One of the reasons Matt's book is so solid is that he follows his logic through to the end. For many years, he has always insisted on looking at "contradictions" in your thought process to see if you might be overlooking something -- since it's hard to mathematically prove lots of points. It's with that critical method in mind that I think there are a couple of seemingly contradictory points in the book.

PREFLOP vs FLOP VALUE TO BLUFF RATIO:
Matt uses two different methods for determining the value to bluff ratios of preflop vs flop. I think the postflop technique is stronger, and he skirts the issue preflop. Let's first look at postflop. Matt determines the value to bluffing ratio by looking at a perfectly polarized range and determines the ratio of value to bluffs to make the opponent indifferent to calling on any street. He digs in deeper to non perfeclty polarized ranges as well, but the thought process is nearly the same which is to make your opponents bluff catcher indifferent to calling and folding. However, preflop he doesn't use the same methodology when determining proper 3-bet value to bluff ratios.

Preflop, as best as I can tell, he uses a process of determining the maximum that either player could "bluff" so that their opponent can't re-raise any two cards. The problem with this method is that it assumes that our opponents are always either raising or folding, and while he says that's likely the case when facing a 4-bet, it's very rarely the case when facing a 3-bet. Our opponent will often defend against a 3-bet by flatting. Matt makes an assumption on how much the 3-better wins back with his bluffs which will help us determine the minimum the flatter needs to defend, but he never reexamines the 3-better value to bluff ratio to make sure that the weakest hand in the defenders range is indifferent to calling or folding (or nearly 0 EV-- otherwise, he won't be able to defend or he'll defend much wider).

I can understand Matt's hesitation in addressing preflop the same way he addresses postflop because equity changes so much between preflop and the flop (since you see 3 cards, instead of one like on the turn and river), so it's difficult to model. While this might not be contradictory thought processes, it does suggest using two different methods for the same question without distinguishing why or validating the preflop version.

finally got my hands on the book. so far it looks like what I expected - top notch poker theory material.

first think which struck me tho was the foreword on "bluff catching". Such a mediocre advice in otherwise highly professional book. I guess Matt wasn't too happy about this being part of his book.

My guess is that he would have preferred to have Ben Sulsky do the into... who doesn't want Sauce to write an intro to their book. However, taking a theoretical line when you're confident an exploitative line is higher EV doesn't make sense and the "bluff catching" forward doesn't say much more than that. It's obviously targeted at live low stakes games which will evolve as well and is kinda obvious but outside of that i didn't have any problem with it and doubt Matt did either. And from the structure of the book, like all 2p2 books, and the acknowledgements Matt probably learned quite a bit from Mason and David on how to organize his material... although i've always thought his stuff was pretty accessible.

@page 55, there is a tiny error in the SB resteal calculation, is more a typo however...

Amount won by stealing = blinds + open raise = 1,5 + 2,5 = 4
3bet size = 3bet size - SB = 9,5 - 0,5 = 9
4 ( X ) - ( 9 ) ( 1 - X ) = 0
4X - 9 + 9X = 0
13X = 9
X = 0,692

The book states 0,693


@page 56
4bet success math
( amount won by stealing ) ( X ) - (4bet size ) ( 1 - X ) = 0
ex
BTN opens 2,5, SB 3bets to 9,5X, BTN 4bets to 19
amount won by stealing = blinds + 3bet size + open raise = 1,5 + 9,5 + 2,5 = 13,5
4bet size = 4bet size - open raise = 19 - 2,5 = 16,5

The book states 12,3 instead 13,5, where I'm wrong?

Great read.

Mind posting your revised SB and BB 3b ranges v CO open like you did in post 62 v BTN open?

If there is a mistake where 0.693 is put instead of 0.692, this isn't something I plan to correct. It may be a rounding difference, and even if it isn't it's so insignificant it's probably a waste of time to even have someone read the correction.

You're double counting blinds in the second question. If the button opens to 2.5BB and the BB 3-bets to 9.5BB, then there is 2.5 + 0.5 + 9.5 = 12.5BB in the pot before the 4-bet.

This is more or less my thoughts as well. I like it when you have two or more people give their opinion and thought this was a useful addition to the beginning of the book. Also, I've never really played live poker, so it's good to have advice from people who have.

Hi,

I bought this book three days ago. Since i am german and my english ist not so good i have problems understanding some things in the book. So i´m here to ask if there are any plans to translate this book into german?

Greetz from Germany,

macius

P.36
Another point to consider is the 5-bet bluff will usually come from the original 3-bettor, and since he has already invested around 11 big blinds, he’s only risking on average another 89 big blinds (assuming he starts the hand with 100 big blinds) by going all in. Therefore, since the final pot will be 201.5 big blinds, his expectation is just over 2 big blinds back for every 1 percent ofequity.

How did you calculate this?
Because 201.5 / 89 = 2.26

We open 3x AQo UTG and BB calls.
FLOP: AK9r, BB check/calls our 3/4 flop bet.
TURN: 7 - flush draw opens and BB donk bet 1/3 of pot.
His leads on the turn does not make sense, since 7 does not improve significantly his range. Does is means that opponents range is left with hands that crush us?
What is our GTO play when opponent takes non-GTO line (means 0% of time he should) and do something exceptional like at this hand?

201.5 * .01 = just over 2bb

GTO lines never change based on your opponent play, so your question doesn't really make sense. But when your opponent deviates from a GTO line his strategy becomes unbalanced and you can take advantage of this by not using a GTO strategy and take an unbalanced exploitative line.

I dont get it.
another example, suppose our opponent 10x UTG.
1) we can agree that his line is not GTO line since he should take this line with 0% frequency, right?
2) if we know that our opponent does this with only AK we can take exploitative line and play accordingly, but
3) we can take also GTO line and play accordingly to his preflop size, cutting down many speculative hands, right?

