MIT Grad v. Joe Schmo -- The Perfect Move (long)

[Knockwurst]

D.S. raises some interesting issues in his response(s) to Lee Jones et al. Besides the psychological dynamics, astutely analyzed by andyfox, another question D.S.'s response brings up is the following: Is there a knowable objectively "correct" play -- the perfect move -- in the play of a poker hand? Is best play knowable?

In chess, the position and candidate moves can be calculated so that often the best move can be determined -- sometimes hours later, days later, or even years later, and after much analysis, the correct continuation is determined. See Kasparov's book "My Great Predecessors" Vol. II Tal-Smyslov where the best continuation for both players was found decades later. And the debate is still not closed. There are also positions in chess that are so complex and unclear that they cannot be exhaustively analyzed even by the strongest computers such as Deep Blue. In these positions, some lines are more correct lines of play than others but the most correct line often cannot be determined.

It seems as though poker is more often analogous to the second type of chess position, the unclear one where some lines of play are certainly more correct than others, but the "objectively" best line of play is difficult if not impossible to determine. To determine the absolute best line of play involves weighing so many factors (ie. static factors such as position, cards, number of players, money in the pot etc. as well as dynamic factors such as the personality of the players and how they are affecting the dynamics of the entire table at that point in time)that it is impossible to determine what is the most correct line of play -- or at least the most correct play is debatable. Even S&M have recommended making a play that is less than best (ie. -EV for that particular hand) in order to affect future play against you. (Likewise World Chess Champion Em. Lasker would go into inferior lines of play because he knew his opponent was not adept or was uncomfortable in such positions).

I guess this is a long way of saying I'm surprised that D.S. would insist so strongly that there is one correct play instead of a range of plays, some of which should be done with a greater frequency, as well as his insistence that if a 2+2 author says it's so then it's so. End of debate. Even the 10 8 8 flop when holding pocket queens would seem to allow for a range of plays. Perhaps betting out most of the time, check raising some of the time and check folding on occassion or even check calling are all correct in a given situation. It's the debate that gets us closer to the answer. (If Kasparov were ever to suggest that his analysis was the last word because he was the best player in the world -- he would be duly ridiculed among chess players)

By the way, in a heads up match of a random MIT Grad v. a random 2+2 poster, I'll take the 2+2 poster every time [img]/images/graemlins/grin.gif[/img].

[EdSchurr]

Well, chess is significantly more complex than poker.

However, you're probably right that with imperfect information, there would be no correct play. But that doesn't mean there isn't a play better than another.

[razor]

[ QUOTE ]
By the way, in a heads up match of a random MIT Grad v. a random 2+2 poster, I'll take the 2+2 poster every time.

[/ QUOTE ]

If you are capable of drawing this conclusion, I'm not sure why you would have any difficulty drawing the conclusion that Ed is very likely correct on any point Ed disagrees with Lee Jones on that Ed is very sure about.

[Andy B]

Ed Miller is neither a random MIT grad nor a random 2+2 poster. If the random MIT grad were given the necessary time to evaluate the game, I'd probably take him over the random 2+2 poster, given that there are some pretty random posters around here. If you make the 2+2er somewhat less random, say only people who have been posting for a year or who have > n posts, I'll go with the 2+2er.

[jedi]

I think you're comparing different things here. DS and Ed are pointing out Lee's errors in default situations in a low limit game. These are monstrously wrong errors. Even if Lee were right in certain situations involving player types and the like, he didn't articulate them, therefore the recommendation to check-fold for example is made in a default game, which is horrible.

In a hand like the QQ on a T88 board, there IS a play that garners you the best expectation in the absence of more information. Of course, if you do have more information, then you should use that to adjust your play accordingly, but when DS and Ed say there's a right way to play this, they're talking about a default situation against typical players for this type of game (low limit in this case).

[Easy E]

any chance that Lee is taking into account the stress of poker playing on beginning players? And trying to remove some traps that could tilt them off?

I haven't read the book in a while- I may have to go back and look at it again.

[jedi]

[ QUOTE ]
any chance that Lee is taking into account the stress of poker playing on beginning players? And trying to remove some traps that could tilt them off?

I haven't read the book in a while- I may have to go back and look at it again.

[/ QUOTE ]

Ed and DS's criticisms of the book aren't on the parts that are geared towards beginners and are conservative. They are on the parts that are plain wrong. By plain wrong, they don't mean the close wrong decisions, they mean the very wrong ones, the ones that differ greatly from the correct play.