Book Discussion: Theory of Poker, chapters 11 and 12

Semibluffs!

The semi-bluff - do you understand what makes it a good play in some circumstances (and which these circumstances are)?

Defense against the semi-bluff - difficult!

On semibluffs: I saw a guy check-3-bet the turn with an open-ended straight draw into two opponents yesterday. How's that for a semi-bluff? Is it good or bad? Why?

NOTE: Please do not quote the book. This thread is to help broaden the understanding of the book, not rip off the copyright of it. Feel free to discuss, but try doing so without infringing on David Sklansky's and 2+2 Publishing's intellectual property rights. Thank you.

Defense against the semi-bluff is so hard, because if they have a legitimate hand, you are just donating money to the pot, but just calling can be giving them the odds they need for their draws. Often the correct move is to fold!

The semi-bluff - do you understand what makes it a good play in some circumstances (and which these circumstances are)?

Click to expand...

- When a scare card on the next street will make them fold the best hand.
- When you hit the card you need.
- When they immediately fold.

First of all TOP is probably the best all around book for poker games that have a structured limit although there is some NL stuff, it isn't the focus.

On semibluffs: I saw a guy check-3-bet the turn with an open-ended straight draw into two opponents yesterday. How's that for a semi-bluff? Is it good or bad? Why? This kinda misses the point about semi-bluffing. The situation in quotes is technically a "bluff". I will respond to semi-bluffing later because sem-bluffing is pretty easy to me and I will throw a bone about it later in the post.

The point you suggest really is quite simple. The vast majority of players don't play their strong draws aggressively enough. The rest either fold too often or don't pay attention to ALL the cards on the board, example: paired boards mean get out of the way the majority of the time.

From a math perspective you want to be in there raising your strong draws because you NEED the hand to payoff as much as possible to cover the times you miss or make a mistake. Plus you will be killing other drawing hand's drawing odds and will likely get them out of the way.

From a deception perspective the overwhelming majority of inexperienced poker players don't even consider the possibility that you would be betting a draw. As such you are very likely to get paid off when you hit.

Not a huge poster here anymore but every once in awhile great topics popup and semi-bluffin is one. IMHO once you get past the basics, semi-bluffing is the skill that will make you the most money against passive or inexperienced opponents! Semi-bluffs mean you have a hand just not an extremely strong one and will likely have outs to save you if you if you are wrong. Also large pots are not the time to be pulling a semi-bluff your opponents aren't going to away when there is a lot of money in the middle of the table.

Well hopefully I made sense my head keeps bouncing off the desk maybe I can put some more into this post later today.

I have found that the semibluff is quite an advanced play. more so than i ever thought it could be. i like Joosebuck's statements on the semibluff

- When a scare card on the next street will make them fold the best hand.
- When you hit the card you need.
- When they immediately fold

defense against the semibluff is very hard to do. I understand the raising aspect of the book, yet i've found that it can be a hard play to make.It makes sense that calling is the worst option that we have facing a semibluff. If we just call we have no information on our opponet. It seems to me that pot odds also mandate the semibluff/defense.

It creates a lot more swings and variance than i am used to. I still have a hard time betting when i don't have the best of it. It's an aspect that i am working on.

I made some calculations in another thread. It has some bearing on this discussion:

https://www.cardschat.com/forum/cash-games-11/expected-value-article-fpaulsson-66175/post-392656

Okay, that actually surprised me. Apparently, for the turn semi-bluff raise to be a better move than to just call, we need to trust that the bettor will fold 85% of the time or more. I feel like my math is messed up somewhere, because this is higher than I expected.

In this example would you fold if you were the bettor? I can't imagine you would fold that many times. Could you math be off because of the small pot? If there were more people the pot could get bigger yet you have less of a chance of them folding. I can't see how this play would work out correctly long term. You would need the pot to be correct and aleast 2-3 players paying you off for all the times you miss.

There are more favorable situations than the one I calculated, though. For instance, if it's checked to us instead of bet into us. He signals some strength when he bets, so it makes him less likely to fold.

The semibluff is hugely overdone at the limits I've played, simply because most of the time it's tried, it doesn't actually stand a chance whatsoever to fold everyone out. Then it's not a bluff, it's just a dumb bet. There has to be an actual, reasonable, chance that everyone else will fold, or your semibluff is a worse option than calling.