No, GTO solution almost certainly requires multiple bet sizes preflop and might even include a 10x some small %. I'd ask a more direct question. River situations are the easiest.

Ehi Matt, can you answer to this post plz?
http://forumserver.twoplustwo.com/sh...&postcount=221

You misunderstand it. What he's saying is that you want hands which do well when your opponent is betting or calling on the river. For example, when you check-raise the flop, if your opponent is never stacking off with TPTK, and only 2-pair+, then it might not be best to raise with an overpair. Instead it would be better to check-raise with a set because you can be confident to be ahead if all the money goes in by the river.


He's saying that when you have a hand like 7h 6h that you're going to fold it on a Ac Tc 2d board because it's signaling to you that it has little chance of outdrawing your opponent because it would likely need to runner runner a two-pair. On the other hand, the same hand on a Ac Tc 5h board plays differently because the hand is signaling that it has two-back door draws. And if it hits one of those backdoor draws on the turn it can continue, and it'll be easy to tell if it does.


It's not that you will really be confused. Think of it this way, when you have a flush draw on the flop and the flush comes in, then you're going to be able to value bet it and will need to balance that with a bluff. On the other hand, when the flush doesn't come in, then you can use the same hand as a bluff.

Let's look at it in terms of value to bluff ratios. Let's say that your flush draw hand has roughly 33% chance of being able to bet it as a value hand (if the flush comes in). So one out of three times, it will be a value hand. Furthermore, let's assume that your value to
bluff ratio should be 1:2, so one value hand for every two bluffs -- ie 33% of your hands should be for value. This is the exact same frequency that your flush draw will be able to be bet as a value hand. Therefore, when you're betting your flush draw, you DON'T need to count it as a bluff because it will be able to be bet as a value hand at a high enough frequency that it makes up for all the times that it'll will be need to be bet as a bluff, when the flush doesn't come in.

At the end of many sections, Matt gives a step by step process on how to think at the tables. But to answer your question, "Why am I betting?" Matt gives a few different reasons throughout the book. Bet for value, bet as a bluff, bet to get high equity hands to fold, bet to charge draws.

Thanks for replying EmptyPromises, and I'll respond as well.

All this is saying is what's better-- A hand which has 20% equity but flops the pure nuts 20% of the time, or a hand which has 30% equity but will only flop bluff catchers? Clearly, it's better to flop the nuts 20% of the time because we can win a massive pot when we do, but if we have a bluff catcher we'll rarely find ourselves in a great situation.


If the flop comes Ac 9d 5s and I have JdTd, it's pretty easy to see I should call (or raise) on a diamond, Q, 8, and probably even J or T turn card. So I'm not going to fold if I get a good turn card, and might outdraw the opponent when he has a pair of aces on the river.

But if I have 44 on this flop, I'm usually going to fold on the turn even if I would have hit a 4 on the river. This makes calling on the flop less effective since the turn card won't "signal" to me that I'm likely to improve on the river, since I can't turn a straight draw, flush draw, or pair with (probably) 5 outs.

I think you're oversimplifying things and you won't feel strongly about labeling a hand as a "value bet" or "bluff" as you get better. There are usually 6 or 7 things I'm asking myself for determining whether or not I want to bet, but after you play probably 1 millionish hands and have spent a few thousand hours talking poker your brain will ask and answer these questions very quickly.

Anyone that has read "Thinking, Fast and Slow" (I'm reading it now and would highly recommend it) will know what I'm talking about.

There will still be a GTO response to an opponent who takes a non-GTO line, but it's something I spend very, very little time thinking about or discussing. If someone donk bets 1/3 of the pot, I'd probably just treat it as slightly stronger than a check and make a note that my opponent does not know what he's doing (assuming I agree that the donk bet is bad from a GTO perspective). We can't conclude what our opponent's range looks like when he takes a non-GTO line, just like we won't know what his range looks like even if he takes a GTO line (as bad players take GTO lines all the time, like opening with aces in the cutoff).

Hi Macius,

I have no idea, you'd have to ask TwoPlusTwo since they published the book (Mason would probably be your best bet for knowing). One of the great things about working with a professional publishing company is all of this stuff is out of my hands, which is great since I'd have no idea what I was doing regarding any translations etc.

FWIW I think anyone who thinks a GTO pre-flop 3-betting range will be either just polarized or just linear is incorrect. In other words, if someone says "I think I should 3-bet a 15% linear range OOP" and they just take the top 15% of hands, that's probably wrong. If they say "I think I should 3-bet a 15% polarized range" and they try to use 3-bet there bet 6% hands, call with a bunch of profitable calls, then 3-bet 9% "bluffs" which aren't strong enough to call then they're doing it wrong.

At the end of the day the question is "How useful is my hand in a raised pot, and how useful is my hand in a 3-bet pot? Is it worth 3-betting this hand to sometime get to play a 3-bet pot despite the fact that I'll often be 4-bet or my opponent will fold? Or should I just call with it since I'm (in the case of the BB) guaranteed to get to see the flop?" This isn't a easy question to answer, especially when you're considering what to do with a hand like K9s in the BB vs a button open.

Thanks for the response, Matt.

To clarify, I was just trying to explain the definition of a polarized and linear range, not comment on the (in)correctness of their use. We have the book for that

50 to 54 or
60 to 54 percent?

Is there a list of errors somewhere? I'd like to go through the book and correct them before reading.

This seems to be the place to go: http://forumserver.twoplustwo.com/sh...33&postcount=1

Perfect. Had read the 1st page when it was originally posted but didn't realise it was being updated.