Here is an updated list of some corrections. I tried to stick with the ones that I thought might actually cause confusion rather than put down every grammatical error or misrounded number. Thanks again to B3lly for this.

Page 64: We almost always know <if> we hold the best hand with suited connectors…

Page 99: This means <the minimum bluff success rate> should be 33.3 percent of the time.

Page 128: And probably should <not> be made.

Page 182: Should read “as well as defend their checks as the preflop <caller> are often unwilling to discuss it.”

Page 189: For example #2, it says “nines and eights” when it should instead read “eights and sevens.” Nines are top set here and clearly want to get as much money into the pot as possible on many turn and river combinations.

Page 343: Y = ½ (1/(1+X)^2). In other words, the (1+X)^2 should be in the denominator.

Page 366: 0.386 should be 0.368

Page 380: In the table, 84.4 should be 79.4 and 60.1 should be 50.1.

Page 381: Since 60.1 should be 50.1, this now weakens the argument that only a small change in stack depth causes a large change in the fraction of value bets (to some degree).

Probably when the cold caller called some hands pre-flop in a mixed strategy, like calling a CO open in the button with AK or AA once in a while. In reality, if I'm at a table with no reads I probably never call an open then continue when someone 3-bets and then 4-bets.

I think so, but this one is trickier. I imagine QQ and JJ may sometimes call IP when UTG opens and MP 3-bets (or something similar), and likewise when button opens and SB 3-bets the BB might want to cold call some AQ or AJ hands (no way to know this of course though). What makes it especially tricky though is I absolutely hate cold calling AQ or AJ only to find out one of the other players had the same hand. In other words, it sucks when I have AQ, one of my opponents has AQ, and then the other guy has T9s. Pretty hard to get a good outcome in that situation so 4-betting might just be better.

Thanks.

Sorry to dredge this back up again, but I finally had a chance to think about it, and I am pretty sure now that what you say in the first paragraph is false.

Again, look at examples 20.1 and 20.2 in TMOP. In the first one, X has a bluffcatcher and Y has 7/8 nuts and 1/8 air, pot size 6 so A calls 6/7 (standard). In Example 20.2, we reach an identical situation (with identical ranges) on the river but X only needs to call 1/7. This is not contradictory, because we are maximizing two different things in the two examples, in the first one we wish to find the Nash equilibrium for a one street game, and in the second we want an NE for the entire two street game and dont care about the river in isolation.

I think all this is related to the concept of sub-game perfect equilibria, but it's late and I'm not about to go dig up old classnotes, so if someone who is familiar with that stuff can elucidate that would be great.

What an amazing book! I'm only a micro-stakes player but this has been incredibly useful for strengthening previously learned concepts. As a math nerd, I find it very helpful to be able to prove certain plays are seemingly optimal.

However, since there are no direct references to squeezing (from what I remember), I'm wondering how my 3-betting ranges will differ, if at all?

If someone can guide me onto how to use the pre-flop concepts to deal with possible squeeze situations, that'd be awesome!

I don't have TMOP in front of me right now (on vacation), but at the end of day a GTO player always just takes the most +EV line with every hand in his range. How can the line that is most +EV be different in two situations if both situations have the exact same ranges/stack depth/etc?

I of course understand how if you have different goals or are using different models you'll get different answers, but I don't see how that will change the fact that if everything is the same then both ranges will have the same maximally +EV lines.

I think chess provides a useful analogy. If a supercomputer is playing two different opponents where the games start off and progress much differently, but then somehow at some point reach the same position, the computer will choose the same move against both opponents. A grandmaster however might choose two different moves, especially if it was a close decision between them, realizing that the theoretically slightly suboptimal move would work best given the mindset that this particular opponent demonstrated by his earlier moves.

I have only played 22k hands since reading the book and trying to make adjustments but results have been good so far (0.1/0.25nl zoom). Wondering in general how others are going with adjusting their play. I feel i struggle a bit with defending weaker parts of my range, eg things like A3o, or K7o in the BB vs a BTN open. Also defending 3 bets by calling with some of the weaker holdings. Feels like I am just folding to much on the flop.

I don't know if this is allowed but Cardrunners currently have this video of Matt's up for free viewing:
http://www.cardrunners.com/poker-vid...matthew-janda/

Wow you certainly can talk fast. Anyway when did the hand examples, there was the examples of KQ and check-calling on A97 flop or something. Just not sure how this continues out on various turn/river scenarios. We seem to be hoping to turn a K or Q that is better than villains QJ or KT hands, and also use it as a bluff if the turn checks through we can bluff at rivers. What about other turn cards, are we just giving up on blanks, maybe check raising some turned gutshots (i dont think check calling could be +ev)

Thanks for the book and all the responses in this thread.

If you don't mind, how are sales going?

