NO!

I think you may be confused about what a theoretical GTO strategy is. It has a very specific definition as originally proposed in the book, Mathematics of Poker. A GTO strategy is one where if you and your opponent are using it, neither has any incentive to deviate and neither are exploitable. If one does deviate, then in the long run, he will be a loser. That doesn’t mean that against certain opponents that you can’t do better by exploiting a non-GTO player but by doing so you put yourself at risk of being exploited.

*edited to add* YES i am definitely confused about what most players MEAN when they mention GTO*

Ok so this is defining it a bit better.

So you would agree that 2 perfectly GTO players cannot profit from each other and they would in theory be break even? Or would you not agree? I can't seem to comprehend how 2 players playing a perfect GTO strategy (even if we haven't discovered fully what that means) could profit from one another.

GTO seems to best in a world where your opponent is also GTO - because if your opponent is not able or willing to exploit you when you deviate from GTO then there is no incentive for you to practice GTO vs that specific villain type.

Or are you suggesting GTO is a system in which there are many true/correct/profitable solutions that can all in theory be profitable but there are clear outsider solutions that are clearly wrong/bad/not profitable? So then one GTO players might have an internally consistent but different GTO strategy from another player???

Do you have either book and if so how far are you in them?

This stuff is explained in a lot of detail within both books. I'm glad to help out but I'd rather not try to explain concepts in a few paragraphs that I've spent many pages explaining in the books.

I just bought the first book, the applications one - But before I begin to read it My brain was being zapped with different posts and answers within this thread - And now I am left believing that nobody has provided a coherent and internally consistent definition of what GTO actually IS.

It almost seems like when people discuss God. Everyone has a different personal opinion about what God actually is or means to them and many people have a difficult time providing a definition for it, even though they tell you they believe or do not believe in this God, etc etc.

Some say GTO is NOT exploitative play. Some say that by playing GTO you are naturally exploiting villain's (whether it's the intention of GTO system to do that or not, it's like an inherent outcome of the system) - Some say GTO is just "well reasoned and good poker" , etc

By defining GTO you can also define what GTO is NOT. And are there multiple variations of a proper GTO strategy that 2 players can develop?

But if you truly feel all this is defined ad nauseum in the book then I will just read the book...

Yeah just read the book. Or start here and just keep reading and clicking new wikipedia links or reading sources as they pop up (this is probably going to be a harder read than any poker book though):

https://en.wikipedia.org/wiki/Nash_equilibrium

The new book pretty much just starts out with explaining GTO conceptual stuff but if you've already got the old one just give it a go (*cough* potentially skipping the pre-flop chapter *cough*) and if there's something specific that seems wrong let us know. But a lot of stuff gets explained bits at a time since it's impossible to explain everything simultaneously, so if something seems unanswered just remember it's probably coming in later content.

Unfortunately, when you write a book it's not like coaching someone 1-on-1 poker, so you'll probably have to read a lot of stuff you already know or read some stuff you're less interested in to get to the stuff you most care about (in your case it sounds like technical explanations of what GTO is and how it works). But that information is definitely in those two books with lots of detail.

Ok thank you, I will study the book and see if I make sense out the wikipedia link you provided.

Found this video , so my questions about whether or not GTO includes exploitative play is answered (They seem to be mutually exclusive) and also the question about if two perfectly GTO players were playing versus each other, the neither would profit - I assume that is the conclusion because GTO is defined as "a strategy in which your opponent cannot beat you" - this obviously implies that if player A is perfectly GTO and player B is perfectly GTO then neither palyer would profit and they would just lose to the rake.

I just got the book in the mail! Your book looks about as intense as I expected. It should be a very good read. I opened up to a few random pages and I found small little math equations on each of the first 5 pages I opened up to. This is EXACTLY what I'm interested in.

So to say I am eager and determined to seriously take my understanding of poker theory and math to the next level...can you recommend if your other book (the newer NL holdem for advanced players) has the same material just rewritten, is it MORE advanced and math heavy, is there NEW material?

Will I be disappointed when the 2nd one arrives and find it is very much the same book with the same information because applications book is over 400 pages and seems pretty decisively exhaustive in it's approach.

I'd just go one step at a time and read the book you've already purchased before worrying about buying another one.

Isn't this like saying "An optimal boxer will not counter any of your techniques or footing inside the ring. He will of course follow his own strategy of throwing 2 punches from the left, one hook from the right and side stepping in alternate fashion to whichever hand throws the punch. Because if he changes his strategy, his opponent will be able to exploit him. By not adapting to his opponent and remaining steadfast in this own strategy, this boxer cannot be beaten. He is optimal."



If my HU opponent folded 1,000 times in a row to my 1,000 consecutive jams pre flop, this is so obviously NOT optimal! This is a huge leak and he is being exploited to the ends of the world.

Why is the Terminology of GTO using within it's main concept a word "optimal" that does not actually mean optimal?

Reminds me of this...



Maybe we should change GTO to GTP (Game Theory Perfect) or GTEV (Game Theory +EV) or GTB (Game Theory Balanced)

I think I read somewhere that the originators of the term Game Theory Optimal are sorry they used the term because of the confusion it has caused. The strategy is optimal in that it is the best strategy against someone also playing GTO. Against someone not playing GTO, it is not necessarily optimal – not necessarily the best; however, aside from rake, GTO still guarantees against long term loss. In other words, you can’t beat a GTO player heads-up in the long run.

From the sticky thread, started by Will Tipton in 2012(!), in the Poker Theory forum:

Apologies if this point has been brought up before (if someone could point me in the right direction if so that wouldbe much appreciated). I notice in the preflop section that the cold call IP ranges are much wider than a lot of good players recommend, and I believe pokersnowie flats much tighter than the ranges in AoNLHE. Has your view on this changed significantly since time of writing? Or do you think that in a lot of relatively soft games flatting much more hands is a good exploitative strategy?

