Making the Pot Bigger vs. Keeping it Small?

[Max Weinberg]

I've been thinking about this one lately. Ok, you pick up some premium hand in the blinds and the table has limped to you, yadda yadda. You either raise and give everyone odds to call to make whatever crap they're holding, or you just check your option and make moves when you know that they're making mistakes when they call. Chicken or the egg stuff.

The question is, even if you raise and give them correct odds to draw, does that somehow offset your status as a 4:1 favorite (I'm just making it 4:1 for the sake of argument, I'm not going to use actual hands).

You made the pot large enough that they are correct to call all kinds of hands, but aren't you more correct to be raking in that money as a 4:1 favorite? Their being correct in calling a gut-shot on the flop doesn't detract from your EV on the hand does it? He's making a call that makes him money in the long-run, but it's also making you money - you're the favorite. Maybe I'm just misunderstanding this theory from HPFAP and someone can point out where I'm wrong.

If my goal was to systematically bust every player on the face of the earth, keeping the pot small to screw with their odds would be my weapon of choice. But in a pure cash sense, raise it up, right?

[William]

Right!!!

[CMangano]

The other thing to think about is the players you play against. If they are the type who calculate odds before deciding whether or not to chase, then keeping the pot small would have its benefits. But if they are going to chase you regardless of the pot size, then I see no reason not to build as big a pot as possible.

I am sure some will disagree.

[1800GAMBLER]

[ QUOTE ]
Ok, you pick up some premium hand in the blinds and the table has limped to you, yadda yadda. You either raise and give everyone odds to call to make whatever crap they're holding, or you just check your option and make moves when you know that they're making mistakes when they call.

[/ QUOTE ]

You answer your own question there. When you raise preflop and they call they are making their mistake there, when you have big holdings that is. So by that mistake they have then added dead money into the pot and then get the chance to make +EV plays if the flop hits them.

So when you wanted to keep the pot small and know they are making a mistake on the flop, you always know they are making a mistake preflop.

Big unsuited cards you generally want to thin the field out though.

Take a deck of cards take the 2 cards out you want as your big hand then deal out a 10 handed game, take out the percentage of cards that usually fold preflop (you'll be doing this tighter than it actually is due to the blinds) then check out the EV on twodimes.net - if you want to make things interesting remove a few hands and see how the difference goes up/down for hands multiway vs short field.

[Max Weinberg]

I guess what I'm trying to ask is does their making +EV plays eat into your already +EV move of taking a big premium hand into a field of bad players.

If you're the 4:1 favorite and by your raise in the BB, some guy has +EV odds to call for a gut-shot, does that eat into your EV on the hand? You're still the favorite and making good money on your hand. Whether or not gut-shot boy has odds to profitably call doesn't really have any bearing on your EV, does it? He's correct in calling and you're correct in betting/raising.

Of course this whole question goes out the window if we're assuming players who are aware of pot odds, but I'm still curious to know about these implications. I'll deal out some hands later tonight and run them through twodimes to see if I can't figure it out.

[Dynasty]

How are you calculating what your odds of winning are? Often, this type of analysis assumes all hands are going to the showdown. If so, then your pre-flop raise doesn't change your favorite status.

[ACPlayer]

Raising in the blinds has other considerations that are perhaps more important. Specifically position and how the hand will play on the flop and beyond. Then factor in how the hand will play after the flop if you raise and pretty much have to bet into a large field, yada yada yada

[baggins]

if the guy limped then he's getting the same odds to call the raise from the BB. so if it's a mistake for him to limp based on pot odds, then calling the raise preflop is also a mistake. you're asking your opponents to either make mistakes preflop, or make them postflop. might as well get their chips in while you're a favorite then to let them see a cheap flop and then have correct odds to draw out on you.

[Max Weinberg]

The way this hand will play out is not possible by any standards, but I think it answers some of my questions and perhaps brings up some more theoretical ones. Here it goes, and feel free to chime in if I'm wrong in some spots as I haven't done much math in a while.

You pick up QQ in the BB after the entire table has limped in. What to do? What to do?

