Some top players dislike NLHTAP: Why?

[Albert Moulton]

[ QUOTE ]
I've heard that Brian Townsend don't think highly of the book, why is that exactly? It's one of my favorite books and I respect Townsend a lot.

[/ QUOTE ]

I play live, 9-handed NL 5/5 1000 max buy in. I'm rereading this book for my 3rd time. And applying just about every bit of advice I can manage in that book I've gone from a break-even player to a $23/hour player. So, I'll obviously keep my day job. But I know from first hand experience that it is a solid book. I think it's as good as any, and better than most.

People accusing the book of being "cook book" didn't read it very closely. The examples are included to illustrate a concept, not to promote a certain play or style of play in every similar situation. Without the illustrations, the concepts would be very difficult to understand. And in most cases, the illustrations include EV calculations to explain why a particular move in particular circumstances would be better than others. The "all-in with the nuts" calculations is one example. That section is nothing but a math problem. It doesn't say, "Always go all in with the nuts." It says that you should consider going all-in with the nuts because it will often have a higher expected value than milking the nuts with small bets that might get called more frequently. And all throughout the book's examples, the authors include multiple caveats to "mix up" or "randomize" your play vs. observant opponents.

The overwhelming theme of NLHETAP seems to be that NLHE is a game of implied odds in which you seek to manipulate your opponents into making big mistakes in big pots while seeking to avoid being manipulated into doing the same yourself. The tools for that manipulation are laid out fairly well: calculating pot odds and implied odd; basic hand reading and EV estimation; adjusting for position, both relative and absolute; bet sizing both pre- and post- flop; pot control; deception and multiple level thinking; the concept of trading small mistakes early in a hand for an opponent's potential big mistakes later in the hand; occasionally risking free cards to manage pot size and induce a bluff; etc.

Many of the criticisms I've seen, like "Who would ever limp on the button" are missing the point of the "limp on the button" section. If you have a good drawing hand, deep stacks, and week players in the blinds who will stack off with top-two pair for 500bb if you have J9s, and flop a Q8Tr straight vs. BB's QTo top-two, then you would be silly to raise 4bb pre-flop and fold out the guy with QTo. Instead, you should let the QTo player have enough rope to hang himself. Now, if you are playing a TAG/LAG 6-max table with 100bb or less effective stacks and only 5 guys at the table, all of whom are never going to stack off with Q8o from the BB, then obviously this particular tactic in this one example doesn't apply to your game. But the larger concept of playing for implied odds and creating situations in which weaker players can hang themselves OOP is still relevant.

I also think the pre-flop bet sizing section is particularly good. Stack sizes, position, starting hands, and whether you want to play a big pot or a small pot should dictate whether and how much you raise. Blindly raising 4xbb+1/limper is less useful than raising big when you want to play with few players in a big pot (for whatever reason), or small if you want to play with many players in a medium pot (for whatever reason), etc.

The principles in the all the examples are true in all games: play for maximizing your EV. Sometimes that means playing for small pots. Sometimes that means playing for big pots. Sometimes you bluff. Sometimes you slow play. Often that means trading small mistakes when the pot is small for your opponent’s big mistakes when the pot is big. Sometimes that means folding what is most likely the best hand because bad position + marginal hand = reverse implied odds. Etc. Etc. Etc.

No other book packs as much relevant and useful information about how to think about NL play than this one. It is not cook book. And those who think it is didn’t read it very closely.

[Albert Moulton]

[ QUOTE ]
I don't identify with the pre-flop raising chapter at all. There is nothing wrong with raising similar amounts with various hands, because you do want action, it has a higher expectation than taking the blind because you "defined" your hand with a 20x bb raise. A pot sweetener bet can be inviting disaster, especially at a small stakes full ring game.


[/ QUOTE ]

Both NLHETAP as well as PNLHE advocate raising different amounts to achieve different objectives from different positions and different hands vs different opponents.

Standardizing your raises is less useful than raising a wide variety of different amounts to manipulate the pot to be the size you want it with the number of players you want in it.

In short, I think you would make more money mixing up your pf raise sizes than always making them the same size artificially. And pot-sweetners with good hands OOP are only disasters if you can't find the fold button when you know (or at least should know) that you're beaten.

[ QUOTE ]
There is little need for deception in poker, attempting to randomize your play at the beginning, you will be opening yourself to mistakes. The biggest problem with the book is that you would be attempting to play randomly, but in a total vacuum. There isn't enough consideration to how your opponent plays, and how they react to your bet.


