Posted Sun Jun 19, 2005 2:26 pm GMT by snoogins47
Laggy Larry says that if you're playing heads up limit hold'em and your opponent does not defend/3bet his BB frequently enough, and assuming he will not alter his strategy, you should raise from the SB 100% of the time.
Is Larry's statement true, or false, and why?
Did you know that participating in a poker forum can help you improve your own game? Be it by sharing experiences or simply asking for help, participation in a forum helps you focus and keep 'on topic' which will help you improve your game. You can learn from other players feedback and from their experiences. Why the THP poker forums? We offer one of the best managed texas holdem poker forums available, and the community within is far more friendly than those typicaly found on other sites. We've made a 'lurkers edition' of the poker forum available here on Holdem Poker Online, but we encourage all visitors to register and join in on the conversations on TexasHoldem-Poker.com
Posted Sun Jun 19, 2005 2:50 pm GMT by suitedaces84
If I'm investing $3 to steal a $2 blind, and planned on check/folding when met with resistance I'd have to pick the blind up 60% of the time. Of course, I'm not going to check/fold everytime I'm met with resistance. I will often get a hand and profit more as a result of my unsuccessful blind steal. Dammit, I guess this can't be broken down into simple numbers.
But I'm going to say you shouldn't raise everytime. If he folded the "correct" amount it would be "correct" to raise with a certain range of hands. If he folds less frequently than he should it will be correct to raise with a smaller range of hands. So, if he folds too often it would be correct to raise with a greater range of hands. However, unless he folds way too much this range will not include all hands.
That's my stab at it.
Posted Sun Jun 19, 2005 5:31 pm GMT by zeroswarm
I don't play limit but my guess would be that you should raise him every time. If he folds say 30% of the time then that would add up to a considerable "stolen" sum. When he calls you could still have a better hand and if not u might hit the flop anyway.
Therefore I would say the statement is true. 
Posted Sun Jun 19, 2005 6:29 pm GMT by Iron Butt
I don't get it. By "100% of the time" do you mean 100% of your steal opportunities? Is this heads up or short-handed? Is this a particularly weak table? Because just as it's stated and making reasonable assumptions of a normal strength ten player table I don't see any case for your proposition.
84 about covered it... to put it another way you're asking whether you should pay an extra bet and a half every round to play a random hand in the SB. I don't think anyone would like the sound of that, even given that the player on your left is weak in the BB. In fact I don't think I like it even if we further state that the BB is going to fold every single time. What about the other players at the table? This guy may be putting one extra bet per round in the pot, but you are not going to be the only one contesting for it, and you're going to be at a serious disadvantage risking a bet and a half to try to win one bet with random hands in lousy position. By maybe 3, 4 rounds tops everyone's going to know what's going on, it's going to be at least a raise by the time it gets to you (which it would be often anyway), and there you are making it 2 or 3 bets with your random hand LOL.
So if the player to your left is weak in the BB should you lean on him? Sure. I don't see any possible case to "raise from the SB 100% of the time" though.
Posted Sun Jun 19, 2005 10:17 pm GMT by snoogins47
Correction: I meant to type the word "heads up." I guess specifying that we were only concerned with "your opponent," singular, could lead to that conclusion, but I didn't specify. This is a heads up game. My apologies.
| Quote: | | "However, unless he folds way too much this range will not include all hands." |
There is something I wanted to specify but purposefully left out of the first post, for a number of reasons. Firstly, by his defending too infrequently, I basically meant that he is doing nothing to thwart your steals. That is (if we just say it's a $1/$2 game, we're investing $3 to win $3 immediately), if we pretended he never 3bets you, or that you folded to every 3bet, he is folding >50% of the time.
If we assumed it was from his perspective, "defending too infrequently" would rely completely on your range of hands, which our opponent in the initial post is ignoring.
Short version: He's folding enough so that you'd make an immediate profit by raising without your even having cards. Therefore, in this situation, raising any two cards will always show a profit. Zero sort of led us to this step.
So we have established that a raise will always show an immediate profit. That's part 1. Now, keep going.
Posted Sun Jun 19, 2005 10:23 pm GMT by suitedaces84
Are we talking about folding vs raising or calling vs raising?
Posted Sun Jun 19, 2005 10:25 pm GMT by snoogins47
| suitedaces84 wrote: | | Are we talking about folding vs raising or calling vs raising? |
We're just thinking about his contention that you should be raising every hand from the SB.
Posted Sun Jun 19, 2005 11:03 pm GMT by suitedaces84
Are we talking about a BB who folds KK because you may have AA or a BB who folds Q9s?
When playing the guy who folds KK it's obvious that one should raise everytime. Vs the other guy the line is a little harder to draw. You have to consider how frequently he will call you and how far behind you will be. You also have to consider what kind of player he is post flop, and what effect your raise will have on the hand. But I'll assume that you both check it on all three streets.