If you do think that in theory IP cold call ranges should be tighter than the ranges in AoNLHE, are there any adjustments you would make to the post-flop chapters because of this (e.g. OOP checking some boards 100%) or do you think the concepts apply as long as you consider the new range instead of the old ones?

Thanks for any thoughts on the topic!

I think the pre-flop section of the book aged worse than any section of the book by a pretty wide margin.

GTO pre-flop ranges likely will look very, very different from the best ranges to use in the actual games you're playing though. IN other words, if the games you're playing give you a big enough edge to beat the rake, you probably can play many hands pre-flop that in theory would be BE or -EV vs GTO opponents (especially due to the rake).

I think defending OOP and the bet-sizing information in Applications aged really well. I think there's probably a fair amount of updating I could do if I re-wrote the book, but I think for the most part it's conceptually sound and wouldn't be worth the time (or extra $40 readers would have to pay) to get the updated book when Applications mostly got it right, mostly.

IIRC when Applications came out, the most common PM I got was how I was recommending way too much checking vs strong villains when the biggest winners CB way more often even at MSNL (and probably HSNL). So like I'd recommend CB 40% check 60% in some spots, and people that that was way too much checking. Solvers/Snowie in many of those same spots actually recommend checking 90%+.

Hi Matt,

I'm really enjoying the book. With respect to part 5: 'whether to bet or ch in position' are there any notable adjustments you would make to the summary points you make at the end of the section with regard to deeper (300bb) live full ring NLHE games? I would say that overall pf ranges are usually wider due to higher implied odds and often many players targeting playing pots with deep weaker players and also the population is generally less aggressive esp in terms of flop ch rz. Thanks

Hi Matt,
after discussing 4-betting and 5 betting % on page 46 you mention that AK is not part of a value 3 bet range against UTG open (since opener never felts with worse) and that this complies with "maximum 3-betting ranges" on page 41. However the table on page 42 shows AK as part of the value portion of a 3betting range vs UTG open (or the portion you defend with against a 4bet). Can you explain how AK is not a value 3 bet in that case? Its also still ahead of some of the 4 betting range by a 15% RFI open as seen on page 46 (TT+, AQs+, AKo) so we could still be 4 bet by worse by the opener.

Thanks!

Sorry about the late response.

The deeper you get the more I'd emphasized robust equity, position, and overbetting.

You have to keep in mind the term "implied odds" only works well in specific situations. Calling extra wide out of position 300BB because of "implied odds" is likely a terrible idea, as the player in position is going to be able to capitalize on the stack depth much better than you.

I dislike the terms "value bet" and "bluff" for any situation other than the river, unless we're dealing with very very polarized ranges (such as check-raising the flop with only air or sets).

I would not worry about whether 3-betting or 4-betting AK is a "value bet" or "bluff." If it's denying equity and making the pot bigger in case we win in an effective enough way to justify the raise, that's all that matters to me.

Hi Matt! Im almost finished with the book.. But i found something odd in one page..
On page 438 in sample hand 3, in "positional range after flatting flop" we have AcQc and KcQc in our range and the turn brings a club draw to the board, but these hands disappear from our continuin ranges? We end up to the river with mostly spade draws that were on the flop, but we fold turned club draws? This doesnt make sense to me since wouldnt those club draws be better in terms of implied odds?

Hi Matthew !

I'm French and very happy to talk with you !

Preflop Question : How did you calculate how much the BB loose (ex : -4,5 BB) when his worst 3bet is called by the BU ? (p69 of the book) I know this is a recurrent spot, and i would like to calculate how much do i need to call the 3bet in my RFI range against different opponent in my pool.

Other preflop question : You also said that's pretty hard to build a 4bet calling range, so when I try to build on, i can actually call lots of hands due to pot odds. Am i wrong to only look at pot odds to determine my call frequencie ? i've heard about needed 35-40% equity to call, How can we get this stats ?

Thanks for your time !

There is no way to solve what the true EV of the BB's worse 3-betting range is once called. We could determine how much the BB needs to win on average once called to break even after looking at how often the BB (after 3-betting) gets a fold or faces a 4-bets, but there's no way to directly calculate EV after 3-betting and getting called.

You can call small bets with a 4-betting range due to pot odds. You cannot convert equity into EV though, so there's no way to know whether a call is +EV or -EV based on equity alone. In other words, if you get 3:1 odds on your pre-flop call, you can't simply say "Well, I only need 25% equity for my call to break even." That's not even close to how it works, so if you make this false assumption (which is a pretty common mistake for players to make when they start trying to implement theory) you're likely to end up taking massively -EV lines.

I would never consider AK as either a 3-bet "bluff" or "value" 3-bet anymore. The terms just don't work and largely for this reason pre-flop does not model well.

I wouldn't consider TT or JJ a "value" 3-bet either.

You are 3-betting to deny equity and make the pot bigger in case you win. There's no threshold of strength where a hand magically becomes a "value 3-bet".

Almost certainly a mistake, unless I assumed we raised them (which sounds unlikely just based on what you've written).

Thanks for your answer, so what are the others informations to take account to build our 4bet calling range and how to use them ? ( i assume pot odds, 4bet frequencie, position, sizing)
And with this informations can we try to estimate the equity needed ?

Hi Matthew,

I am looking at the samples at the end of the book and have a question about the continuation betting frequency. The book discussed that your betting frequency on a later street is dependent on the prior street's bet sizing, but the samples doesn't seem to take that into account. It assumes all streets are bet for 75% pot, then doesn't this mean that a later street should be bet 70% of the time? The continuation percentage seemed arbitrary in these samples.

Thank you.