[/ QUOTE ]

This really isn't true at all, especially the bold part. It just isn't. Maybe you play with enough random, on-line fish where you can bet good hands fast and get called by worse ones, but most tough games have good players who won't give up much money unless you mix up your play enough to trick them into making big mistakes in big pots. But you need deception in your game to do that.

Moreover, NLHETAP is very careful to consider what range of hands your opponent might have, and what range of hands your opponent might think you have.

And I bet you will occasionally open raise with a hand like 89s. That is a raise for deception, not for value. So, you have deception mixed into your game already.

[ QUOTE ]
I have AJ...

[/ QUOTE ]

In many games you should fold this preflop. In others, this is a reraising hand or even a push. It depends on a lot of factors, and your hand with your factors aren't specifically covered in the examples, nor were they meant to be.

No book will tell you every thing about every thing.

Moreover, the section you quoted about drawing is not the only relevant section of the book. There is also a section that vs. aggressive opponents when you have a 1-pair hand with deep stacks that sometimes it's better to risk a free card for pot control and to induce a bluff than bet and risk getting raised or c/r'ed. And there are other relevant sections, as well. Once again, each section illustrates one concept. In a real hand, you may have to consider several concepts and alternatives before finding a play that you conser to be max-EV. Don't blindly say, "Opponent drawing = me c/r'ing." That wasn't what the authors meant. That's not how readers should interpret it.

[daveT]

I enjoy your thoughts, but you are misinterpreting my post.

In general, you should be playing straight-forward, boring poker, especially when you first start. Open limping with Aces on the button is a flimsy attempt of deception. I don't discount the value of bluffing and bet merging. The second is not so much deception as it is a quality of betting several similar hands with similar values. If that is deception, then yes, I am guilty. Bluffing and game theory was covered in TOP. I don't remember if NLTP even mentions GT and Bluffing.

I did not talk about my position or table conditions. I did not say how much I raised, how many players where in the pot when I raised, blah blah blah. If you are talking about "trading mistakes," this hand illustrates. Against certain players I may raise, but I have the discipline to fold when he bets. I would know better than bet. I said in my example that the person drawing is check-raising, that is a play that is becoming so cliche it is down right irritating. I play tons of hands a day, because I play heads up poker. I can attest that the over all story is not based on deception. I play uber-LAG, and I am telling you that I am not deceptive.

Yes, there are good sections in the book, and I am sure you can find great advice even in "Play Poker Like the Pros." Among the sections that are good is the multi-level thinking. The weak-tight section had me laughing so hard I almost fell out of my seat. I simply wish that the book focused more on "playing your opponents hand," not in the sense of deception, but by configuring an optimal strategy for facing an implied range of hands. It needed to focus on level 2, i.e. what does my opponent think I have. I can't imagine playing poker without this consideration.

[Albert Moulton]

[ QUOTE ]
I play heads up poker

[/ QUOTE ]

lol, I don't know much about heads up cash games. I suspect the fundamental principles are all the same, but many of the tactics in NLHETAP probably don't apply since they are mostly written for full ring games with generally good FR cash players.


[ QUOTE ]
I don't remember if NLTP even mentions GT and Bluffing.


[/ QUOTE ]

GT is mentioned briefly, but not in detail since the authors felt that was covered in TOP. However, "Concept No. 58: Any strategy relatively close to a game theoretical strategy is at least almost as good as the optimal strategy, and sometimes it's better," has an excellent, short discussion of just how much "mixing up" you need to do in order to avoid being exploitably easy to read.

As for bluffing, the "Hammer of Future Bets" section, and the importance of timing a bluff to reserve the credible threat of big bets on later streets (particularly the turn and river) has improved my in-position bluffs considerably against weak-tight opponents and good LAGs who will go 1/3 their stack w/ TPWK, but not the whole stack OOP. That whole bluff timing and planning section was the best I've seen on that topic.

[ QUOTE ]
It needed to focus on level 2, i.e. what does my opponent think I have. I can't imagine playing poker without this consideration.

[/ QUOTE ]

There is an excellent section on this topic, p. 168-175, "Multiple Levels of Thinking." Basically, it explains that you need to think one level ahead of your opponent. That seems pretty obvious, but I hadn't thought of it until I read it. For example, what good does it do to act on what I think he thinks I have (2nd level), if he only plays his cards (0 level) and doesn't think about what I have (1st level). In that instance, thinking on the 1st level is optimal, and 2nd level thinking will give you problems. Against a 1st level thinker, you need to think on the 2nd level. And so on. It's a really interesting concept that seems well explained.