I'll start with a hand 23o where I'm likely a 3:1 dog (or there abouts). If I can pick this pot up preflop as opposed folding my EV will be $0 (the blinds are 1/2 and I'll win a $4 pot one in four times and it will cost me $1 to do so). If I raise and there is an X probability that he calls me my EV will be X*8*.25+8*(1-X) - 3 = -6X + 5. So in order for a raise to be better than a call -6X + 5 must be greater than 0, or X must be less than .83. If this guy is defending his blinds 83% of the time he is certainly not folding too much. But this does not take into account post flop play.
BTW: pokeredge told me I fold my SB to a steal 82% of the time, and fold my BB to a steal 50% of the time. Is this about right?
Posted Mon Jun 20, 2005 1:50 pm GMT by snoogins47
I was hoping this post would have more buzz around it. Oh well.
Thanks for the discourse everybody who participated, but I'm just going to come out and explain the reasons for this discussion flat out.
Larry is wrong. Here's why (and there's shockingly little to no math that I'm going to use here.)
It is true that a raise with any two cards will show an immediate profit, if you never win the pot when he calls. Which is NOT the case, so raising becomes even more of a tempting play.
However, that doesn't mean you should do it. This popped into my head after listening to a hand recounted to me.
The player inquiring was playing a smaller NL game, flopped an open-end straight flush draw, and pushed all in on the flop, which was a very significant amount more than the pot.
"I figured it was profitable."
I'm fairly positive that his statement was true, and I'm 100% convinced he played the hand incorrectly.
We all have a tendency, to apply calculations only when it is convenient, and it is often very hard to perform an adequate and complete analysis while actually taking part in a hand. It seems only natural to abuse and misuse the mathematical concepts that we learn, and it can easily lead to leaks in our game.
How is this abusing the mathematics? Sure a push all in was profitable. Shoving all in with AA preflop into a small pot carries a positive expectation too. That doesn't mean you should be doing it.
One of the worst leaks I consistently see from "decent" players is related to this problem. Once we figure out how to evaluate plays mathematically, the course of action many take is to look at one play, and if it carries a positive expectation, to take it. I'm sure at least some of this stems from the omnipresent cliche about pushing small edges.
Pushing small edges is good, but your goal should not be to push every edge you have, but to give yourself the HIGHEST possible expectation at all times. Sometimes this number is less than 0.
For instance, if we're not getting the odds on a draw, our call might cost us, in general, .25Big Bets. Folding would carry an expectation of 0. It is very apparent that folding is vastly superior to calling.
However, when the expectation becomes the difference between +1.5 Big Bets, and +1.75 Big Bets, it is easy to overlook the details. But there is hardly a difference between the two.
I can make a comparison to Blackjack pretty well. For instance, say you're dealt
8 8
and the dealer is showing a Ten.
Common wisdom says split, correct?
But splitting carries a negative expectation of 0.4803. Splitting 8s against a Ten is a money losing play.
So, the obvious answer would be to hit, like you would on any other 16 against a Ten.
But hitting carries a negative expectation of 0.5398.
Another money losing play. Splitting is the lesser of two evils, and thus, optimum.
How does this apply to the poker example?
Well, the reason Larry is wrong is that while raising every hand shows an immediate profit, there are many factors that are not taken into account. I will illustrate with a very extreme example.
Pretend Larry's opponent raises and caps every street when he has no pair, and just calls when he has one pair or better.
It's apparent that a lot of the time, you would want to play to the flop with this opponent, because you will be exploiting his weaknesses to a much larger degree.
On the flip side, if your opponent is a excellent post flop player and truly out-classes you, the scale tips way toward the "raise every time" side of things.
There also may come a point where if your opponent folds preflop too often, but plays somewhat reasonably post flop, limping may carry a higher expectation than raising, when you hold powerful hands. This doesn't just apply to heads up limit. One other illustration of this point comes in a NL game when you are on the small blind. It can often theoretically correct to push with trash hands, while merely completing with better hands. This is because the pure steal equity you get carries a positive expectation, but with a better hand, taking a cheap flop may carry a higher expectation than pushing.
The bottom line: Don't try to determine if a play is profitable. Try to find the play that wins the most, or in some cases, loses the least, and play accordingly. That's where the money comes from in this game.
Posted Mon Jun 20, 2005 1:58 pm GMT by Dave B
Raise every time
Posted Mon Jun 20, 2005 3:11 pm GMT by Iron Butt
OK, thought there was something like that. Who plays limit heads up LOL? That didn't occur to me. There's that banker in Texas, plays like 20K/40K limit heads up against Doyle's crew... OMG you are moving up aren't you snoog? :D
OK so now that I'm up to speed, I disagree with you, raise him up in the BB every time - while it works. Why? I guess I disagree with your position (if I'm "getting it" LOL) that this would be a trivial leak and you would make more by seeing flops etc. I mean, that may very well be true, but if the guy's leaking big post-flop too just take the money, what's to analyze?