You Raise:

The flop is a 23T rainbow. You bet out and only get called in one place by a guy holding 56. You put the 21st bet into the pot and he calls getting 21:1. He doesn't hit on the turn and calls your bet again getting 12:1 on his call. The odds of completing a gut-shot by the river are 5:1, so five times your QQ wins when he folds on the river after getting no help. Five times you win 14BB for a total of 70BB. One time in five he spikes and you will lose 4.5BB (he gets a raise in there along the way, and you call him down). Net total of 65.5BB per six times you play this hand out.

The guy with 56 loses 2.5BB five times for -12.5BB. One time he hits and wins 17BB for a net profit of 4.5BB per six play-outs.

You Don't Raise

The flop is the same 23T rainbow, but this time you check to the bluff-aholic button who bets, and you check-raise him. He folds, but 56 calls-cold getting 6.5:1 on his call. He'll call the turn also if he doesn't hit. Five times your QQ wins 9.5BB for 47.5 BB net, and lose 4.5BB once for a total profit of 43BB per six hands. 56 will lose five times to the tune of 2.5BB for a total of -12.5BB. But he'll also win once and get an extra bet for a single 14.5BB. He makes only 2BB per six play-outs.

The Interesting Part

I couldn't contrive an example in a family pot where you don't raise and the gut-shot is not getting at least a little profit on any call he could make, so I had to make due with giving him less pretty odds than in the first hand.

When you raise in the first situation, the profit that 56 shows over an average six hands is 0.069% of yours.

When you check-raise in the second situation, the profit that 56 shows over an average six hands is 0.047% of yours.

Now the 0.022% isn't a monster of a percentage, but what I've inferred from this is that your edge is greater when you choose to manipulate the pot odds in your favor. You don't take down as big of a pot when your hand holds up, but you are making better money on the hand when you play it out the same way in similar circumstances. You get paid off slightly better in the long-run when you manipulate the pot odds to "widen the gap" between their EV and yours.

Is this what Sklansky is talking about and did I just re-invent the wheel on this one? I think I finally understand the concept behind this one now, and I'm open to any comments or criticisms regarding this post.

Thanks for reading the novel.

[Schneids]

[Ed Miller]

This is actually the opposite of what S&M suggest. They suggest keeping the pot small against people who chase too much and building pots against people who fold properly (or too much).

[muck_nutz]

I'm not sure I follow your conclusion. I'm not particularly interested in maximizing precentage edge, I'm interested in maximizing my net win. I'd far prefer to have a net win of 3BB which corresponded to a 65% edge then 2.5BB which corresponded to 85% edge (numbers totally made up).

[Max Weinberg]

Using your numbers I'll run through it to try to explain what I'm saying.

You say you'd rather have a net win of 3BB corresponding to a 65% edge rather than a 2.5BB win corresponding to an 85% edge. Let's just assume you lose an even 4BB every time your hand doesn't hold up.

Playing it out over a hundred hands as a 65% favorite you stand to make a net of 195BB while losing 140BB, for a grand total of 55BB in the green.

If you take the smaller pay-out with better odds you stand to make 212.5BB while losing 60BB, for a grand total of 152.5BB in the green.

So my argument is that any extra thing you can do to screw up your opponent's pot odds is a very good deal for you. Any extra few percentage points you can squeeze out of a situation will pay off dividends in the future.

I'm not really advocating raising or not raising. There are way too many variables to contend with like player tendencies, skill, how often they fold in a huge pot, and other things I can't think of right now. In some situations raising might be the best play, and in some, just checking might be. Sometimes it might be worth it to squeeze out those extra percentage points and sometimes it might be worth it to go into the fray with guns blazing.

I'm probably more confused than I ever was, now that I know the concepts behind it than I was when I just took the raise vs. manipulation issue at face value.

[CMangano]

I understand what they mean in HPFAP, but it seems to me that if they are going to call no matter what, then making the pot bigger would be the best option. If you're going to get drawn out on, you're going to get drawn out on. I may be way off here though and would love to hear from others who disagree.

