My current approach to completing from the SB when there are limpers and you have a good price is to stick to the reasonably strong stuff - pocket pairs, suited aces, suited connectors - and to fold the weaker stuff like suited kings, suited gappers, and lots of offsuit stuff. I take this strategy from https://redchippoker.com/complete-sm...in-live-poker/. After reading the article, it seemed like it makes sense, and didn't seem worth my time to study it in much more depth. But now I find myself curious, and wanting to explore it more deeply.

Let's take a spot where you're in the SB with 74o and there were three limpers. You've got 9:1 pot odds. Say we're playing $1/2.

When I try to analyze this with some math, it seems like a pretty slam dunk call. It seems like folding is leaving a ton of money on the table.

Of course, this could just mean that my analysis is flawed. So I'm not necessarily arguing that it is a slam dunk call, just that it seems like it is to me, and if it isn't, I'm not understanding why.

Two pair and better mining
Suppose that you use the strategy of check-folding whenever you flop something worse than two pair, and go for value when you flop two pair or better. With 74o, you flop two pair or better about 4% of the time. If you make $25 on average when you do hit your 2p+, the SB complete is profitable. Think about it - suppose you complete and miss 24 times. You lose $1 each time, and you're down $24. Now, on the 25th attempt, you win $25. Now you're up $1. So if you can expect to win $25 when you hit, it is +EV. So then, I think the question is whether you can expect to win $25 on average.

When you complete and the BB checks, the pot will be $10 (let's just ignore rake for now), and it will be a five-way pot.

If you bet $6 on the flop, get one caller, and then bet $11 on the turn or river and get a call, you've made enough money on your 2p+ mine.

If you bet $8 and get two callers, you've also made enough money on your 2p+ mine. If you make more money on the turn and/or river, then you're doing even better!

Being in a multiway pot, it seems pretty plausible to make $25 when you hit.

Of course, there are some issues to consider:

• Reverse implied odds. Sometimes you hit and still lose. Eg. you flop trips but your opponent outkicks you.
• Being out of position, you're at a disadvantage and it's hard to maximize value.
• The textures where you do hit, your opponents probably miss. Eg. T44.

But still, even considering these issues, the bar of averaging $25 seems pretty low. Sure, when the board comes T44, you may run in to A4 and lose some money. But more often you'll win some money from Tx or pocket pairs. Sometimes you'll lead out and everyone will fold, but more often than not in a multiway pot someone will have Tx, or a pocket pair, or perhaps a flush draw, and give you action.

If we factor in the fact that we may get to see a free card on the turn, or the fact that we may have implied odds to draw vs a flop bet, then things get even better for us. And when we have a stronger hand like A2o, we get a slight bump in EV from our showdown value.

Playing single pairs harder
Perhaps you could boost your EV a little bit by not just check-folding all single pair hands. It seems pretty tough to pull off, but possible, perhaps.

I get the impression that a big reason why playing junk hands from the SB is -EV for so many people is because they overplay them postflop. Eg. not being disciplined with middle or bottom pair, or perhaps overplaying top pair. It makes a lot of sense that just folding these preflop would save a lot of people money. I don't disagree with that. Here, I'm interested in exploring whether playing these hands is +EV if you play correctly postflop. Or, at the very least, if you just check/fold everything that isn't two pair or better.

Let us assume that the 4% of the time you quoted means we win 4% of the time.

In the small blind your folding range would have an EV of -.5 bb (assuming a sb is 1/2 a bb in your game).

So let's say you started off with just calling say 10% of the hands you would normally fold. Let's also stick with the convention that you just fold if you don't make 2 pair+.

Then you could approximate your EV with this equation:

.9*-.5 + .10*(.04*(4+W)+.96*-1)

Where .9*-.5 is how often you still fold and lose .5 bb and the rest is what happens when you call. W here stands in for some unknown amount you might win with the best hand. .96*-1 is what happens when you fold after you called.

So now, in order for us to choose to call we need the above equation to be at least as good as folding, which was -.5 bb (note all values are in terms of bb)

.9*-.5 + .10*(.04*(4+W)+.96*-1) >= -.5

So let's solve for W here and use that as an estimate of our post flop winnings to see if calling is better than folding.

Simplify the equation:

-.45 + .016 + .004*W - .096 >= -.5

.004*W >= .03

W >= 7.5

So with our basic assumptions it seems we'd have to win approximately 7.5 bb ($15 in 1/2) post flop on average to make the play at least break even.

If all players involved are at 100bb then that means you only have to win 7.5/300 = .025 or 2.5% of the remaining stacks in play. That seems like a pretty good deal right?

But there are certainly flaws with our model:

1. We don't always win when we hit 2 pair
2. We don't close the action when we call so may not even see a flop
3. We can potentially win without showdow (i.e. bluff).