I've been working on my preflop game first. I want to get ranges pretty well memorized, adjusted to what I feel comfortable playing, etc. I too think it's pretty difficult to play those hands post flop, esp oop. It seems like whenever I call with hand like A3o in bb v sb open I flop absolutely nothing, no backdoors etc, and just end up folding to flop c-bet a lot. But prob the biggest leak in my game is I just c-bet way to often and then c-fold the turn way to often. I wasn't defending near enough checks so the video and he hand examples in the book have helped with this leak a lot. Planing to move to postflop adjustments after I get preflop down pat.

I really do wish cardrunners didn't have that $100 signup fee, I would snap get a sub to watch Mathews other videos to help get this stuff down.

I want to talk too fast in general. It's pretty easy for me to slow it when talking person to person, but I still haven't quite got it down to talk slower in videos (despite telling myself about 20 times per video to talk slower). Good to hear someone else mention it again though as I definitely need to work on it.

It's really hard in general to talk about what I'd do on all different turns and rivers, and if the situation looks really close then I think a lot of times it'd be a mixed strategy (though there's rarely a way to know for sure). The KQ example I'd probably do a lot of different things on different turns so it's really hard to say, but obviously we are check-calling the flop and hoping our opponent checks back the turn rather than bets.

Are you able to read the thread regarding that video if you don't have a subscription? Because I've answered some questions that are similar to this one but I'm not sure if you need a subscription to access it.

I actually liked the fast pace in the video, just was impressed that you could deliver it so fluently as well. So don't slow it down too much.

I don't see the comments to the video but that's OK, i think i get it more now and things become clearer with practice.

Hey Matt. Have Always been a fan, I always watch your videos and I bought your e-book
There is this guy who plays up to 10/20 and I think 25/50 sometimes with good results on PGC Forum here on 2+2 who said he didnt particulary like your approach to GTO in poker. Here is what he said:

"I dislike the book mostly because his overall approach (making worst hands indifferent between betting and openfolding) for approximating gto lines for preflop, flop and turn is wrong, and not just a little bit."

After being asked why he thought that, he said:

"Not that much to explain why his approach is wrong, it just hasnt got much to do with GTO, finding frequencies such that betting/raising ATC is 0EV is just completely arbitrary and irreleveant on any street but the river.
Better approaches are studying toy games or brute force imo."



Im really a noob at math and GTO in general, could you express what you think of what he said, Im kinda curious as I respect both you and him alot, I dont know what to think cause Im hugely uninformed.

thx!

The problem with making hands indifferent between betting and open-folding can probably best be seen pre-flop. As of right now (it may change), I don't think the BB should be 3-betting any hands against a button open that aren't also +EV calls. So, if you start with the assumption that the worst hand in your 3-betting range should have an EV of close to 0 and that's not true, then it will likely get you inaccurate results. I think this is pretty problematic for the pre-flop ranges and why I wrote the long(ish) response to LorenzoVMatterhorn's original comments earlier on in this thread.

Other times I think that assumption is less problematic and pretty useful. For example, if cutoff opens with 6h5h and the button calls and the flop comes Js Ts 2d, then I don't think it's a bad approximation to assume that it should be -EV for the cutoff to bet the flop with that weak of hand and the button should thus defend aggressively enough to prevent cutoff from being able to profitably bet any two cards. And if the cutoff bets and gets raised, he definitely needs to defend enough to prevent the button from being able to profitably raise any two cards (or else the button will never fold to the original bet, and always call or raise instead). Most people, especially as SSNL and lower, are constantly not defending enough in these types of spots (and if you're playing against opponents who aren't defending enough you should be exploiting them, and likewise you should try to make sure you're not getting exploited).

As for whether toy games or brute force are better for studying GTO poker, I think at the end of the day different people like different approaches and it's probably best to use a mix of everything. Someone playing NL$5000 is probably in the top 0.001% or so of poker players, and it's pretty easy to imagine a player like that thinking ranges made from models with loose assumptions are pretty bad and inaccurate whereas a SSNL player may think they're awesome.

It seems that people tend to overlook the subtle difference between making the worst hand in a whole range (or ATC) and the worst hand in a theoretical correct continuing indifferent.

- If we think that a GTO strategy in a certain situation does not involve continuing with 100% of the range (e.g. by calling or raising when facing a range), then inevitably there is a worst hand in the GTO continuing range that is close to 0EV (or the next worse hand is negative EV, i.e., the 0EV is between those hands)
- While frequencies for making ATC do not yield GTO ranges, they certainly can be used to prove that a certain strategy can not be GTO for situations as described above. Observe that being able to continue with ATC in asituation where we assume GTO play does not involve continuing with 100% range contradicts us playing GTO.
- There might very well be situations where a continuing range can consist of 100% of the original range, in which case the making ATC indifferent approach clearly ist not beneficial.

All of the above is stated explicitly in several spots in the book. Yet it seems that people who criticise this approach fail to recognise this. The frequencies might not lead us close to GTO ranges (and the further away from the river we are, the more this is probably true), but saying it is irrelevant as a tool for disproving optimality and exposing significant imbalance seems ignorant or a little bit lazy.