[CMangano]

I would much rather profit 70BB and make his call correct, then profit 43BB and make his call incorrect.

[Max Weinberg]

I think that can just be attributed to the fact that in my example there ends up being 8BB of dead money in the pot on the flop when you raise. My main point in that post was basically to show that manipulating the pot size to where they are incorrect (or less correct) in calling, you make a bigger profit per hand than you would have had you given everyone correct odds to draw to whatever. They eat into your EV on the hand when they are correct to call.

I'm sure a better example hand could be found instead of me artificially creating a bunch of dead money in the pot. If I could find a good example, I think the numbers would work out to show a greater EV when you manipulate the pot size to force your opponents into making mistakes. I'm not sure if it would be a huge discrepancy, but I think there would be a noticeable difference between when your opponent had odds and when they didn't. You'll still get sucked out on with the same frequency, but I'm thinking that you'll show more expected profit per hand on the times you kill their odds. I'll see if I can't figure out a real world situation to put this theory into test.

[Max Weinberg]

I think my previous post above this makes no sense now, as your odds to get win or get sucked out on don't change, simply the profit per hand that you expect to make.

Raising the pot or not doesn't change you to a 65% or 85% favorite. My conclusion in that post doesn't logically follow the information I've given, so just chalk this last post to being a dumbass and not paying attention to what I'm writing. I'll try to think of a better example.

My advice to take a 2.5BB profit as an 85% favorite over a 3BB profit as a 65% favorite stands, however. It just has nothing to do with what I was trying to get across.

[Mason Malmuth]

Hi Max:

I haven't read the other posts, so perhaps someone has already covered this.

First, you need to understand that if you raise before the flop, your raise will show a profit. So the question is whether you can show more profit in the long run by not raising than by raising?

In this spot an important thing to consider is: "How well do they play?"

Notice that if your game contains a bunch of good players who just happen to all limp in as opposed to a bunch of terrible players who all limp in, the profit on your preflop raise won't be as great. That's because the penalty for playing a terrible hand is more severe than the penalty for playing a decent one.

To see this, let's suppose you have a pair of kings. Would you rather raise someone who has limped with with 98 suited, or someone who has limped in with 93 offsuit. Notice that you make money in either case, but your raise obviously makes more in the second case.

So this is frequently the answer. If they play really bad, the amount of profit that you give up before the flop can easily be more than what you gain back on the flop and beyond. If they play pretty good, the opposite can easily be true.

Bst wishes,
Mason

[Mason Malmuth]

In Poker, Gaming, and Life there is an essay by David that addresses how to play QQ in the big blind after everyone has limped in. It also addresses how to play AA and 99.

Best wishes,
Mason

[Copernicus]

The original question (at least as I understand it) is much easier to answer in the context of EV then it is in terms of "pot odds".

I'll reask the question in terms of a specific example: You have a hand that is the nuts unless a chaser hits his two outer. Are you better off if you had built a huge pot early on, or kept it small to minimize the chasers odds?

Round the 2 outer's probability of hitting to 5%, and call his probability of folding to a bet F%.

From your point of view:

EV = Pot* [F%+(1-F%)*(1-5%)] + Bet * 2 * [(1-F%)*(1-5%)]

Three things should be obvious. First, your EV goes up when the pot goes up (ie the bigger the pot the better) holding everything else constant. Second, your EV goes up when your Bet goes up (holding everything else constant). Third, the only thing that can reverse "the bigger the pot and the bigger the bet the better" is if you can influence F%.

When can you NOT influence F%? A) When the chaser is ignorant of pot odds B) When the pot is already so big that the chaser always has proper odds to call (as often happens in a limit game.) C) The trivial case where one of you is all in. D) The equally trivial case where he has overwhelming reasons to call despite not having pot odds (eg small stack in a tournament.)

F% is generally going to be 1 or 0....he will fold if he thinks he doesnt have the proper odds (ie youve manipulated the pot so he really doesnt have the odds, or bluffed him into mis-assessing his W%).

This then, I think, brings you to the essential "skill" difference between limt and PL or NL.

