|
|
#131
12-10-2006, 02:13 AM
|
|||
|
|||
Re: Poker question from alphatmw
[ QUOTE ]
[ QUOTE ] [ QUOTE ] You are under the mistaken impression that a game theory perfect strategy is optimal against everyone. [/ QUOTE ] No i'm not-My whole point is that this statement is wrong. [/ QUOTE ] Well, there's not much to say other than that you're wrong. "Game theory optimal" and "best" are not the same thing, particularly when "best" is defined by "highest EV", and the highest EV strategy also happens to have large exploitable flaws that the opponent is not expected to exploit. [/ QUOTE ] I don't know your background, but I have absolutely no idea where you got this idea. I studied math and economics as an undergrad at the University of Oregon. I am currently a graduate student in Economics at the University of Pittsburgh. Without counting them I would estimate that I've taken a dozen courses (give or take a couple) that are either specifically about game theory or heavily involve game theoretic models. I have read all too many scholarly articles in theory as well as behavioral and experimental economics. I'm currently working on my dissertation which will be about modeling situations where there is one (or more hopefully, we'll see where this goes) informed agent who must decide whether to reveal her information to an uninformed agent. In all of my courses and reading (and writing obviously), I have not even 1 single time come across the use of the term optimal as you are using it. Optimal means optimal. In equilibrium each player is acting optimally obviously. However, nobody I know of would ever refer to a strategy as optimal that is not a best response to whatever the other players are playing. In short, "Game theory optimal" and "best" are the same thing, and are pretty much always followed by "conditional on" or something similar. The word you are looking for is "equilibrium," not "optimal." 
|
|
#132
12-10-2006, 03:17 AM
|
|||
|
|||
|
Re: Poker question from alphatmw
Curious,
[ QUOTE ] Addt'ly, in light of a recent thread, I'm fairly certain that resident game theorist JaredL will agree that what we're referring to here as exploitive/ative (and he calls optimal) is very much a part of game theory. [/ QUOTE ] It is. In equilibrium every player is maximizing their payoff given the play of their opponent(s). We should expect the game theorist and the psychologist to attempt to do the same. In reality, both will fail at this task as poker in any real form is much too complicated for people to play optimally all the time. Assuming the theorist knows an equilibrium strategy, she will stop playing it at the latest (in reality she wouldn't start playing it because the last part is true at the beginning) immediately upon discovering a flaw in her opponent's game, because it is no longer optimal. However, it is abundantly clear that a perfect game theorist will win. The reason is that she can play an equilibrium strategy and guarantee at least breaking even. If she deviates from the equilibrium strategy, which she undoubtedly will, then she must have a higher EV than she had from playing the equilibrium strategy.
|
|
#134
12-10-2006, 12:34 PM
|
|||
|
|||
|
Re: Poker question from alphatmw
I thought game theory was a way to play against an opponent who was also playing perfectly. i thought that was the point. If i'm wrong, i really would appreciate someone letting me know. Thanks.
|
|
#135
12-10-2006, 01:04 PM
|
|||
|
|||
|
Re: Poker question from alphatmw
[ QUOTE ]
[ QUOTE ] [ QUOTE ] keep in mind that i asked for the world's greatest game theorist, not a hypothetical perfect game theorist (unless this level of game theory is attainable). [/ QUOTE ] It is possible for no limit MTT short stack play. Only one decision. All in or fold. [/ QUOTE ] It has only been approximated. It uses a finite set of stack sizes, [/ QUOTE ] For all intents and purposes, this restriction is clearly negligible. [ QUOTE ] and ignores the fact that both opponents have the option of just calling the blind and then playing postflop, which will make the game tree much, much larger. [/ QUOTE ] But for short stacks (the only time we'd consider employing the JoF solutions) this restriction is again clearly negligible. Further, people have found solutions for the BB that allow arbitrary bet sizing from the SB. Just because you haven't heard about them/seen them doesn't mean they don't exist.
|
|
#136
12-10-2006, 01:07 PM
|
|||
|
|||
|
Re: Poker question from alphatmw
[ QUOTE ]
I thought game theory was a way to play against an opponent who was also playing perfectly. i thought that was the point. If i'm wrong, i really would appreciate someone letting me know. Thanks. [/ QUOTE ] If you know you're not sure of what you're talking about why did you just attack jogsxyz so authoritatively?
|
|
#137
12-10-2006, 01:15 PM
|
|||
|
|||
|
Re: Poker question from alphatmw
[ QUOTE ]
[ QUOTE ] I have always wondered if it is correct in Hold'em to bluff (or call) on every street with the game theory frequency even if it doesn't appear to make sense, or if you should just pick your spots and only bluff when scarecards hit or draws appear on the board. That's part statement, part question. [/ QUOTE ] Yes, game theory will make the proper bluffing frequency adjustments. With no threat of a bluff, opponent will not need to make any crying calls. [/ QUOTE ] If you scroll up to my first post in this thread (or alternatively open your TOP's to p. 186) you'll see that David clearly agrees w/ jogsxyz here.
|
 |
|
|
Linear Mode
