resulting play makes player 1 better off and player 2 worse off than they would’ve fared under the Nash equilibrium, then let’s say player 1 is being “too pushy”.
For example, take the PD and additionally give player 1 the option to trigger a nuke and kill everyone. The resulting game still has a unique Nash equilibrium (defect,defect), but player 1 can threaten to trigger the nuke unless player 2 cooperates, then defect and get higher utility. Looks legit?
No idea what to say about games with multiple equilibria (e.g. bargaining games) or about the decision-theoretic context, which is way more tricky.
Couldn’t parse what FAWS’s comment was saying, but the following may be similar in spirit:
Assume we have a two-player game with a unique Nash equilibrium. Now allow player 1 to make a credible threat (“if you do X then I’ll do Y”), and then player 2 must react to it “optimally” without making any prior precommitments of his own. If the resulting play makes player 1 better off and player 2 worse off than they would’ve fared under the Nash equilibrium, then let’s say player 1 is being “too pushy”.
For example, take the PD and additionally give player 1 the option to trigger a nuke and kill everyone. The resulting game still has a unique Nash equilibrium (defect,defect), but player 1 can threaten to trigger the nuke unless player 2 cooperates, then defect and get higher utility. Looks legit?
No idea what to say about games with multiple equilibria (e.g. bargaining games) or about the decision-theoretic context, which is way more tricky.