In limit you are limited in your ability to influence F%. When you cant influence F%, you should maximize the pot (which may be by jamming, slowplaying, whatever). When you can influence F% it is usually by aggressive play that causes the opponent to misjudge the odds he needs to call. That is why aggressive play is mandatory in limit.

In PL or NL, if you know your opponents will heed the call of "pot odds", each hand presents a challenge to balance W% and F% (ie your opponents perception of their pot odds). This is restated often in English as "playing the man, not the cards").

[Mike]

Perhaps this will help? You play QQ 100 hands. You lose twenty five times to straights and flushes. You lose 25 times to either a bigger pair or two pair. You lose ten times to a set. So out of your forty wins when do you make the most cash?

If you do not raise on the forty hands you win, but did raise on the sixty you lost, you are in trouble as you aren't showing much if any profit.

If you do not raise any time, your win and loss rate are in a ratio to each other. If you raise all the time, the ratio is about the same, but the actual $ difference is greater. So there is not much difference except the $ you are willing to risk.

I am guessing you read the more important previous post. So, you should understand the trick is to mix your calling and your raising to the quality of the players in that particular hand. Bad players play to the river because they came to play. Better players play to the river because they feel they are going to win the hand.

If I am seventh in with a strong draw preflop and it's capped by the time it gets to me, my odds are the ~ same as if there were no raise at all - assuming the prior raisers also call the raises. The amount of money I have to put in the pot is greater, but the odds are the same.

You really have two variables to manipulate depending on your players. How good are they and how much do they value their chips? Anything you do should be done with those two parameters in mind.

In wrapping up there is no magic in pot size and players, only a good idea of how well they play, what they are willing to call, and when will they throw in the towel. Include this idea in your raising requirements and it will help you plan when and how to manipulate the pot.

[muck_nutz]

"My advice to take a 2.5BB profit as an 85% favorite over a 3BB profit as a 65% favorite stands, however. It just has nothing to do with what I was trying to get across."

I think maybe a nomenclature issue (or a misunderstanding) made you come to this conclusion, but I believe its wrong. Lets take an all heads up all in equity example. YOu get all in with 65% = 3BB net of a pot (total pot is 20 bets; .65*20 = 13 = .5*20 + 3). Over a hundred trials you will expend 1000 bets to get back 1300 for a 300 bet profit. In the 2.5/85% case (total pot = 7.2 bets; .85*7.2 = 6.12 = 3.6 + 2.5) you will expend 360 bets to get back 612 for a 250 (plus round off error) bet profit.

The reason I made what I thought was an obvious statement (you want to favor making 3.0 bets/hand over one of 2.5 bets/hand) is that in the examples in your post you seemed to be favoring making 43 bb/100 hands over making 65.5 bb/100 hands. Thats why I prefaced my earlier post with some confusion.

Unlike my examples the typical limit holdem multiway pot contains a lot of dead money. In those cases it might make sense to control the pot size, but in the end your expected average win on the hand should be higher (not lower) then if you didn't control the pot size. In many cases, eventhough you pump up the pot to a size that makes your opponents future calls correct, you will make so much earlier in the hand that getting a smaller fraction of a bet later in the hand still makes for a larger average win. Its only in those cases that what you get from earlier bets isn't so large that making more on fewer later bets makes sense. This is far more frequently common in NL (and to a lesser extent in PL) when you can take a big chunk of your opponent late in a hand.

[cbloom]

This has been answered pretty well, but let me try to sum up the facts clearly.

You have the best hand, so just calling or raising are both +EV moves. The question is which is more +EV?

The answer to that depends on the details of the game. The problem is that when you raise before the flop, the winning pot will be much bigger, but there will also be many more people going to the showdown, so your chances of winning are less. Exactly which way is more +EV is depends on a lot of details.

One thing you can say for sure is that raising increases your variance *a lot*. If the EV both ways is not very different, most people will prefer the move with the smaller variance, which is to not raise.

[Max Weinberg]

Thanks for the replies everyone. I think I understand the differences between, and reasoning for both options now.