Well there's another subtle difference I probably should have mentioned yesterday which is just like a pre-flop BB vs button 3-betting range may include mostly strong hands which are also all +EV calls, it may be this is the case on the flop in some spots too. In other words, maybe every hand in a flop raising range would also be a pretty +EV call, but we choose to raise instead. While I do think there are a lot of spots where we should raise a pretty strong hand even if our hand isn't super strong (so calling it a "value raise" isn't a very useful term), for the most part I think it's pretty easy to see why in a reasonably polarized flop raising range probably includes some hands which are close to indifferent between raising and folding. That's especially true since optimal flop raising ranges will probably have some hands which are part of a mixed strategy (folding sometimes and raising others).

Oh, I've been betting the hands with least showdown value like 65, and checking the most showdown value like lowpairs/Ahi(Khi some situations). Trouble is though once i've checked all my strongest showdown value hands, my delayed cbet bluffing range on the river ends up being close to the villain's optimal call range, hands like AJhi..

Apologies if this has already been asked, but this is quite a lengthy thread to trawl through.

I was wondering what sort of results you've had playing the strategy you advocate in your book, Matthew? Like what sort of win rate do you have at what stakes?

Bought the book yesterday and although I've not got too far yet, I've really enjoyed what I've read so far.

I've answered that question (as well as put some additional details) in my blog:

http://www.cardrunners.com/blog/Matthew%20Janda

Just another quick question - I was wondering roughly what the overall VPIP/PFR/3bet stats work out as in practice, if you employ the preflop ranges you suggest in the hand charts section?

Short review:

Really eye opening book! Def def worth the 25 or w/e it is bucks. Almost finished it, def gonna reread it , and maybe another time after that

Hi Matt,

I watched your free video on Card Runners Classroom: You're doing it wrong part 1 and it seemed to me you were check/calling with a lot of 3 straight, 3 flush cards that I thought were bluff bets in the book.

I'm thinking specifically of hands #1 and 6.
1:
Open AcKh in CO, button calls. Flop is 9c 6c 4h, you recommend checking. Would this be a bet with Qc 6c 4h or AcKc on Qc 6h 4h flops?
6:
You open in the button with 8h7c, big blind calls. Flop comes Th 7h 4s, you recommend checking.

Is that because the board is more coordinated and lacks a single high card?

Hand # 5 is where I usually check:
You open in the button with Jd9d, big blind calls. Flop comes Th 9h 5d, which I like a check on also, because you hit so many broadway type hands with this flop.



Thanks,

Eric

PS - I'm very impressed by the book so far, and I've only completed thru the flop play section. I think a lot of it will apply to even my low limit live games.

Hope you don't mind me asking a question like this since I'm not paying you for coaching, so if you don't want to answer that's fine.

So I was watching your OOP 3-bet pot videos and you said in the third one that you don't like donking the flop unless your range has a decent amount more equity than the 3-bettors. So I started looking at the BB calling range vs BTN open in single raised pots and I have only generated 1 flop where we have a decent amount of more equity than the buttons. The most I have found was 54% and that was on a QJ9 flop. Just about everything else was around 50% or lower. I also noticed just like our OOP 3-bet calling range that a low flop like 965 just sucks for our range. We can have all sets but can never have a straight or 2p since all SC and 1 gappers are in our 3-betting range. So what I'm getting at is that it doesn't look like were hardly ever going to be able to donk from BB and low flops sucks to defend in single raised pots. So would you tweak this range any? Does it really matter if we can't donk that much or defend on 9 high or lower flops in the BB? Plus low flops are flops that we need to be c/r pretty aggro but we just don't have many strong hands in our range.

I was using the BB calling range vs BTN open you have in your book vs a 50% BTN opening range.

I don't really think this is a good thread for me to comment on videos I've made at CardRunners, as it doesn't have much to do with the subject of this thread and I'm not sure if that's fair to CardRunners or TwoPlusTwo.

That said, whether I bet, check-call, or check-raise 3 to a flush + 3 to a straight is going to depend heavily on the board texture, each player's range, how much showdown value I have, etc. I'll talk all three lines pretty regularly.

I don't think it matters at all if you can't donk from the BB on certain board textures (or even most or all board textures). Most people cold call much weaker ranges from the BB than they used to (especially against the button), so whereas donk betting was probably pretty good with much stronger BB cold calling ranges it's not a line I'd take often (if at all) in usual button vs BB spots. I've seen spots where I think the BB should have a donk betting range vs a button open, but I think they're pretty rare. Obviously it's not solvable so no one knows for sure, but I think you're spot on with QJ9 being one where donk betting sometimes looks pretty good (since if we always check our checking range may be too strong).

Again though, I'm going to have to be pretty good about not derailing this thread to talk about CardRunners videos.

can you give me the nickname of the guy here on 2+2 or a link where he writes those comments, i would like to learn more about what he said.thanks