Let's say that you're playing 20/40, $1000 each on the table. You have determined that your man folds 3 of 5 times you raise his BB. Let's say that you're playing roughly even otherwise, because he must be offering some resistance or there would be no need for analysis. So raising every BB, taking the BB 3 of 5 times, and otherwise breaking even: You're making $6 a hand and will bust him in about 170 hands. That seems like a pretty good result to me... what, 2, maybe 3 hours heads up? 9 big bets an hour at 60 hands an hour? Maybe you're used to doing better than that  but as I understand it that's a huge win rate.
In fact that seems like a pretty scary weapon there. If you let up every so often he might not even notice until it's too late.
Posted Mon Jun 20, 2005 3:34 pm GMT by suitedaces84
No offense, but it appears that you've missed the point of the post entirely. The idea is not to turn any profit, the idea is to turn a profit that is as great as possible. If he folds preflop too much you would make more on average by limping with certain hands. If he raises with AA he will likely pick up the blinds, okay, however by limping he will make more on average.
Like Snoogins pointed out, I could go all-in preflop with AA one million times. I would turn a profit one those million hands without a doubt. Does that mean it's the optimum way to play AA? Of course not. I would make much more if I put in a preflop raise and went from there.
Posted Mon Jun 20, 2005 4:50 pm GMT by Iron Butt
None taken. But again re profit, 9BB/hr is not a trivial leak, and if you can do better against this guy, where's the need for analysis?
But anyway so if your point and the point of this thread is that you shouldn't do anything in poker 100% of the time, even make a certain +EV decision the same each time, because for instance there will be certain situations where there will be an even higher EV option, hey, I totally agree. In fact, and no offense intended here, I thought that was such a basic principle that it couldn't be what this was about. As I touched on earlier I think it's obvious that you wouldn't actually in a real poker game as opposed to a theoretical discussion want to raise anyone's blind 100% of the time; sure, there may occasionally be a higher EV play but far worse eventually your opponent(s) will play back and then you're playing random hands for too much money.
What I thought this might be about is the whole blind stealing issue; how big a problem is it, how to do it and defend it effectively. Seems like it could be a much bigger weapon/problem than I had thought. I'll have to start a thread on that.
Posted Tue Jun 21, 2005 12:00 pm GMT by red_pen
| Quote: | Laggy Larry says that if you're playing heads up limit hold'em and your opponent does not defend/3bet his BB frequently enough, and assuming he will not alter his strategy, you should raise from the SB 100% of the time.
Is Larry's statement true, or false, and why?
|
Got me thinking this post did. Got me thinking lots. And here is my take on it:
Larry's statement is 100% correct. Not only that, Larry's statement is 100% correct regardless of how often the BB folds, as long as he folds sometimes, and as long as does not adjust his folding strategy.
Only one thing really matters here and that is the pot equity the BB can claim to have. There is not one hand the BB can hold for which he is getting less than about 17% (1 in 6) of the pot. I'll admit there may be plenty of BB / SB combinations which would justify a fold to a raise, but there are no justified folds the BB makes to any raise on the basis of his cards alone. So he must call. It also helps that he's the last to act and knows how much he must stake to see the flop.
Everything stated about outplaying the BB later on the basis that post flop trickery is worth more than getting more chips in the hand pre-flop when you've got the best of it, is perhaps misplaced at limit poker when there are already 4 small bets in the pot. There maybe a hard distinction to draw between the two strategies when the BB folds to 5% of raises, but at 30-50% levels the distinction is clear cut.
i've gotta leave work now but I'd willing to expand on my point of view if anyone is interested.
Posted Tue Jun 21, 2005 1:04 pm GMT by red_pen
| Quote: | | Only one thing really matters here and that is the pot equity the BB can claim to have. There is not one hand the BB can hold for which he is getting less than about 17% (1 in 6) of the pot. |
This maths is wrong - where it came from god only knows.
It should read 25% (2 in 8 ). It should match the size of the pot and the size of the call.
It's been a long day.
Posted Tue Jun 21, 2005 3:42 pm GMT by NickFlynn
Ok, lemme crack on this one.
Let's assume Larry raises every single hand when he is on the button. Passive Paul is facing a $2 call with $6 in the pot, so he should be calling at least 75% of the time.
Let's say he only calls 50% of the time. How big is the leak? He's losing $2 once every 8 hands. That's 6.25 BB/100 hands.
Now let's look at Snoggin's idea that you might do better EV-wise if you got to the flop more often against Passive Paul, because he'll lose more money there.
