From the article in highstakesdb: http://www.highstakesdb.com/3437-int...awrilenko.aspx

" the language of all the top players is the language of game theory, distributions, bluffing ratios, etc".

What??

In all the interviews with Ivey, Antonius, Dwan, Isildur, Jungleman I have never heard them talk about using game theory and math, quite the opposite in fact. Where does he get this idea from?

Guys like Sauce, Kanu and Ike all have strategies that are more game theory optimal.

Check out the latest pokercast for an interview with Sauce where he discusses this.

Even the players you mentions still think in similar ways.

I've seen past posts on here were Dwan discusses game theory optimal play.

Hahaha in today's games that's the language of anybody who is a long term winner at anything above $100NL! As the games have gotten tougher and edges have gotten smaller players have gotten more tight lipped in terms of strategy. Go search the archives (http://www.twoplustwo.com/archives.php) for some of durrrr's posts back in the 'golden age' of poker if this surprises you.

shaboy

Maybe they aren't using it directly but they are using things that work and likely a ton of math, game theory, etc. is involved. Maybe they don't know why they do it, but likely do it because it's an effective winning strategy.

I remember durrrr once boasting on High Stakes Poker that he could not play poker for 10 years and still be able to beat the games, so either he's really arrogant and was incorrect or he has a great handle on strategy

Poker is cool because people with vastly different strategies can succeed. the winningest cash game player (ivey) has no idea what any of those things are, but he knows enough from gameflow and experience that he could beat any of the top online players (was up 20m against game specialists like galfond at plo and hoss at limit)

ultimately the best players aren't going to be people who are smart and discover poker and make good analytic judgments even if they are broadly going to be more successful than gamblers, its going to be people who need the money from poker more than anything, who've grown up in desperate situations, so poker is their life (ivey). he's in his own class for that reason

inb4 poker evens out

How is this a thread?

It wouldn't surprise me at all if a few of those names don't think about it like "I need to have a bluff in my range here 20% of the time to be GTO" but rather while they're playing they deduce "based on this guy's image of me/this guy's currently tilting a bit/I flat out feel like gambling a bit" they ultimately end up running bluffs and altering their range distributions just by feel accidentally the right amount to be close to GTO. Maybe they realize what they're doing, or maybe their feel is such that they're just getting it right without realizing it...

I know one LHE player in particular who's super successful and definitely is far from a math wizard (you could say "hey what's 8 times 6" and he wouldn't spit 48 back at you in a nanosecond like most poker players could) and isn't thinking of things in terms of GTO and math. I think he's just accidentally getting it right just from his own instincts and has been for years.

Can't remember what book I read (some game theory nonsense) but it cited an example study of Italian penalty shootouts whereby both the striker and goalkeeper were making near enough GTO decisions on which way to kick/dive based on the players strength for one side or the other. I am pretty sure no professional footballers in Serie A are especially versed in game theory and I doubt anyone has even attempted to teach them. I guess it's just experience.

Edit: book was this: http://www.amazon.co.uk/gp/product/0...ms_ohs_product

durrrr and jungleman have made strategy posts using those concepts and similar ones.

I'm pretty sure Durrr argued something along the lines of "GTO does not exist in NLHE." He argued the point extensively with sauce in a HSNL thread. Not sure where that discussion ended, or if it ever did.

Anyways, I know lots of very good mid-high stakes regs who are terrible at math, but they seem to have an intuitive sense of near-optimal frequencies. Also, there are certain things that "feel" players can do SO well that it makes up for money they may lose playing less than optimal in other spots. I would expect that the same is true of the higher/highest stakes player pools as well.

I have heard him mention in interviews and posts that NLHE (Heads Up at least) was pretty much solved in from a Game Theory Optimal perspective.

even if some of the top guys don't specifically use the terminology they still know what they are doing from an abstract perspective, whether they know it or not

This is also discussed in some depth in Soccernomics.

http://www.amazon.co.uk/Soccernomics...3604005&sr=8-1

It basically states that despite some teams doing lots of analysis into which way their opponents penalty taker(s) / keeper are more likely to go and then passing this information to their own players to form a counter strategy, that actually the top players instinctively undertake a near "optimal" mixed strategy for this analysis to be of little use.

DaycareInferno: many scientists use clip boards.

Many truck drivers use clipboards too. I believe it's more instinctual than fully understanding the physics and math behind it though.

Jungleman did a strategy video (free i believe) on the topic of game theory and it's about 25% correct, 50% kind of correct but missing the point, and 25% factually incorrect. I also remember a very highly regarded DC coach also put out a video that covered game theory which largely was incorrect (this was a couple of years back though). I remember a long comments thread where i was pretty much ignored when i pointed out the mistakes until some of the fixed limit coaches came in and backed me up.

Of course they are both excellent players, much better than me theoretically and in practice. Just GT seems less important to their games, even at the high level. I think the majority of mid-stakes+ fixed limit players are fairly well-versed in game theory, have tackled MoP, and a lot of strategy discussion focuses around trying to balance ones range etc. From what i've seen from some at least public discussions, big bet guys are less well versed and/or get by just fine without it.

Of course these are broad generalizations and I think a lot of the top NL guys do a lot of range theory work too.

Software i wrote (shout out to Combonator) is used a lot for study of your own ranges to come up with balanced strategies; I think at least 50% of its pro users are LHE players, which given the 90/10 split in game popularity, is heavily limit-weighted. But I know for sure of some high-stakes NL guys who use it too; mostly HU i think.

As another example, I think if you said "optimal" to a fixed limit player many would assume you meant "game theoretically optimal". Whereas i've heard many NL coaches in vids say "optimal" to mean "most profitable."

On the contrary, I can't imagine that professional-level players aren't taking advantage of this stuff, or if not the players themselves, some coach or trainer.

Vanessa Ruosso definitely does, I believe she studied GF at Duke

Maybe we should take this to Poker Theory

I remember that too. I think durrr mostly just gave up when he realized he didn't know what he was talking about.

Optimal frequencies aren't known, so it's hard to say much, but there's a lot of reason to think that modern good-reg strategies actually don't look too much like GTO ones.

This is certainly true, but only insofar as other players are making big theoretical mistakes for them to exploit. It's probably not true, for example, in limit holdem, and won't be true forever in NL assuming people continue to improve.

I doubt he said this, and it's certainly not true.

GTO (game theory optimal) implies there is a solution for poker... like chess or backgammon etc. there isn't for hunlhe (or plo etc), and I'm very confident (but can't be sure b/c simpler game and I'm worse at it) there isn't for lhe etc. just about every good nlhe specialist disagrees with me on this... except jungle/galfond/few others who r actually good. altho i think galfond wasn't sure iirc or something like that (limit games probably ruined his brain).

Here's the relevant discussion between sauce and durrrr about game theory optimal poker/bots etc - http://forumserver.twoplustwo.com/19...46/index2.html

(starts around post 180)

Very interesting read, even if (like me) you're not smart enough to understand a lot of what they're talking about.

Players like Ziigmund balance their ranges very well, it's probably based on experience, feel and is somewhat opponent dependant I imagine. Sure it's not based on math but just like those soccer goalies you mentioned it's an approximation that is probably pretty close to optimal or no?

Because it's an interesting discussion