Furthermore, many of the examples that depend on 2nd level thinking include that in their analysis. For example, in "Concept No. 6: It can be right to call with decent hands that have little chance of improving even if you plan to fold if there is a bet on the next round," there is an example in which the authors ask, "What does your oponent put you on? You called twice ..." In that case, you need to think about what you think your opponent thinks you have. Now, that isn't in every example, because the tools illustrated in each example don't need it. But they are there when needed to explain the concept in the example.

I really think that you might be undervaluing the book because the interpersonal dynamics and subsequent choice of tactics in a heads up cash game are so much different than the interpersonal dynamics and subsequent choice of tactics described in NLHETAP's full ring examples.

I do know that at live, medium stakes, NLHE this book rocks.

[fraac]

[ QUOTE ]
There is an excellent section on this topic, p. 168-175, "Multiple Levels of Thinking." Basically, it explains that you need to think one level ahead of your opponent. Which seems pretty obvious, but I hadn't thought of it until I read it. For example, what good does it do to act on what I think he thinks I have (2nd level), if he only plays his cards (0 level) and doesn't think about what I have (1st level). In that instance, thinking on the 1st level is optimal, and 2nd level thinking will give you problems. Against a 1st level thinker, you need to think on the 2nd level. And so on. It's a really interesting concept that seems well explained.

[/ QUOTE ]
I'll say it until you're all dead, Sklansky knows nothing about levels. There are two: game on and game off*. Making the right moves at the right time is purely about multidimensional pattern modelling, not the pseudo-rigourous Sklanskian logic of 'levels' ffs.

Who actually thinks "What does he think I'm thinking?"

I bet Sklansky doesn't. He couldn't explain the process he really uses, so he rationalised a load of 'logical' crap. I'm certain of it.


* Truly enlightened ones see only 1 level, of course. Maybe less. I don't know. They won't tell me.

[Gelford]

[ QUOTE ]
[ QUOTE ]
You know with two 'weak' players in the blinds, if you look down on J9s, then you should prefer to just limp on the btn, in order to give the blinds the chance to make expensive postflop mistakes

[/ QUOTE ]

This is something that I am not sure if you are being sarcastic about.



[/ QUOTE ]


I'm not sure if I being sarcastic or not either. I is just an illustration of the general wierdness of the book.

A 'weak' player, where you prefer to limp and see a flop instead of raising. So basically you have some dude, that will fold pf if you raise with your J9s (which is obv less EV than playing a flop with him (them))

It is hard to envisage such a player, and simply the description 'weak' as all there is to guide us does not cut it. Is it weak as in someone who folds too much, but if it is, then why does he suddenly start to commit too much post? (or do you want to let him hit something of medium strengt and then bluff him with a big bet on the river, since he is laying every thing down when faced with really big bets, ei. the the weak-tight chapter ...)

Or is it weak as in fish, but those guys call with a far to big a range both pf and post flop, so you prefer to pad the pot pf, so that if they call post, their mistakes will be even bigger

And finally, what is it exactly that J9s is supposed to flop, that will allow us 'punish' this weak player, most likely is some sort of (combo)draw, which is fine is you are planning around stealing, as that is what gives your strategy an extra bit of punch, your pounding away and flopping draws gives you extra leverage.

But here, where the previous section clearly states that planning around stealing is not really desired in NL ??

I am not saying Sklansky is clueless .... but his book at times seems so or maybe not clueless, but very confused, it lacks the rigor of an expert player teaching, the organisation is just shot from page 100 and onwards.


Luckily Flynn and Metha seem to have that rigor, so in the end 2+2 still brings home the bacon [img]/images/graemlins/smile.gif[/img]

[Albert Moulton]

[ QUOTE ]
[ QUOTE ]
[ QUOTE ]
You know with two 'weak' players in the blinds, if you look down on J9s, then you should prefer to just limp on the btn, in order to give the blinds the chance to make expensive postflop mistakes

[/ QUOTE ]

This is something that I am not sure if you are being sarcastic about.



[/ QUOTE ]


I'm not sure if I being sarcastic or not either. I is just an illustration of the general wierdness of the book.