What if you only raise half the time? You've lost 3.125 BB/100 hands of pre-flop equity, but you've got Passive Paul playing another 25% of the flops. It seems to me (and this is where I am probably making a big mistake) you need to beat him by 12.5 BB / 100 hands after the flop to make up what you've given up pre-flop. Seems like a reasonably tall order, but if you can do it, then that's the way to go.
- Nick
Posted Tue Jun 21, 2005 3:54 pm GMT by NickFlynn
Maybe I got my numbers wrong, but now that I look at that again, I'm not so sure Snoggins is right about this. 12.5 BB/100h is quite a lot - I'm not sure anyone plays that badly after the flop. Even if you bias the whole deal, by raising with your crap hands and only calling with the good ones, so you have an advantage going to the flop, I'm not sure you are going to find the right opponent to make this work very often.
- Nick
Posted Tue Jun 21, 2005 5:45 pm GMT by zeroswarm
Well its all interesting stuff but me being me I would still be raising the hell out of this guy. 8)
Posted Tue Jun 21, 2005 6:24 pm GMT by Iron Butt
Yeah, I made pretty much the same argument, Nick, and was told I was missing the point. Nice touch, it hadn't occurred to me to apply pot odds to the blinds.
I think we more practically minded types are just getting in the way of the philosophising LOL. Which is fine... I think the practical aspects of the question are more interesting, think I'll go start that thread now.
Posted Wed Jun 22, 2005 12:21 am GMT by suitedaces84
If the BB defended his blind 100% of the time would you never raise? Of course you'd raise when you got a good hand, that's how you would take advantage of his leak. By raising a player who folds too much with a great starting hand you're allowing his leak to benefit him. I *think* that was the point--it makes sense to me anyway. Does that make sense to you guys?
A few other things I'd like to point out:
-the leak is no where near as big as you guys are guessing it is. Keep in mind it will rarely be folded to the SB, so his "leak" is not even close to what some of your math has it as.
-would you call an all-in from the SB with any two cards if his all-in was only doubled your blind? I'd hope so. There are very few hands that you will be greater than a 3:1 dog to, and you're getting 3:1 on your money. By that logic wouldn't it be correct to call a preflop raise from the SB with any two cards in limit hold'em?
Posted Wed Jun 22, 2005 1:01 am GMT by snoogins47
I'm too tired to go into great detail this evening, but a few comments.
| Quote: | | ...even make a certain +EV decision the same each time, because for instance there will be certain situations where there will be an even higher EV option, hey, I totally agree. In fact, and no offense intended here, I thought that was such a basic principle that it couldn't be what this was about. |
This was actually one of the major points I wanted to make. And it IS a very basic concept. But it's one that's easily overlooked, and many people don't always keep it in mind. Too many people end up making stupid decisions "because it has a positive expectation," and don't even stress about it, whereas they will say that their opponent is the worst player in the history of the world because his call on a gutshot on the turn carries a negative expectation of 4/10ths of a big bet. Clearly not everyone suffers from this, but it's really almost reaching epidemic status amongst the mediocre players of the world, especially on every poker message board ever.
We're so caught up in trying to find +EV situations that thought processes often become flawed. In essence, we shouldn't think "we fold here because calling isn't profitable," but "we fold here because folding carries a higher expectation than calling." And likewise, we don't raise with the nuts because "it's profitable." Calling heads up with the immortal nuts carries a positive expectation. But it's asinine to do so, because raising MUST carry a higher one. It's obvious in some cases, but in general, I think it's a concept that most people would do well to have beat into their heads.
| Quote: | | What I thought this might be about is the whole blind stealing issue; how big a problem is it, how to do it and defend it effectively. Seems like it could be a much bigger weapon/problem than I had thought. I'll have to start a thread on that. |
That is another factor. Some people don't really grasp the impact that blinds and dead money really have on poker strategy, and how powerful steal equity really is. One of the most fascinating things I ever read was a chart that... I dunno who did it, but I want to say Sklansky was involved.
Anyway, the premises were that you are in the SB (I believe) and your opponent was in the BB. You have hand X. If you were to push all in, show your opponent your cards, and he was only going to call when he was a money favorite, how large of a stack can you have for this push to still carry a positive expectation?
