This was my point exactly. Proving that an action is a "mistake" means showing that it is not the optimal play according to the GTO strategy and that it is not exploiting your opponent enough to risk being exploited back. In other words, unless we know the GTO strategy, it's hard to say for sure what is a mistake and what isn't.

You also forgot Zadeh.

Thank you, I hadn't heard of him before. I found that you (David) wrote some articles about his work for the twoplustwo strategy magazine last year are they still available online anywhere?

Some of Zadeh's more recent publications also look interesting, but in a different way

I'll see what I can come up with then

I'm not a full-time player so I can't really even give a guide on beating NL10. Most of my best received posts so far (e.g. my 1000th on Average Bubble factors and 2000th on Kelly Criterion in MTTSNGs) have been maths ones where the numbers stand objectively on their own merits, rather than ones where I have to say "In my limited experience ..." so I might try something along those lines. A lot of microstakes tournament players have trouble with dividing their bets over the remaining streets (e.g. they bet half pot on the turn then find themselves with 1.4x pot on the river and not sure what to do) so that's something I might look at.

Thread resurrection.

GTO is useful as a baseline for play against unknown opponents you assume to be good. Couple of fatal problems with focusing on GTO as far as I can see -

- if you enter a game against opponents you don't know but assume to be good, you're a lousy game selector.

- if you stay in a game where you can't find an exploitative strategy that works better than GTO because everyone's so good, you're a lousy game selector and some kind of masochist.

- if you're in a game with a bunch of weak players then you of course exploit all their mistakes, not protect against being exploitable.

- if you play low stakes and try to approximate GTO instead of exploitative plays the rake will eat you alive.

What are their mistakes if not regular deviations from optimal play? So, an understanding of what theoretically correct play is is necessary to exploit mistakes. Or am I missing something?

People knew what mistakes were and how to exploit them long before theory heavy GTO discussions came into vogue.

Studying GTO and playing around with solvers is definitely not useless, but seems to me its utility is rather marginal relative to the hype.

Yes but those 'mistakes' were much bigger back then. Poker is tougher and mistakes are much more difficult to spot and thus exploit well, unless you have a decent idea of what is theoretically sound...

'Im going to 4bet wider v this guy cuz he seems to 3bet a lot' does not cut it anymore

How would one go about playing this GTR strategy and does it apply to BJ?

Since this thread got resurrected...

True, I also forgot Norman Zadeh, even though I have his book in my library. I also forgot to mention Nesmith Ankeny, who described a GTO-strategy for 5 Card Draw. Unfortunately I couldn't find any reports about practical tests.

If anyone is out there, Andrew has published a book on GTO, called Play Optimal Poker.

1. Not true. You may know they aren't good, but don't know what mistakes they will make. Imagine playing someone who has literally never played a hand of poker before.

2. You may be in a game like this because you have run deep in a tournament. Staying in the game is probably optimal.

3. See one.
4. It depends on how bad the players are. Also, had bad the rake is. But, once again, if you know your opponents are bad, it is better to start off GTO, then exploit once you see the errors.

1 You don't have to "know". There's a probability space based on human nature - e.g. that people who call on the flop are likely to also call the turn.

2 is a fair point.

I agree on 1, we can definitely guess at the kinds of mistakes an unknown, and clearly inexperienced player will make. My example was probably too extreme.

Imagine, however, that some day an ET comes to play poker with us, and has never played poker before. It understands the rules perfectly. We wouldn't have any idea of the kinds of mistakes it makes to start out, therefore, playing GTO would be correct, at least until we figure out its weaknesses.

OK, I realized I just gave an even more extreme example...

Game theory and "playing GTO" are not the same thing. In fact I kind of hate the term "playing GTO" period but I accept that it seems stuck in the poker lexicon. Game theory is a tool for thinking about poker/analyzing situations. Sometimes even when you know what mistakes a player is making, the maximally exploitive strategy is not obvious, and understanding game theory is useful for finding less obvious exploits. It can also be useful for finding mistakes in the first place - if you don't know what equilibrium would look like, you're going to struggle to spot potentially exploitable deviations from it. It's not always as simple as "he calls river too much!".

Not All Games In This World To Wich GTO Can Be Applied Are Alike.


GTO breaks even (guarantees 0 EV) in some games or situations.

GTO is plus EV against non-GTO play in other games or situations.


GTO flat loses bigtime in other games or situations, like when military science fellows try to apply it to the inferior side in asymmetrical war game situations-- sometimes finding the actions that lose the least is the best we can do.

Zadeh's 1974 book is of great historic interest and was way, way ahead of it's time.

A number of players successfully employeed his lowball and draw system, with a few modifications, including myself in lowball.

With draw and lowball, Zadeh provided tables with:

Uninformed Strategies
Informed strategies

The body of the book and the appendix were rich sources of poker theory that others built on but rarely credited.