I think when people link the nash equilibrium strategy with poker they fail to realise that one of the variables in the GTO approach is that for every possible stacksize you would have to slightly adapt your distributions.
Ofcourse GTO applies to poker, but the rigid strategies people commonly think of (push fold) when they associate Nash with are totally wrong, it is exponentially more complex than that.
And who wants to play GTO anyways? Then the game reverts to luck :P. exploitative is way more fun

People confuse GTO with best. GTO doesn't mean 'best'!

The whole point of GTO is to minimize the maximum loss possible against any other strategy. This also has the consequence, in many games where the GTO solution can be easily shown, of minimizing wins. For instance in roshambo, GTO is obviously to randomly select one of rock/paper/scissor - but it is literally impossible to ever show an advantage over any player with the GTO 'solution'.

Exploitative maximizes wins vs a specific strategy with the obvious downside that it is almost certain an exploitative counter-strategy exists which would result in accordingly large losses. Again roshambo is the obvious example. If you're facing somebody who's doing nothing but rock,rock,rock then anybody with a brain would start going paper,paper,paper until they adjusted, but the GTO strategy would still just blindly pick randomly since the adjustment of playing paper would open yourself up to exploitation. So maybe you're asking well then couldn't GTO be random unless your opponent has chosen the same symbol for the past 20 rolls in which case you roll the opposite the next time. No. That wouldn't be GTA because if somebody knew your strategy then they could exploit you by rolling paper 20 times and then rock on the 21st time. GTO is about making it impossible for other strategies to exploit you - not about exploiting other strategies. In GTO you literally don't care what your opponent does as you are simultaneously playing against all possible strategies - not your opponent's particular one.

GTO has a very specific meaning and it's unclear GTO poker is even desirable, let alone something top players would actually strive for! I'd generally assume that when people say GTO they actually mean balanced+exploitative, that's not at all what GTO is, but in the poker player vernacular that's what it's turned into: "My opponent is raising 2bb to win my 1bb 40% of the time. If I want to play 'GTO' I need to defend at least 33% of the time or he shows an instant profit!" There's at least two things technically incorrect with that statement, but isn't the meaning more important than the delivery? And given that while there may be only 1 true GTO solution there are effectively infinite poker player 'GTO' solutions.

I'd be more surprised if the GTO bot beat me, at 100bb deep hunl, for less than 5bb/100 than I would be if it beat me for more than 20bb/100.

In some games, like rock/paper/scissors, GTO play automatically breaks even and only exploitive play beats suboptimal play. Optimal play in RPS makes your opponent's decision irrelevant, every play they make is 0ev.

I don't think HUNL, or any full scale poker game, is like that. To take a very obvious example, optimal HUNL play clearly does not make the opponent indifferent to folding AA preflop rather than raising with it. It also doesn't make shoving 32o for 100bb a breakeven play. These are extreme examples, of course, but I think it's clear that the concepts apply to more realistic decisions. Playing too loose or too tight is just worse than playing well regardless of the opponents response. You don't need to "exploit" these mistakes to profit from them.

re: thread title
lol, no they don't

someday when i have more time on my hands i'll explain y this is wrong- but it clearly will take a while. If i had to disclose my strategy to u and u to me n we played then theres an optimal strategy (i think this is nash equilibrium but it doesn't interest me so i dunno much abt it). if you play a strategy that is the most optimal were i to know it... then ur gonna lose a lot- b/c its pretty easy to figure out what that is, and i can play similarly back but exploit occasional spots and you won't be sure if I'm doing that. if you then adapt ur play based on how I'm playing u aren't playing GTO anymore (GTO is a strategy that doesn't need to be adapted).

edit: i think most (probably all?) 2player games are solvable unless they involve incomplete information (and some of those are but only the very simple ones).

u can stake these bots, happy to play and we lockin to a buncha hands. but the bots strategy can't be changed from whatever it starts as. id finally need to use a hud

hoss is right btw

in after durrrr omg i can still feel his aura

i'm very interested in this if you mean the LHE bots.

@durrrr: you cannot exploit a GTO strategy as it will be equally balanced in all spots - as will it's calling ranges. Bluffing too much will lose, folding too much will lose, calling too much will lose. Doing anything other than the optimal move in accordance to every possible hand you could have in every possible spot will weight your range in such a way that you will be slowly losing.

how about a 100bb NL game where all stacks reset to 100bb after the hand?

I'm pretty sure there's no LHE bot playing actual GTO at the moment. The only question is whether it's close enough that basically no human player could beat its strategy and would end up slowly losing the way you describe.

I'd have to go with yes, plus Durrrr's never been known for his limit game, I'd assume he was talking about a NL bot.

HU limit 5 Card draw high is a good example of how a GTO strategy could be "exploited" and unable to win. What if the "gto bot" was set to calling down a player who Stands Pat before the draw some small percentage of the time.... You could just Stand Pat every single time and instant profit... if the Bot cannot adjust to the fact that you are standing pat bluffing every hand it can't beat you. No matter what % of the time the bot would call a stand pat, it could be exploited. 2 player game, 5 card draw high HU... Any strategy in this game that was unable to adjust to opponents could be beat.
If the bot was set to call a stand pat every time, that could be exploited to **** by patting value hands pre and getting paid every time.

Again that's not correct.

Bluffing too much will not exploit it.
Folding too much will not exploit it.
Calling too much will not exploit it.

There's a huge difference between not exploiting it and losing to it. Even starting to prove what/if any strategies a GTO strategy would inherently exploit is absurdly complex.

Either I said it the wrong way round or you are reading it wrong, but I agree with you

For the reason you just described the GTO bot does not care what the opponent does. GTO does not 'adjust' as any direct action taking against a specific strategy can be exploited by a slightly modified counter strategy. GTO play would be ridiculously counter intuitive.

But what I am saying is that if the GTO does not care what the opponent does in this specific game, it will not profit. If it was preset to call down Stand Pats with some specific range and/or frequency, it can be exploited no matter what that frequency is.

What I'm saying is that the stand pat part is irrelevant. It wouldn't 'hand read' but rather use optimal frequencies based upon all possible distributions of hands you could ever possibly have and all actions you could ever possibly take.

I realize it wouldn't hand read or adjust. That is precisely why i am saying 5 Card draw can not have a winning GTO strategy. No matter what % of the time the GTO thinks is an "optimal frequency" for calling down a stand pat, i could exploit it.

If you play against LHE bots, the bots owners will be very happy.

I have a question. Does the gto mean it plays the same way all the time? Or is it a set of euqations with variables that take into account what the opponent is doing?

Nash bots play the same no matter who he's playing, how long they've been playing. It's a strategy that's been stored in a hard drive, and it just looks it up in the hard drive what strategy to use when situation comes up. I think for those top LHE bots, the space they took is often larger than 100 GB.

You keep stating calling down. You are again viewing things as players would - that since you stood pat presumably you are polarized, nutted or bluffing. GTO play would not make such an assumption and would play in an extremely counter intuitive fashion that is very difficult to elaborate upon in a non theoretical level since it's difficult to even imagine what it would look like! The entire point of GTO is that it's impossible to exploit it - again that does not mean it's exploiting you though. Proving that such a solution exists was nobel prize worthy, so I don't presume to understand let alone be able to explain it on a forum!

Before you leap to where you're going with that question - playing the same does not mean taking the same action given the same situation. Randomness will be a huge part of any GTO solution. GTO will play the 'same' but in the same exact situation it will likely take any number of different actions with some degree of weighted randomness that yields an unexploitable strategy.

Yes but what i mean is the frequencies are set and not dependant on the opponent and hows hes playing?