Of course, AA was infinite. KK was way into the hundreds of BBs. The more interesting (and shocking to a lot of people) data was down a bit, in the realm of say, late stage SNG play. When you have around 10BBs in your stack, even if your opponent were to call ONLY when he should (given that he knows what you're holding), it's profitable to push a LOT of hands. We're talking T6s, J6o type hands. If I can dig up the chart I'll post it here. Probably be an eye-opener for a lot of folks.
| Quote: | | Let's say he only calls 50% of the time. How big is the leak? He's losing $2 once every 8 hands. That's 6.25 BB/100 hands. |
You're investing $3. To win $3. He folds 50% of the time. From a pure steal equity standpoint, you're sure as hell not making much money out of that. Of course, you're still going to win some of the time you're called, but since you were talking merely the steal equity here, you might want to go back and run some of the numbers again.
| Quote: | | Maybe I got my numbers wrong, but now that I look at that again, I'm not so sure Snoggins is right about this. |
The point I was making is that a play isn't necessarily best just because it carries a positive expectation. If you're disagreeing with that, you're unequivocally wrong, and I hate to use such absolute terms in anything related to poker. Other than that, I didn't really make any startling contentions: I said that a player who caps every street when he has nothing, and resorts to calling when he has one pair or better would easily cause us to have situations where we should be limping instead of raising, despite the fact that raising carries a positive expectation.
The specific example I gave was to illustrate the point, not to give you an idea of how to play in specific circumstances. However, I'm going to elaborate on the specific situation a bit.
I think the chances of raising 100% of the time being sub-optimal are much higher than ya'll tend to think. I've run into quite a few people who don't defend enough to stop me from making an immediate profit by raising every hand. Some of these people are quite bad post flop as well, and some of them have major leaks in that they overplay hands after the flop, and are very predictable while doing so. Some of these folks, as well, make mistakes which are compounded many times in smaller pots, and like to try to run their opponent out of smaller (unraised) pots. I'm sure not raising preflop 100% of the time against these sorts, though I'm still raising very frequently.
I will also say that in a real life situation, it is probably more likely to be "correct" to complete/fold occasionally, because even an occasional slowing down greatly decreases the propensity of your opponent to start (correctly) playing back at you and loosening up his defense.
Also, it's much easier to envision a situation where you're better off limping frequently, even if your opponent folds his BB too much, in a no-limit or pot-limit environment, and I'm sure it's pretty obvious why.
I'm not sure I've really even begun to make all the points I wanted to, but I'm too sleep deprived to really care at this point. I'll be back tomorrow to ramble some more to the handful of people who care. Wait, nobody cares: rather, to the handful of people who read it.
I hope you started the thread about stealing Iron, because I'm sure I have something to ramble about in there too.
Posted Wed Jun 22, 2005 10:40 am GMT by red_pen
I want another go - beats working for a living.
| Quote: | | The point I was making is that a play isn't necessarily best just because it carries a positive expectation. If you're disagreeing with that, you're unequivocally wrong, and I hate to use such absolute terms in anything related to poker. Other than that, I didn't really make any startling contentions: I said that a player who caps every street when he has nothing, and resorts to calling when he has one pair or better would easily cause us to have situations where we should be limping instead of raising, despite the fact that raising carries a positive expectation. |
Can anyone justify limping against a big blind who folded to every raise, even if he played as stated post flop?
And correct me if I am wrong, but you can't cap in heads up limit - the pot can raised until one player is all-in, no? In either case, ask yourself what is the best strategy against a player who plays every street as described and folds every hand to a pre-flop raise. There may be a case (and it may be Snoogins47's point) that when the number of hands you can play against this player is limited and his stack large, it is better to limp and try and get as much as you can when you have the chance. But this is the exception. Without any doubt the best long term strategy is to raise every hand before the flop. If you even think for a moment that this aint so in any real world  scenario (no infinite bankrolls, no artificially limited number of hands etc), perhaps it's time to think again.
Hopefully this bizarre example shows that I do not dispute that simply having a positive expectation is not necessarily enough to justify one line of play. The dispute arises from the specifics of the problem stated. And the problem as stated relates to a long term strategy, is one about blind stealing, and has a contraint in that the folding strategy cannot be adjusted accordingly.
Lets put some figures in play:
The small blind raises only with pairs "AA-88"
A 72o has about 12% (of the pot equity). 73o (13%), and 94o (14%).
The small blind raises with "AA-55" AK, AQ, AJ
72o (22%) 73o (23%). 94o (25%).
The small blind raises with "AA-22" Ax, Kx, QJ-8
72o (30%) 73o (32%). 94o (32%).
The small blind raises with everything
72o (35%)
The looser the raise, the more correct the call, and the more incorrect the fold. When the sb raises with everything, everytime the bb folds he loses. (on average). With 72o he loses 0.8 of a sb by folding rather than calling. With 94o he loses 1.2 of a sb, with T7o he's losing a big blind. Incidentally if the sb limps and the bb is playing 72o the value of the mistake to the sb is 1.4 small blinds.
the final analysis
So however odd it seems - if the sb starts raising all the time the bb should at least call with anything. But he can't call all the time because this is a constrain of the problem.