I would say 1 and 2 are much more likely to skew the amount we have to win to break even upwards more than 3 is going to positively comtribute and skew the amount downwards.

So let's say you were correct and we actually need somewhere around 12.5 bb to break even ($25 in 1/2).

The problem with that is there is empirical evidence from players who have been tracking their online results for years to show that it's really really hard for many of your unpaired hands to make 7.5bb per hand on average let alone 12.5bb per hand. Some of that is likely selection bias (i.e. people just don't play 74o that often) but a lot of it is just the fact that these hands just don't make strong enough hands often enough AND either win at showdown or have opponent's whose hands can call on boards that favor them, especially in limped pots.

I mean on a board like T44 what can your opponents who have limped to you be reasonably expected to call with for a significant amount of money? Conversely when a lot of money goes in you might be up against a better hand.

So overall your implied odds with totaly garbage just aren't there especially multiway.

Heads up could certainly be a different story since now we have more opportunity to bluff, but we do have the dosadvantage of always being out of position.

To reiterate a point I made a little while ago, with poker you should generally think of hands (or poker spots in general) as being good, bad, and marginal, and you shouldn't devote too much time and energy to the marginal. A hand like 74o from the small blind, with several limpers in a live game (presumably bad players), is marginal. Most players would say to fold 73o off there (its bad), and most (I think) would say call 76o (good). A 75o is also probably marginal.

The idea however that a hand as good as K9s is a "mistake hand" is completely ridiculous. I don't mind when poker authors err on the side of caution when giving preflop advice, but there comes a point where its so nitty that it is simply bad.

I realize I'm not providing actual evidence or math as to my declarations of what is good, bad, marginal, mainly because I'm a bit against the idea of preflop math in order to determine profitability, unless stacks are at least close to going in. There are just too many postflop scenarios to consider when its not shallow. Your best bet is to just follow collective wisdom as much as you can, and use your own stats once your database gets larger.

@just_grindin

Looks like our math is yielding the same result. You came up with the figure of needing to win another $15 postflop (let's assume $1/2) on average. But when you win $15 postflop, you're also picking up the $9 preflop. So winning $24 breaks even.

That's the same number I came up with. Assume you play 25 hands. 24 of them you lose $1, so you're down $24. Then if you win $24 on the last hand, you're breakeven. Win $25, and you're +EV.

I agree about the flaws in the model. And I think that analyzing those factors is really what this question is going to come down to. It just seems to me like the positive factors will outweigh the negative ones. Two positive factors you didn't mention are the implied odds, and the fact that sometimes the flop checks through, or you do happen to have implied odds to make a +EV call on the flop.

The empirical data is a really good point. I think that also is a hugely important thing in this discussion. There's a big caveat though. People who are playing these hands from the SB may be making mistakes with them postflop by investing too much extra money with a mediocre hand. So it'd be really interesting to see the data from someone who we somehow knew was playing correctly.

On T44, I think you can expect two calls from Tx hands usually, two calls from a flush draw if present, and one or perhaps two calls from a pocket pair like 77. But that's just T44. On something like A44 or K44, there's a lot more top pairs that you could get say two streets of value from.

@NMcNasty

I see where you're coming from. I'm tempted to think that too about going with collective wisdom when the math is tricky, but here I think I might disagree. I don't think the math is actually too difficult to approximate. And I think that it is worth exploring at least.

I also agree with the general idea of not spending too much time and energy studying marginal spots. That's why I justified never spending time studying this spot. Here, my justification is just curiosity.

I disagree the positives outweigh the negatives. I doubt we'll be able to prove it one way or another but intuitively if you have a really strong hand on the flop there are several reasons this could cut into your implied odds, especially since you have to win an addition amount that is nearly twce the current potsize. Mostly because your opponents don't have hands good enough to pay you off.



I agree that any players with a significant sample of hands would likely have played them badly, since good players are less likely to play these hands at all.

However any good players that have played the hands are likely to suffer from sample size issues, where the results are highly unfluenced by luck.

But think about this - even players that play these hands in the BEST of situations (i.e. from any position with all strategic options available) will likely have a winrate for these hands hovering around 0. So it seems highly unlikely that if we restrict ourselves to just calling and "nut peddling" we'd be able to do significantly better than that.


I think boards like 244,344,544,644,844 and A44 actually pose some problems for 74o in a limped pot. I agree that 944,T44,J44,Q44, and K44 are probably some of the better boards for it.

@just_grindin

You think everyone is folding on [2-6]44 boards? If so, then we should have +EV bluffs on comparable boards like 633. My impression is that you can get some calls from pocket pairs, flush draws, and some overcards if you don't blast too hard. It's the kind of board where it looks just so hard to hit that some people might not give you credit. And then there are people who are the type to want to fight for these pots who you can get in to a level-y raise game with.