XD speaks about how it adds deception: This is true. People will often misread what cards you're holding when you raise on the come, but you should also consider what it does to your implied odds: If player A bets, B and C call, and you raise, all of them call, and then you hit your flush on the next card, they will all check to you. Chances are you could have made more money by simply calling the turn and raising them on the river when you hit.

I like to semibluff in headsup pots, where it's highly unlikely that I hold the best hand, e.g. 8-7s in the hole on a queen-high flop. It may be unlikely that my opponent has paired up, but he likely has me outkicked. A semibluff can fold a lot of bad hands - that are better than mine - while at the same time I may have some chance of hitting a straight or a flush on the turn, depending on the draws.

Remember, you're not bluffing if you have the best hand.

Slansky said that his math showed that [heads up] your opponent must fold only 4 out of 19 times on a semibluff to make it a +EV move.

yeah - it gets even better when you're heads up and you bluff randomly... crazy stuff...

joosebuck: That has got to depend on the pot size.

I have a ? about this book theory of poker. Im going to buy it on ebay but i see theres 2 versions of it . The original released in 1989 (i think) and another version released in 2003. Whats the difference???If any. Thanks!

Probably just minor updates. As far as I know, ToP hasn't gone through any major revisions.

can anyone else comfirm that ? I want to make sure I dont miss anything if I get the old version.

nothing about poker has changed since then, i dont see what could be different besides the very end part where he explains each game and has a sort of glossary of poker terms.

thanks alot !

also pick up:

super system 1&2, small stakes poker, and mike caros book of poker tells. dan harrington i think also has a book about tourney play, but i dont have it yet.

In the section "When Not to Semibluff," Sklansky mentions that semibluffing when you're last to act (and no one has yet bet) is generally not a good idea, because you run the risk of being checkraised. Let's look at why:

It's the turn, three players still in, the pot is currently 3 big bets, it's checked to you, and you bet with a flush draw, on the button. This is a bad idea (unless the table is one where the two other people have shown tendencies to fold too often, not a common scenario at lower limits), and there are many reasons why:

1) You're betting as last to act. This is - even by very weak players - seen as a possible attempt to steal. Your bet on the button is less intimidating than you may want it to seem. This lowers the chance of a semibluff to work. This is mostly a sidenote, though, and the real problem lies in:

2) If you're checkraised, you're getting a clear indication that you're going to see a showdown, i.e. you're not going to be able to bluff the checkraiser out of the pot on the river. You now HAVE to hit your flush to win. What's worse, you're only 20% to pull this off, which means that now you've invested 2 big bets in a pot (because you have to call him, the pot is laying you too big odds not to) where you only have 20% equity. You just lost 0.6BB immediately by betting, and the checkraise gives you a chance to lose 0.6BB more (to quote Sklansky, "do you see why?"). Folding, of course, is even worse; now that you've made the pot big enough, you're getting 6-1 on your money, and folding means you forfeit equity worth 0.2*6 = 1.2BB. You lose money by calling and you lose money by folding. Raising is the worst option, of course.

This brings us to another point that is brought up in another book (at this point, I can't remember if it's Small Stakes Hold 'em or Hold 'em Poker for Advanced Players): It's more devastating to get checkraised with outs, than without outs. Let's look at an example of that:

If you hold JJ on a K-T-4-T board. You're on the button, and you have a caller. You bet again on the turn, and now he checkraises you. You can happily lay your hand down. If you're behind here (which you always are, barring bluffs), you have only two outs. You're sacrificing very little equity by folding.

On the other hand, if you had had QJ and gotten checkraised, you would suddenly have found yourself in a dilemma similar to the one above, where you WISH you had checked, but didn't.

So, in conclusion: If you have outs, check behind. If you have no outs, bet. "Having outs" means "several outs", having "no outs" means 3 or less, but it varies.