Now it has got to be accepted that for a given card combination (eg sbAA v bb A6 say, with an A6x flop) there is a lot to be said for limping. But this is outside of your control, is randomly determined, and has a low probablity of occurence. You also can't know any of this when you decide to make your play. Besides, how would you feel if the flop was 457, 578, or 345 - or there were running 6s. I believe that the sheer randomness of the flop erodes your edge (it's not to say your edge isn't there, I just mean its not there all the time) to a point well below the advantage you gain raising pre-flop. And that the case for limping some of the time is just not there.
You have two pieces of knowledge, your own cards, and that the bb might fold to a raise. A play isn't necessarily best just because it carries a positive expectation but in these circumstance a raise carries the best positive expectation all the time IMHO.
Posted Wed Jun 22, 2005 11:54 am GMT by NickFlynn
Ok, I'm willing to endure Snoggine wrath a little more here. I've got pretty good reading comprehension skills, so I think I know what you are getting at, but I think I can phrase it differently.
When you do an "EV analysis" of a particular play, you are normally making some sort of simplification of the situation. In the case of a blind steal situation, you are normally assuming that play after the flop proceeds "normally," whatever that means. (I take it to mean that the players play relatively equally after the flop.) In cases where this is not true, it is NOT CORRECT to make that assumption, and hence the EV you calculate is not the true EV of play.
If you are raising every time from the small blind and he is folding 1/2 the time, you are pocketing $3 every 4 hands in fold equity. That's $0.75 in positive EV. That's the conventional analysis.
There is another side to the coin though - you've lost the post flop equity on the hands that he folds. That needs to be subtracted from the fold equity to get the true value. Let's say you make 25 BB/100 hands on the hands that go to the flop - that's $1 per hand. Now raising every hand cost you $2 every 4 hands, or $0.50 / hand, reducing the EV of raising every hand to $.25.
Obviously, at some point the lost post-flop equity could exceed the fold equity and the play becomes EV-. That only occurs in relatively extreme circumstance though.
- Nick
Posted Wed Jun 22, 2005 2:27 pm GMT by snoogins47
Nick: I might easily be missing something, but in a standard 1/2 blind structure game, you're investing $3 to win $3. And you win that $3 half the time. I can't see how the pure steal equity is anything but zero. This analysis is admittedly flawed anyway because it's assuming you're losing every time you're called, but bleh. The only reason a raise becomes profitable here is because of your equity when you're called.
| red_pen wrote: | Can anyone justify limping against a big blind who folded to every raise, even if he played as stated post flop?
And correct me if I am wrong, but you can't cap in heads up limit - the pot can raised until one player is all-in, no? In either case, ask yourself what is the best strategy against a player who plays every street as described and folds every hand to a pre-flop raise. There may be a case (and it may be Snoogins47's point) that when the number of hands you can play against this player is limited and his stack large, it is better to limp and try and get as much as you can when you have the chance. But this is the exception. Without any doubt the best long term strategy is to raise every hand before the flop. If you even think for a moment that this aint so in any real world scenario (no infinite bankrolls, no artificially limited number of hands etc), perhaps it's time to think again. |
If we say our opponent plays very poorly post flop, and that includes a lot of chip spewing, and he folds every hand to a preflop raise, I can't believe it's right to raise with AA, assuming he's not paying attention. Though the math we've been running is a lot of theoretical garbage, I'll continue down that road. A player who's folding every hand in the BB is a 1224 to 1 underdog to be making a mistake by folding if you're raising AA. This is ignoring post flop EV. I will grant the concession that if our opponent folds EVERY hand preflop, and we IGNORE post flop play, you're a moron if you ever complete. That's pretty obvious, since raising carries a +EV of 1 small bet, whereas completing with AA would be somewhere along the lines of .85xSB. The flaw in your discussion is that you're basing your numbers on the premise that the raise is happening every time. If it's correct to complete AA and raise every other hand, the "overall" expectation from the big picture isn't going to change much, since the raiser is still going to be raising over 99.5% of the time.
An undeniable answer to our problem would take a lot more work than any of us are actually going to do, and a lot more work than I'm capable of. I'm not going to play heads up against a player, tell him to play in a certain sucking way, and do this for years. I don't have Turbo Hold'Em, so I can't set up any specific sims. We aren't even working with any set parameters: if a player folds 51% of the time, this is vastly different from somebody who folds until he has AA. A player who is too tight post flop makes this problem completely different from a laggy maniac. When your analysis hits the post flop part of things, you sort of stopped giving any numbers. Just a few random smatterings of possible flops.
Well, there are quite a few very common situations where your opponent will be making sizeable mistakes post flop, when you hold AA.