[Bokonon]

I understand that not raising pre-flop in the BB with AK or AA or KK minimizes variance, but I've always wondered why Slansky recommended maybe not raising in those positions with those cards. Admittedly, he's played a little bit more poker than me /images/graemlins/smile.gif.

It's certainly true that people will be less likely to chase with a small pot, and that if no one hits you'll have a lot more luck bluffing them out with overcards. But . . . well, certainly for the big pairs like AA and KK, it seems to me I would far prefer walking in preflop with it capped than with just one bet each, even if there are six or seven limpers. I mean, you're wrecking the implied odds of all those doofuses with suited connectors and such, right? Even it it means that one or two of them may have the odds to chase to the river.

Someone sharper than me needs to set up a few simulations using the TTH software, where Our Hero gets AA on the button and raises preflop, vs getting AA and not doing so.

[CrackerZack]

Sklansky never advocates not raising AA/KK pre-flop. there is a small section about checking the flop thru, raising the turn to knock out long shot draws that would be getting odds to call if you bet the flop, bet the turn, but most people tend to misunderstand it and misapply it.

[Bokonon]

Oops on me -- I was conflating advice on raising with AA/KK preflop in the BB with raising with AK. I *think* that he may have suggested not raising with AK in a many-limper pot when you're in the BB, but correct me if I'm wrong. (That happens a lot.)

[CrackerZack]

this came up in a seminar he gave. the response contained a quote (paraphrased) "Understand that when you raise PF with AK from the BB, you will win the pot less often, but the pots you win will be bigger" Which is the crux of this argument. You have to evaluate how well your opponents play and will they pay you off all the way when you flop big? Can you manipulate them out of position? Is the most aggressive people on your close right? All of these factors have to be taken into account when deciding whether or not to raise. Raising is +EV, but you have to decide if you can make more EV by not raising. Most players aren't good enough to make back up EV they're giving up PF (myself included I believe), so I raise PF. If I feel very in control of the table, or think I have a large post flop edge, I'll consider not raising, but its rare.

[huzitup2]

alot of us play in games where A/rag [off] and oft-times K/rag[off] - not to mention some other truly horrible hands - are played as a matter of course.

When I know this is the case in the game I'm in - wow - I just can't keep myself from raising QQ/JJ out of either blind.

If an overcard flops, I usually wind up having to check and fold.

However, there is a better than 50% chance that there will be no O/C when I have QQ, just over a 50% chance that an O/C will come to my JJ, not to mention that the occasional flop does come with an O/C AND my set card.

Am I wrong to be raising QQ and JJ in THESE games ?

*

I seem to recall another essay - I think YOU wrote this one - that said (basically), "if your opponents are playing absolute trash it probably IS correct to raise with hands such as AQ/off etc., even in multiway pots. You are gaining so much equity from calls by hands such as 64/off, etc., that you more than make up for the POST FLOP equity you surrender". (This is paraphrased; the ""'s may be inappropriate here).

Best wishes,

- H

[Copernicus]

[ QUOTE ]
this came up in a seminar he gave. the response contained a quote (paraphrased) "Understand that when you raise PF with AK from the BB, you will win the pot less often,

[/ QUOTE ]

How can that be? It makes no sense to me. A raise isnt going to keep more people in, it can only drive people out which means you will win more often. The only scenario where you could win less often is if you are reraised and you fold yourself, but, at least in limit, you would only fold if you believed you were up against AA or KK, and you still would have lost to those hands in most circumstances.

[cbloom]

BTW in this vein - David & Mason - I think your books are badly in need of updating. They all assume that your opponents are basically solid players. That may have been the case in the mid/high-limit games ten years ago, but these days we have a huge amount of 1) novices and "action players" in the low limit games, and 2) experts who are mixing it up a lot, deviating severely from the HPFAP guidelines. You guys should do a whole new version of HPFAP that deals more with the wide variety of opponents you may be up against.

[Bozeman]

Because hands that would have called on the flop in the bigger pot may now fold. Of course, the slight increase in the chance of winning the pot doesn't make up for the extra bets that aren't in the pot.