F Paulsson said:

In the section "When Not to Semibluff," Sklansky mentions that semibluffing when you're last to act (and no one has yet bet) is generally not a good idea, because you run the risk of being checkraised. Let's look at why:

It's the turn, three players still in, the pot is currently 3 big bets, it's checked to you, and you bet with a flush draw, on the button. This is a bad idea (unless the table is one where the two other people have shown tendencies to fold too often, not a common scenario at lower limits), and there are many reasons why:

1) You're betting as last to act. This is - even by very weak players - seen as a possible attempt to steal. Your bet on the button is less intimidating than you may want it to seem. This lowers the chance of a semibluff to work. This is mostly a sidenote, though, and the real problem lies in:

2) If you're checkraised, you're getting a clear indication that you're going to see a showdown, i.e. you're not going to be able to bluff the checkraiser out of the pot on the river. You now HAVE to hit your flush to win. What's worse, you're only 20% to pull this off, which means that now you've invested 2 big bets in a pot (because you have to call him, the pot is laying you too big odds not to) where you only have 20% equity. You just lost 0.6BB immediately by betting, and the checkraise gives you a chance to lose 0.6BB more (to quote Sklansky, "do you see why?"). Folding, of course, is even worse; now that you've made the pot big enough, you're getting 6-1 on your money, and folding means you forfeit equity worth 0.2*6 = 1.2BB. You lose money by calling and you lose money by folding. Raising is the worst option, of course.

This brings us to another point that is brought up in another book (at this point, I can't remember if it's Small Stakes Hold 'em or Hold 'em Poker for Advanced Players): It's more devastating to get checkraised with outs, than without outs. Let's look at an example of that:

If you hold JJ on a K-T-4-T board. You're on the button, and you have a caller. You bet again on the turn, and now he checkraises you. You can happily lay your hand down. If you're behind here (which you always are, barring bluffs), you have only two outs. You're sacrificing very little equity by folding.

On the other hand, if you had had QJ and gotten checkraised, you would suddenly have found yourself in a dilemma similar to the one above, where you WISH you had checked, but didn't.

So, in conclusion: If you have outs, check behind. If you have no outs, bet. "Having outs" means "several outs", having "no outs" means 3 or less, but it varies.

Click to expand...

I understand what Sklansky is saying here, but it seems to fly in the face of the Semibluff as a whole. Not raising because the possiblity of a checkraise you would then never semibluff. You could raise from UTG and get checkraised from the BB in the same situation. would the play then be incorrect also?
This situation seems to be very table/player dependant. Do the players in the game tend to check raise their hands? Or do they just call down? At a table full of weak/passive players I can see the value of a bet here. At a table with aggressive/stong players I can see why we would like to get a free card here.

In Sklansky dollars I see how it works out, but on a whole i think we need to bet here as XD stated
"From a math perspective you want to be in there raising your strong draws because you NEED the hand to payoff as much as possible to cover the times you miss or make a mistake. Plus you will be killing other drawing hand's drawing odds and will likely get them out of the way."

You now HAVE to hit your flush to win.

We knew going into the turn and river against 3 opponets that we needed to make our flush to win. If we don't semibluff this turn, there is no way we take a shot at the river. This seems like our best chance to thin the field/build the pot. Without a bet on the turn we are surrendering our hand, there is a chance with a bet that we could get a couple folds and take a shot at the river against one opponet. Which we could not do if we check and face a river with 3 oppoents.

Xandit said:

I understand what Sklansky is saying here, but it seems to fly in the face of the Semibluff as a whole. Not raising because the possiblity of a checkraise you would then never semibluff. You could raise from UTG and get checkraised from the BB in the same situation. would the play then be incorrect also?

Click to expand...

Not quite contradictory; the very important distinction is that since you were in position you decided to rob yourself of a free card by betting. You could have seen the river (in this case) for free, but you instead tried taking it down right then and there. If you're not last to act, the situation changes, because if you bet, everyone else may fold, but if you don't bet, someone else still might.

Xandit said:

We knew going into the turn and river against 3 opponets that we needed to make our flush to win. If we don't semibluff this turn, there is no way we take a shot at the river. This seems like our best chance to thin the field/build the pot. Without a bet on the turn we are surrendering our hand, there is a chance with a bet that we could get a couple folds and take a shot at the river against one opponet. Which we could not do if we check and face a river with 3 oppoents.

Click to expand...

Good point; Our bluffing prospect in the face of checking with 3 opponents still in, in position, on the turn, makes bluffing the river highly unlikely. I'm not sure where I was going with that, but you're definitely correct.

