| |
BGonline.org Forums
Backgammon, Poker, and PRs
Posted By: johnk In Response To: Backgammon, Poker, and PRs (Robert Wachtel)
Date: Sunday, 28 February 2016, at 12:15 a.m.
I agree with a lot of what Bob says - especially about the Giants list and against "PR shaming". But I do think part of tournament strategy is considering the strength/weakness your opponent even if it doesn't agree with the best bot play. That is not to say we could all play perfectly if we wanted to, just that the best bot play mght not be the best play over the board. This is especially true with cube decisions where for example doubling may gain 0.050 this turn on an easy take but you may be able to give your opponent a tougher cube decision next turn. Other examples are complicating/simplifying the game based on opponent's relative skill and making a riskier play with bigger payoff. Not saying all these efforts are successful or even correctly calculated but most tournament players use this strategy to some degree. All of XGs decisions are based on the assumption that it is playing another "perfect" player - ie, itself which is not the case in real life. Ray is probably the best example of a player who successfully plays the player, not the bot.
How would XG do in actual tournaments? We all agree it plays better than any human but by how much? What would be a fair handicap for XG. Would you enter a tournament where XG paid twice the regular entry fee? how about 1.5 times? Perhaps a paramutual system would let us tap into the "wisdom of the crowd" or XG could start with a point deficit. True it would be short-term results, at least in the short-term, but it would still be interesting to start collecting results.
| |
BGonline.org Forums is maintained by Stick with WebBBS 5.12.