A 'weak' player, where you prefer to limp and see a flop instead of raising. So basically you have some dude, that will fold pf if you raise with your J9s (which is obv less EV than playing a flop with him (them))

It is hard to envisage such a player, and simply the description 'weak' as all there is to guide us does not cut it. Is it weak as in someone who folds too much, but if it is, then why does he suddenly start to commit too much post? (or do you want to let him hit something of medium strengt and then bluff him with a big bet on the river, since he is laying every thing down when faced with really big bets, ei. the the weak-tight chapter ...)

Or is it weak as in fish, but those guys call with a far to big a range both pf and post flop, so you prefer to pad the pot pf, so that if they call post, their mistakes will be even bigger

And finally, what is it exactly that J9s is supposed to flop, that will allow us 'punish' this weak player, most likely is some sort of (combo)draw, which is fine is you are planning around stealing, as that is what gives your strategy an extra bit of punch, your pounding away and flopping draws gives you extra leverage.

But here, where the previous section clearly states that planning around stealing is not really desired in NL ??



[/ QUOTE ]

The point he is making is that implied odds are more important than pot odds. If you raise preflop with J9s in the example, then you cut down your implied odds whether you get called or not.

For example, If you have a good drawing hand, deep stacks, and players in the blinds who will stack off for 200bb if you have J9s, and flop a Q8Tr straight vs. BB's QTo top-two, then you would be silly to raise 4bb pre-flop and fold out the guy with QTo. And yes, there are players in many of the live games I've played in who would fold QTo to a raise, but call/bet all-in with top two-pair on a Q8T flop.

So, in this instance, the book recommends that you should let the QTo player have enough rope to hang himself.

Now, if you are playing a TAG/LAG 6-max table with 100bb or less effective stacks and only 5 guys at the table, all of whom are never going to stack off with Q8o from the BB, then obviously this particular tactic in this one example doesn't apply to your game. But the larger concept of playing for implied odds and creating situations in which weaker players can hang themselves OOP is still relevant.

[Albert Moulton]

[ QUOTE ]
Who actually thinks "What does he think I'm thinking?"

[/ QUOTE ]

I actually heard a good regular at a live 5/5 NL game last Wednesday consider calling an opponent's all-in say to himself, "The question isn't really what you have so much as what do you think that I have..." Then he thought, and folded.

So, he does. And I do.

Asking myself questions at the table about "multidimensional pattern modelling" doesn't help me nearly as much when considering an action in response to a possible bluff by a very good regular as asking myself, "What does he think I put him on, and would he bluff here more often as a result of that belief..." (3rd level).

Nor does pondering the phrase "multidimensional pattern modelling" help me when considering making a bluff vs a good, but straight forward player nearly as much as asking myself, "What would he put me on if I push here, and how much of the range I think he has would he fold based on his read of what I probably have..." (2nd level).

The "levels" thing just makes it easy to consider the right questions. Skill and experience vs a particular opponent will tell you how many of those questions to ask.

[eMbAh]

[ QUOTE ]
[ QUOTE ]
There is an excellent section on this topic, p. 168-175, "Multiple Levels of Thinking." Basically, it explains that you need to think one level ahead of your opponent. Which seems pretty obvious, but I hadn't thought of it until I read it. For example, what good does it do to act on what I think he thinks I have (2nd level), if he only plays his cards (0 level) and doesn't think about what I have (1st level). In that instance, thinking on the 1st level is optimal, and 2nd level thinking will give you problems. Against a 1st level thinker, you need to think on the 2nd level. And so on. It's a really interesting concept that seems well explained.

[/ QUOTE ]
I'll say it until you're all dead, Sklansky knows nothing about levels. There are two: game on and game off*. Making the right moves at the right time is purely about multidimensional pattern modelling, not the pseudo-rigourous Sklanskian logic of 'levels' ffs.

Who actually thinks "What does he think I'm thinking?"

I bet Sklansky doesn't. He couldn't explain the process he really uses, so he rationalised a load of 'logical' crap. I'm certain of it.


* Truly enlightened ones see only 1 level, of course. Maybe less. I don't know. They won't tell me.

[/ QUOTE ]

How do you come to this conclusion?

[fraac]

[ QUOTE ]
Asking myself questions at the table about "multidimensional pattern modelling"

[/ QUOTE ]
I should have clarified: you don't ask questions about it, you just do it. Whether you want to or not. So you may as well pay attention and get good at it. Maybe 'levels' will help train you. I think they'll only slow you down and should only be used in descriptions after the fact, or when relating to David Sklansky.