For instance, when your opponent flops any pair, he's probably going to be putting a bet or two in as greater than 4 to 1 dog. When it's top pair, he's likely to lose a lot more than that. If we assume that every time he flops top pair, he's going to raise us, and we're going to 3bet, which seems a fairly safe assumption (even though sometimes he will wait until the turn to raise) we're making on average over 2.4SBs per flop where that happens, and that's ignoring turn and river play. Sometimes this is even worse for our opponent because he is only drawing to two outs.
Sure, occasionally our opponent will out flop us. But when he doesn't, and gets a piece of it, a generic "bad" opponent will end up spewing a lot of chips drawing fairly thin. Like I've said about 100 times though, this hinges on the fact that opponents overplay their hands post flop while being too tight preflop, and things like that.
Against the extreme opponent I outlined earlier (which I can't imagine you're actually referring to... but just in case) there's absolutely no way, if we assume no cap heads up (I was assuming a cap heads up because without it, the discussion is ludicrous), that raising becomes the best play. Playing a pocket pair to the river carries an EV of somewhere near half of his stack, or your stack, whichver is larger. This is because given his flaws, you can get in unlimited raises with the "nuts" based on the parameters of his play, very very very often.
I'm just fairly certain that it can't be correct to raise 100% of the time just because our opponent folds "too much" if he plays poorly, and us well, after the flop. It's pretty much tantamount to a "slowplay" after the flop. If we have the nuts on the flop it's clearly an immediate mistake to not bet. But depending on the range of hands the player will call with vs. what hands he will fold that have a chance to become hands he will pay us off with, checking can still carry a higher EV.
In the real world I think it's much more likely to be correct to refrain from raising. I've given the reasons for that. I think it's almost impossible to deny that this could happen in a pot limit or no-limit game. And I think the worst part of this argument is that we have to try to quantify post flop play to a magnitude that is essentially impossible.
Posted Wed Jun 22, 2005 2:37 pm GMT by Loonbat
You guys keep this up and we're going to insist you stick to posting on "2plus2" ... whatever happened to the poker concept of "Do it if it feels good" ... 
-Loon
Posted Wed Jun 22, 2005 3:03 pm GMT by NickFlynn
| Loonbat wrote: | You guys keep this up and we're going to insist you stick to posting on "2plus2" ... whatever happened to the poker concept of "Do it if it feels good" ...
-Loon |
LMAO.
I humble beg your forgiveness.
Please do not banish me to the Kingdom of the Know-it-all Arseholes.
- Nick
Posted Wed Jun 22, 2005 3:05 pm GMT by age_of_sages
I like this conversation that's going on here, though I have nothing to add to it.
I just wanted to point out that this:
| Snoogins47 wrote: | | I'm not going to play heads up against a player, tell him to play in a certain sucking way, and do this for years. |
made me laugh more than I have in the past few days, thanks Snoo! 
Posted Wed Jun 22, 2005 3:15 pm GMT by NickFlynn
| snoogins47 wrote: | Nick: I might easily be missing something, but in a standard 1/2 blind structure game, you're investing $3 to win $3. And you win that $3 half the time. I can't see how the pure steal equity is anything but zero. This analysis is admittedly flawed anyway because it's assuming you're losing every time you're called, but bleh. The only reason a raise becomes profitable here is because of your equity when you're called.
|
No, the raise itself is profitable if he lays his hand down more than 25% of the time.
Look at the pre-flop decision from the perspective of the BB. When you raise, he is looking at a $2 call for a $6 pot. Folding any more than 25% of the time in this situation is leaking chips. If he is only calling 50% of the time, he is basically handing you $3 every 4 hands - which is your +EV on the raise.
(This is assuming he never re-raises. If he does re-raise, he can call less than 75% and not be leaking.)
- Nick
Posted Wed Jun 22, 2005 3:26 pm GMT by NickFlynn
| snoogins47 wrote: |
An undeniable answer to our problem would take a lot more work than any of us are actually going to do, and a lot more work than I'm capable of. I'm not going to play heads up against a player, tell him to play in a certain sucking way, and do this for years. |
You don't have to - that's what we have Party Poker for.
(Sorry, couldn't resist.)
- Nick
Posted Wed Jun 22, 2005 3:37 pm GMT by NickFlynn
| snoogins47 wrote: |
In the real world I think it's much more likely to be correct to refrain from raising. I've given the reasons for that. I think it's almost impossible to deny that this could happen in a pot limit or no-limit game. And I think the worst part of this argument is that we have to try to quantify post flop play to a magnitude that is essentially impossible. |
You are so much more entertaining that work.
I agree - in the real world, there are complications. If you raise every single time, even the biggest idiot in the world is going to catch on eventually, and start either calling more, or even playing back at you. If there weren't complications like that, poker would get pretty boring, really fast.
On the other hand, you shouldn't underestimate the vulnerability of the player who fails to defend his blind enough in a heads up situation.
