How far has NL poker come? How far do we have to go? (abstract/long)

[w_alloy]

[ QUOTE ]
Way way way too many variables to have "PSC convergance"... as one thing changes it all changes.. kinda like economics, you can only analyze things when you keep A, B, and C constant, etc.. (I've only taken first year econ courses so don't bother cutting me up.. lol)

[/ QUOTE ]

I definitely agree its all connected, and I actually wrote something about that in the OP but it was kinda convoluted and the post was already long enough, so I deleted it.

But basically I think that the interconnectedness makes studying individual stats tough, but there are some exceptions. Simple preflop stats like VPIP and PFR have certainly become more uniform over time, and definite trends have emerged. Also I always keep fold flop to cbet displayed, and while this used to be all over the place even for decent players on party, on stars this year it has been much more consistently in a certain range for my tough opponents.

Also, knoweldge about what stats actually mean could be gained by watching how they move together. And the way I defined PSC, it doesn't suffer from this problem anyways (debateably).

[franzinator]

Really interesting read .

[LearnedfromTV]

msnl perspective. This is fun stuff but I think we're very very far from unexploitable play.

The following is very hypothetical. But let's say 23/19, distibuted some particular way by position, is optimal preflop. So if you put six 23/19's who play optimally postflop at the same table, they break even (minus rake), and if you replace one with a 30/20 or a 20/10 who plays optimally postflop, the new guy gets exploited into going broke or becoming 23/19.

Fine. That's what it might look like. But even if preflop numbers are mostly converging somewhere among decent players, that hypothetical optimal ecosystem is really delicate. And if you disturb it with a couple fish and a couple mini-fish, like people who play 23/19 but suck (they're everywhere! - I'm one of them), not to mention a [#$%@!, etc] shortstacker.. well, even just for preflop, profitable play and optimal play diverge really fast. And that doesn't even touch on the "optimal play postflop" assumption being ridiculous. Maybe people are approaching something like optimal cbet% for a given pfr, etc, but there's no way it extends much further, like into double-triple barrel frequencies; if anything, the number and type of possible boards makes this sort of thing very unlikely.

I'm not sure what the right scale is, so these numbers are just made up to illustrate a point, but if, say we've gone from an unexploitability factor of 10/100 to 20/100 in the last year, I think the only way we get close to/over 50 is by taking a bunch of really good players, putting them in a closed system - no new fish, and when some of them turn into fish and go broke, they can't reload - and let them battle. With new money flowing into the system it'll never get close.

[LearnedfromTV]

One other thing is that whatever optimal play is for a six max table where everyone has 100bb, it's different for the same table where everyone has 150, or 200, or for five-, four-handed, etc; so the actual situation we have, where every table has a unique distribution of stack sizes, and the stats we're looking at are split over different numbers of players, makes it even less likely we can find convergence toward optimal stats. Too many variables, and in a lot of cases outcomes are very sensitive to small changes in the variables.

[El Diablo]

raptor,

"playing unexploitably is absolutely retarded and not the most profitable way to play, and its not even close."

Not once PSC gets to 11, baby!

[g-p]

[ QUOTE ]
playing unexploitably is absolutely retarded and not the most profitable way to play, and its not even close

[/ QUOTE ]
just found a huge hole in my game

i try to hard to be unexploitable

[king_of_drafts]

[ QUOTE ]
[ QUOTE ]
playing unexploitably is absolutely retarded and not the most profitable way to play, and its not even close

[/ QUOTE ]
just found a huge hole in my game

i try to hard to be unexploitable

[/ QUOTE ]

QFT raptor is a genior

[w_alloy]

[ QUOTE ]
msnl perspective. This is fun stuff but I think we're very very far from unexploitable play.

The following is very hypothetical. But let's say 23/19, distibuted some particular way by position, is optimal preflop. So if you put six 23/19's who play optimally postflop at the same table, they break even (minus rake), and if you replace one with a 30/20 or a 20/10 who plays optimally postflop, the new guy gets exploited into going broke or becoming 23/19.

Fine. That's what it might look like. But even if preflop numbers are mostly converging somewhere among decent players, that hypothetical optimal ecosystem is really delicate. And if you disturb it with a couple fish and a couple mini-fish, like people who play 23/19 but suck (they're everywhere! - I'm one of them), not to mention a [#$%@!, etc] shortstacker.. well, even just for preflop, profitable play and optimal play diverge really fast. And that doesn't even touch on the "optimal play postflop" assumption being ridiculous. Maybe people are approaching something like optimal cbet% for a given pfr, etc, but there's no way it extends much further, like into double-triple barrel frequencies; if anything, the number and type of possible boards makes this sort of thing very unlikely.

I'm not sure what the right scale is, so these numbers are just made up to illustrate a point, but if, say we've gone from an unexploitability factor of 10/100 to 20/100 in the last year, I think the only way we get close to/over 50 is by taking a bunch of really good players, putting them in a closed system - no new fish, and when some of them turn into fish and go broke, they can't reload - and let them battle. With new money flowing into the system it'll never get close.

[/ QUOTE ]

While I think a lot of this post is correct, and I'll be the first to acknowledge that this whole thread has very limited applicability to day to day play, I think you are forgetting something.

That vast majority of high stakes poker decisions online are made heads up post flop between good players with 100bb+ effective stacks. As I said in the OP, mid stakes games usually have a (often delayed) reaction to but dont have much of an effect on the PSC.

[VanVeen]

it does not logically follow that strategies will converge toward game-theoretic optimal play or intrinsic unexploitability. the selection pressures acting on the player pool are not such that this is true. relatively speaking, there isn't a huge gap between your average 50nl grinder and the best players in the world when it comes to grasping the game-theoretic properties of the NLHE strategy landscape. none whatsoever. the computational limitations of the human brain, even the best human brain, do not allow anyone to explore the terrain without a blindfold on.

what is a logical necessity is that strategies will become less exploitable given the parameters defined by: a) the cognitive limitations of the players, and; b) the decision-making biases, both emotional and cognitive, of the players.

the best players probably: a) have frequencies that jive because of their own decision-making biases and quirks; b) a superior ability to exploit the decision-making biases of other people by out 'leveling' them; c) better emotional control than everyone else - information processing and decision making aren't as adversely affected when under stress (or are affected in a non-conventional adaptive way).

i'm sure as time goes on people will get a bit better game-theoretically and they'll probably get a bit better at exploiting the weaknesses of other people. how much room is there to go? who knows. eventually the skill level at the top will pretty much max out and thereafter strategy variation won't trend in either direction - it'll be random until someone either starts crunching #s (and using those #s to do more and better trial-and-error tests) or they invent super smart/emotional equanimity pills.

[ImsaKidd]

[ QUOTE ]
playing unexploitably is absolutely retarded and not the most profitable way to play, and its not even close. on the stt forum all they talk about is playing unexploitably, but who cares? all that matters is what is the most profitable. unexploitability is not it, even at the highest levels.

[/ QUOTE ]

So true.

Sometimes I will make a certain play (usually a fold), and I will think "Wow, this is so exploitable." That usually turns into "I should not make an exploitable play".

Obviously, the problem is that your opponents dont see hands you fold. They will never be able to exploit that because they dont even know that exploitation exists. Very apparent, but might be something a few people still overlook.

I dont play HSNL, and I'm a shortstacker. That probably means my thoughts have 0 credibility, am i right?