Without considering it too much it seems like a simple example would be a game where some invisible dealer picks a number 1-100 and the three players have to guess as close to the number as possible without going over it and take turns going first could result in a similar situation.

Just throwing this out on intuition. I haven't analyzed it yet, but it seems like this should be an example. I believe two people with an unusual strategy could at least ensure the game is -EV for the third player.

Not sure if this question makes sense as I don't know much game theory, but with respect to game space (someone above said that the game space is much larger for chess than it is for poker), would not the game space for poker depend on the stack sizes.

Similarly, to the guy who says that ranges are very narrow in games with a huge number of players, surely again this depends on stack sizes. With 1 000 000 bb stacks, I'm pretty sure every hand that can make the nuts should be in your opening range, even from UTG.

I think the effect of stack sizes saturates at some point where it is unlikely that you get your stack in anyways except in a huge cooler situation that comes up very infrequently. Maybe this point is around 1000-1500BB in a no-ante game.

lol this game is massively rigged IN FAVOUR of the third player.

however if you were to make this game symmetrical (rotate the button every hand) maybe it would work, because the other 2 players can screw the first guy quite easily.

A much more simple game.

3 people play. The rules are simple and based on some sort of twist of the prisoner's dilemma. An invisible dealer randomly picks a number 0 or 1. The goal is to guess the number with one specification: if two people guess the same number then they lose regardless of the result. So it's clearly impossible to solve as the 2 other players can very clearly make it -EV for the first player by simply having one of the two always guess the same thing, and the third player guess whatever remains.

EDIT: So he will never win when he's first or second to go, and given that he'd be getting at best 3:1 on his answer, he'd have to be correct 100% of the time when he's third to go (to break even) and that's impossible.

Define 'solvable'. I am not even really sure what this means. If you are playing against ******ed droolers, sure, I agree that there are systems that will work. If you are playing sngs, once you get to the later stages, sure, you can come up with a system that will break even at worse. However, you will also encounter people that know how to exploit these tendencies when you get to the higher levels, making it near impossible to come up with a fixed system that can consistently beat them.

Now, at some point i think it is entirely possible to figure out some kind of system that will break even or maybe even slightly win, but so much depends on the players you are up against, how they adjust to certain situations. I also think it would be hard to convince me that this system would ever be optimal, and would merely get by with marginal winner at best, and that is a huge concession.

Late-phase SNG strategy is solved to an extent that there is no exploiting these tendencies by mysterious super-players.

Solving poker imo means deriving the game-theoretical optimum for all streets and also devising exploitive strategies against opponents who make mistakes (against those the GTO-strategy still at least breaks even but might not win the maximum).
Obviously we are far from achieving this, but it could theoretically be done imo. The same can be said for solving chess, with anything resembling today's computing technique the tree of moves is far too big for it to be solved.

What's game theoretical optimal for a game where the optimal strategy is -EV? Like the game I suggested, you can't just say there's an optimal strategy that breaks even at worse - you have to have some sort of reasonable proof of such. Multiway games make it possible where there's no strategy that can avoid losing.

To solve poker, you have to solve the play versus every different player in the world. That is at least my view.

It is optimal to bluff vs player X in this spot, but vs player Y it's losing. And then comes the variable of playing on tilt, or some other feeling that distorts your play, so you have to know exactly what mood your opponent is in and solve every different mood to have solved play versus one opponent. But of course, before it is solved, you also have to prove that if he is for example tilted he will play the same everytime, if he's drunk he will play the same everytime. I doubt this is the case, but it could never be proven anyways.
i'm talking headsup right now, but put this into a full ring game, and add the factors of multiway pots etc and see how he reacts.

Poker isnt complicated, it's just unfathomable because of certain factors. The right move is ever changing, while in chess, the right move is constant.

not really: the right move vs a strong grandmaster might be to go into a trivially drawn ending when the right move vs a weak player would be to go into a position where you are worse with lots of chances.