- Nick
Posted Wed Jun 22, 2005 4:56 pm GMT by Iron Butt
Clearly this issue has divided into the "hey, I'm kicking his ass, why stop with the blinds?" camp and the "a bucketful of cash made in the blinds is easily worth a possible two bucketfuls post-flop" camp. Who's right? I say you both have a point and no one's "right". My question is how does this weird hypothetical situation relate to real poker? Not sure it is at all except I now have the uncomfortable feeling that I'm not aggressive enough in and around the blinds...
| Snoog wrote: | | I'm too tired to go into great detail this evening, but a few comments. |
...followed by 1134 words. Yeah, if you're going to post make an effort, willya?
| and then Snoog wrote: | | One of the most fascinating things I ever read was a chart that... If I can dig up the chart I'll post it here. |
Yeah please do. Or let us know where to get it.
| and furthermore he wrote: | | I hope you started the thread about stealing Iron, because I'm sure I have something to ramble about in there too. |
I did... kick it off willya, it's not getting any action. It's here if anyone missed it or cares:
http://www.texasholdem-poker.com/forum/t11531/blind-stealing-and-defending
I'd really like to hear what you guys have to say about blinds beyond the mythical "raising every time" situation...
Posted Thu Jun 23, 2005 3:16 am GMT by snoogins47
| NickFlynn wrote: | | No, the raise itself is profitable if he lays his hand down more than 25% of the time. |
I'm close to certain that this isn't right. We can't look at it from his perspective really, because after all, we're basing this on our own.
The bottom line here is that regardless of whether he's making a mistake or not based on his pot odds and your range of hands, you're betting $3 by raising.
When he folds, you win $3. When he calls (for the sake of this argument) you lose $3. Despite any mistakes he may be making, you're not going to show an immediate profit by raising if he folds less than 50% of the time. I can't imagine this not being the case, regardless of how sub-optimal his strategy may be.
Anyhoo, I don't really feel I'm underestimating the vulnerability of said opponent, since I'm only really advocating refraining from raising with big pairs against bad opponents. Either way, I'm kinda glad this thread has spiralled out of control from what I originally intended. Always fun to discuss.
| Quote: | | My question is how does this weird hypothetical situation relate to real poker? |
Well I just wanted people to think. This hypothetical may not ever directly apply to you, but the concepts contained are very important, especially the "rudimentary" concept of maximizing expectation. I use quotes because it's quite a simple and basic concept, but one which is not only often misinterpreted, but often overlooked.
Posted Thu Jun 23, 2005 8:21 am GMT by NickFlynn
| snoogins47 wrote: | | NickFlynn wrote: | | No, the raise itself is profitable if he lays his hand down more than 25% of the time. |
I'm close to certain that this isn't right. We can't look at it from his perspective really, because after all, we're basing this on our own.
The bottom line here is that regardless of whether he's making a mistake or not based on his pot odds and your range of hands, you're betting $3 by raising.
When he folds, you win $3. When he calls (for the sake of this argument) you lose $3. Despite any mistakes he may be making, you're not going to show an immediate profit by raising if he folds less than 50% of the time. I can't imagine this not being the case, regardless of how sub-optimal his strategy may be.
|
I really think you are wrong about this. The flaw in your argument is that you are assuming that we lose the $3 automatically when he calls. Even granting that he has, on average, a better starting hand than we do when he calls, his edge is probably no better than 55%-45% on those hands.
The folds he is making that are -EV are, by definition, +EV for us - it's the poker equivalent of Newton's law.
Anyhow, that being said, I completely agree with you that against a bad player, you don't want him folding when you've got a legitimate hand, because you stand to win so much after the flop. Real world, if I think I'm up against a passive bad player heads up, I'll raise with my bad hands, and check/call my good ones before the flop. At least until he gives some sign of catching on.
This isn't an example of a bad player per se, but a good example of how you have to mix it up and adapt to your opponent. I was in a prolonged 3 way battle at the end of a SnG. All three of us had roughly equal stacks with blinds about 10% of stack size. Whenever it was folded to me in the small blind, I put in 4xBB raise anytime I had any sort of hand, and called when I didn't. After about 12 circuits, the guy on my left started re-raising all-in every time I called. A few hands after it dawns on me that he's figured me out, I look down and see KK, call, he raises all-in, I double up.
- Nick
Posted Thu Jun 23, 2005 1:15 pm GMT by snoogins47
Alright, we're arguing semantics here then... I was merely trying to calculate the equity the raise gains from stealing. Of course in practice it's more effective than this, because as I had mentioned, we win sometimes when he calls.
Posted Wed Jun 29, 2005 2:58 pm GMT by Iron Butt
| Quote: | | My question is how does this weird hypothetical situation relate to real poker? |
I retract this question. I've been ripping people up heads up lately, I credit in part this discussion. Thanks guys.
|
|
