Also worth checking, out of interest: How does each of the bots perform when playing against itself?
Good question—after running some matches:
AnderBot Cooperates until one of the copies randomly triggers a Defect, then both copies defect for the rest of the match.
CheeseBot, DMRB, SHTFT, SimpleTFTSeer, Switch, TatForTit, TwoFacedTit, and VOFB always cooperate with themselves.
Pip always defects against itself.
For a next version of the tournament, perhaps it would make sense to force each bot to play a match against itself.
This ensures that everybody playing DefectBot is no longer a tournament equilibrium (not that this is probably going to happen, anyway).
Self-cooperation was one of my core test cases, so I guess this probably indicates that many other people also designed for and/or tested self-cooperation.
Huh.
Also worth checking, out of interest: How does each of the bots perform when playing against itself?
Good question—after running some matches:
AnderBot Cooperates until one of the copies randomly triggers a Defect, then both copies defect for the rest of the match.
CheeseBot, DMRB, SHTFT, SimpleTFTSeer, Switch, TatForTit, TwoFacedTit, and VOFB always cooperate with themselves.
Pip always defects against itself.
For a next version of the tournament, perhaps it would make sense to force each bot to play a match against itself.
This ensures that everybody playing DefectBot is no longer a tournament equilibrium (not that this is probably going to happen, anyway).
Self-cooperation was one of my core test cases, so I guess this probably indicates that many other people also designed for and/or tested self-cooperation.
Huh.