[Copernicus]

Either you are agreeing with me that you will win more often (not less) due to other players folding (I was talking pre-flop, you are talking flop), or you are saying your AK may fold more often on the flop.

The only way AK will fold more often on the flop is if the bigger pot invites more aggressive action on a ragged flop. I would think that will happen less often than the number of wins gained from pre-flop folds.

[pudley4]

The idea is that by raising, you are making the pot bigger, so more people will call (correctly or incorrectly) to chase their draws. With more people staying in the hand longer, you will obviously lose the hand more often. What needs to be figured out is if the bets that go in preflop outweight the number of times you lose due to the larger pot (I think they do).

Remember that when you raise from the BB, you are not doing it to knock people out - it is extremely rare that a player will fold preflop for that one bet.

[Louie Landale]

You've mixed two points.

[1] Bet and call. I have the best hand and the opponent has the correct odds to call. It is correct for me to bet and for him to call. I do better when if folds, and he does better if I check. There is no quandary here. It is the equity of my hand that makes the bet correct, and the equity in the pot that makes his call correct.

[2] Manipulating the size of the pot. So you are contemplating NOT raising PF as a clear favorite in order to keep the pot small. That's a mistake. Its only justified if that mistake causes the opponent to make an even greater mistake later. [2a] If he is going to call you on the flop regardless of your raise (he calls correctly) or you don't (the pot is small and he calls incorrectly), then you lose whateve equity you would have gained by your PF raise. Yes, you manipulated the size of the pot and caused the opponent to make a mistake, but it only costs you money: your chances of winning are the same but you engage a smaller pot when you are the favorite. [2b] If he is going to CORRECTLY fold on the flop if you do not raise or will CORRECTLY call on the flop if you do, it works out that you usually make more money by raising. Work it out.

Manipulating the size of the pot only has a chance of being correct if you can [1] get the opponent to CHANGE what he would have done, and [2] this change is a mathemetical "disaster" for the opponent, such as willingly draw dead or fold the best hand.

- Louie

Note: not raising on the flop figuring that will let you raise on the expensive turn is a valid play, but its NOT "manipulating the size of the pot".

[Louie Landale]

In addition to my other post, I would like to say that raising or not raising PF in this situation, say with QQ, is MOSTLY about [1] whether you can make up lost ground with a clever check-raise or 3-bet later, and [2] whether you can save a little money by waiting for a favorable non-AK flop.

[1] Raising PF pretty much announces your hand and its unlikely you will get raised later in the hand. Checking figuring to auto-fire on the flop MAY let you 3-bet since players can and will raise a bet from the blinds fairly liberally, and rightfully so.

[2] Check and fold if an A or K hits. I don't know how much this is worth, but its GOT to be less than a bet.

Not raising is "disguising your hand", not "manipulating the size of the pot".

- Louie

[Bozeman]

Read my response and Pudley's carefully, and you will see the error in your thinking.

But as I said, this does not make it a good play.

[Rook1]

in the end scarded money loses
but yeah having AA in the SB is dubious

[Copernicus]

Not an error in my thinking as much as being blinded by the correctness of the play in the context of the original question. If you can get more people to stay post-flop because you have "manipulated" the odds pre-flop to favor their call (but you will win if they miss their draw) then you have increased your EV.

Just to nail this down a little more directly, in the long run a raise from the BB will not, IMO, result in winning fewer hands. On balance you will get more folds pre-flop then you will get winning hands staying in post-flop solely because you have built the pot with the raise. Most often the players you lose to would have either had the correct odds whether you raised from the BB or not, or would have ignored the odds and played anyway. In this (imo) plurality of the cases you have not influenced the folding equity, so not only do you win more often, but you win more.

I disagree with the post that said a raise from the BB is not likely to get additional folds. In a limit ring game not populated by total fish, a raise from the BB is often read as AA or KK, or maybe a "loose raise" with AK, and will often induce folds from limpers. In a tournament setting a raise from the BB almost always loses the loose limpers, getting you heads up.