I still think chess is clearly a more complicated game, but you are going the wrong way about demonstrating that (or the opposite).

damn you gotta catch up i already agreed that it's not actually solvable.

Heads up is definitely solvable though

poker is more complicated if u count in $$$$. strong BRM and strong nontilting mentality is essential. if u look at poker as just a "freeroll" than it doesn't make sense but then chess wins. good BRM isn't so important for chess.

There's only four streets in omaha/hold'em, and seven streets in stud games. Draw games typically have even less streets of play. This means that there's much less ground to cover logically than chess. And since both games are just as flexible IMO, I'm gonna have to go with chess on this one.

The only thing that separates poker from chess in terms of complexity is the role of intuition, which has almost no role in chess, unlike poker.

Most top players would think that Intuition plays a relatively large role in chess, especially in speed chess. In poker, since you generally have such a short time to act on each street then intuition is more likely to play a big role.

unknown information vs more pieces

/thread

True. It's like comparing apples and oranges, but still interesting to cover in detail.

I don't think speed chess in particular is being discussed in this context, though -- speed chess, if anything, reduces the complexity because you're forced to think about less than you would otherwise anyways. Also, I think you're mistaking intuition for good observation, in which case factors such as anticipation are derived logically.

comparing the games themself chess is obviously way more complicated. However there are other factors in poker that make it more complicated in some aspects. For examplein poker you play based on reads because its a game of incomplete information and adjust because of this.

Poker is WAY more complicated than chess.

Set up a chessboard midway through a played game. Take out one of the players and replace him with another chessplayer.

The incomming player does not need to know how the pieces arrived at their states. The incomming player does not care about the play style of the opponnent or his tendicies. The incoming player need to care about who the previous player he was replacing was. The opponent the player is facing is irrelevent so is the previous history of the game. The correct move doesn't depend on any of those factors.

Now, switch out a poker player midway through a hand a replace him with another on the internet.

To make the correct play the player would have to know what happened in previous betting rounders.

How the opponent has played in the session leading up to the hand.

How the player he is replacing has played so far leading up to the hand.

Then the new player would have to figure out way play correctly balancies into the range of the previous players plays.


Thats the thing, You can set up a chessboard take a situation in a previously played game analysis it and find out the correct move.

You take a poker situation midway a session and try to analysis it and you cannot find out the correct move without knowing everything that has taken place from everyone perspective and even then you probable will still not be able too.

Chess is simple you can find concrete linear answers. Poker is hard you find nonconcrete nonlinear answers.

That being said, Poker appears easier because the compitition is no where near playing a form of optimal poker where in chess players are playing much more closer to optimal for that game so the game apears harder because the compitition is much better.

Carded, you need a better reason than the psychological aspects alone or you just said that roshambo is more complex than chess. And that aside, what you said isn't entirely true for chess. The vast majority of positions have countless different possibilities that all players are going to handle very differently and the opponent is very important. Eg - if you're playing somebody who likes quiet positional/maneuvering games then going for the most complex variations is a good idea, and vice versa for a guy who loves complexity. This, I think, was one major reason Kramnik was so effective against Kasparov whereas if you gave another player of a similar ability but a different style the same positions Kramnik had then the results could easily have been drastically different.

Range balance is not psychological it is mathamatical. You are only pointing out how little you know of poker which basically agrees with me claiming people think poker is easier than it is because they aren't even close to scratching past the surface of how complicated poker can be.

This whole post is really random. There's a game-theory optimal answer to any HU poker problem and possibilities to exploit through deviation based on knowledge of the opponent. There's a game-theory optimal answer to any chess position and possibilities to exploit through player knowledge (if you don't think chess has that aspect, it shows how little you know about chess). Multiplayer poker has an added complexity. HU limit seems clearly simpler than HU NL or chess.

I disagree. For chess their is a game theory optimal answer for a situation. For poker there are game theory optimal answers.

I'm sure people will refute that statement better than I can.

One thing : if there is an optimal move everytime , 200 000 000 positions per seconds computer and the best players in the world can't find them all.
It's so complex that it does not really matter does it?