My overall point still stands: Semibluffing in position with plenty of outs is not a good idea, if you run the risk of being checkraised. Another reason it's bad is because with plenty of outs, you can actually gain some value with it if you get many callers (e.g. flushdraw on the flop in Hold 'em - two or more callers, and you're showing an equity surplus), but if the first player after you raises (or you get checkraised when on the button) that checkraise may drive others out and again you've managed to get plenty of money in with the worst of it. This is actually a fairly common mistake that I saw as recently as 20 hands ago on a $3/$6 6-max table. It looked something like this:

Completely loose (calls everything - yes, everything - preflop, for one bet) player calls UTG. I had ATo in middle position and raise. Semi-loose player calls from the BB, meaning that his range is fairly big. Not 72o, but big. The flop comes A-6-4, where 6-4 are diamonds, and both BB and UTG check to me.

I bet. BB checkraises, loose idiot folds. I call.

Turn comes another 6. BB checks. I don't like the look of middle pair, but my kicker is still alive, and this guy hadn't been CR:ing middle pair that I had seen earlier, and had been pretty good about betting his big hands, so I decided he either had a weaker ace and was testing me, or he was on a flushdraw and tried to take control. I bet. He called.

River is another 6. He donkbets. I call. He has a busted flushdraw, and decided to give the river a go.

His biggest mistake here was checkraising the flop. If he had bet out, the loose maniac might have called. If he had checked, and then called my bet, the loose maniac might have called. Both scenarios would have meant that he'd have gotten money in with an equity surplus (I'd like a better way of putting that, but I've come up short so far), but now he decided to scare the other player off. Not smart, if he wanted to play his flushdraw. It's possible he was running a more advanced semibluff, but in that case I think waiting until the turn to pop me would have been smarter. But oh well.

the reason you check the flop in that situation @ the button is because you have the option of giving yourself the free card. it's like checking the river with a moderate hand that you might be a winner with. if you check, you robbed him of a CR. if you bet, the only hand that will play back at youi is one that will beat you.

F Paulsson said:

Not quite contradictory; the very important distinction is that since you were in position you decided to rob yourself of a free card by betting. You could have seen the river (in this case) for free, but you instead tried taking it down right then and there. If you're not last to act, the situation changes, because if you bet, everyone else may fold, but if you don't bet, someone else still might.

His biggest mistake here was checkraising the flop. If he had bet out, the loose maniac might have called. If he had checked, and then called my bet, the loose maniac might have called. Both scenarios would have meant that he'd have gotten money in with an equity surplus (I'd like a better way of putting that, but I've come up short so far), but now he decided to scare the other player off. Not smart, if he wanted to play his flushdraw. It's possible he was running a more advanced semibluff, but in that case I think waiting until the turn to pop me would have been smarter. But oh well.

Click to expand...

I see what you mean now, thanks for the clairifcation, since in this example we have so many outs to make our hand and possibly make a second best hand for another player, taking the free card then raising/reraising the river if we fill up with the flush will pay off.
By not taking the free card we are decreasing our implied odds (by getting others to fold to the bet/checkraise), It still seems to be table depenedant. in your example the loose idiot could of still called 2 bets on the flop and a turn bet. we've all see this done. So while his play was faulty,one could see his line of thinking...

To be honest, I've noticed that for every level I climb in stakes, the more table (and person) dependent all of the tactics become. Including preflop play. With a hand like ATs on the button, I can either fold, call or raise, all depending on who else has called or raised already. And not just whether or not someone has, but who.

Properly making these adjustments depending on table and opponents - which is something I've started seriously doing only recently - makes such a huge difference in winrate that I've completely stopped multitabling. I make almost as much money playing a single table as I could playing three - where if I play three, I wouldn't be as picky about table selection, I wouldn't pick up specific reads, and I wouldn't be able to use the information properly - and I have a lot more fun doing it.

Anyway, I'm getting off topic now.

adjusting from player to play is extremely important. ill make my money off of only 1 or 2 people at the table most of the time, so if theyre extremely tight, ill bluff a lot more, and wont give them any action. but to the loose aggressive ill snap his bluffs off with a more moderate hand.