The only issue I see with A44 is that people now aren't calling with pocket pairs and there are no overcards. But that seems pretty balanced by the fact that someone is likely to have an ace, and they'll usually be willing to give you at least two streets with it.

Let's look at some other textures.

• 77X - Seems pretty similar to 44X. I think you can get calls from pocket pairs, flush draws (when possible), top pair when X is big, and when X is small, nonbelievers and people who want to fight for the pot.
• 74X - Sometimes the board will be very connected and there will be a lot of draws you can get calls from. When X is low, you get the nonbelievers and people who want to fight for the pot, plus weaker stuff being willing to put money in. When X is high, you make money from top pairs.
• 865 - There are some pair+draws that will want to call, and any sets or two pairs you make a ton of money from. There shouldn't be any overpairs since the pot was limped, and I could see stuff like A6 folding. I also don't see people fighting as hard for the pot, but something like T9 or 54 could get into a raising battle with you.
• 653 - Still some pair+draws. Two pairs and sets of course. But now I could see the nonbelievers and aggros fighting for the pot.
• 774 - Obviously this one is tough to get value on the flop. But if you check, someone else could pull the trigger. And even if it gets checked through, someone is bound to improve to something on the turn.

Didn't read all the responses so sorry if it was already stated.

The biggest problem I see is position. You are always first to act. You need to flop your 2 pair and be up against a strong pair or a good draw. Can you win pots by showdown? Of course but in most situations no one is paying you off. Board come 4 7 9 rainbow who is paying you off on that board? AA KK QQ JJ TT are probably not in the hand since it was limps all the way around. So you 2 pair mine your 74o and catch it.

Are you firing out on the flop?

If so who is calling (79, 44, 56, 77 maybe if someone would limp that, A9, A7, T8, T9, etc etc)? Either no one calls and you win a small pot the small percentage of the time you catch a hand or you will be up against a better hand.

In other words you are missing 96% of the time and winning small pots when you hit. You need to win $26 the one time you hit the 2 pair for this to be profitable.

I think there are 2 problems. You don't get many calls in a limped pot because not many hands are good enough to call when you bet into 3 other people.

I also think that you run into better hands more frequently on those boards than the other boards because pairs with 22-66 are more likely to limp and hands with a 4 are more likely to contain 3's and 5's and to a lesser extent 2's and 6's and those hands show up more often in limped family pots. Also A4 is more likely in limped pots as well because it's an ace and people love playing A high hands.


I am sure it's not unheard of but aggressive opponents don't usually limp preflop and go crazy post in my experience. You are right you might get calls from pairs, overs, etc. You may even get multiple streets. I am just not sure on average all things considered that means your hand makes 7.5bb over the long haul.

Agree that overall Ax will compensate for running into say A4 or 4x better kicker. But you might also have to split with lower 4x depending on the runout. Again I understand that what I am describing happens rarely but it also happens to be when I think significantly more money enters the pot.


I guess my overall point is that I think you are drastically over estimating the intersection of how often we actually hit the board, how frequently another player has a hand to pay us off, and how much money players will invest to do so.

Do I believe there are tables with opponents bad enough this strategy might work? Sure. I just think in general it's not likely to be profitable vs any regular pool of players.

Also we've completely ignored the fact the big blind has the option to raise behind you so you may not even see the flop as often as you're expecting so you're just adding an additional -.5 bb to your EV some non-zero percent of the time.

I should also note that I could be completely biased in the other direction that I am over estimating how often we win only the pot or get our money in bad :-)

@just_grindin

All good points. That makes sense that that on the boards we hit, we have to be particularly worried about someone hitting harder, because people play those small pairs and stuff. And it's also a good point that people give you more credit when you bet in a multiway pot. I also agree about players who limp generally not being the type to get aggressive and fight postflop.

I think it's just a matter of us having different intuitions, and like you said before, it'd take a deeper dive into the math to really figure it out. Anyways, thanks for the conversation

@Lozgod

I talk about my thoughts on how we make money on different board textures in comment #7 on this thread.

But to address your comment on 974, that does seem like one of the harder boards to get value on. 9 is low enough where there aren't as many top pairs as there would be if it was like an A or K, but scary enough where weaker stuff might fold out. Still though, with four opponents, it's reasonably likely someone has a 9 and gives us a street or two. And if we're talking 1/2, you could run into people who overplay something like 87 or J8 with a backdoor flush draw. And even if we're talking stronger opponents, then we'll face bluff-raises more that we can make good money from.

I agree with all of that in theory and if those boards and situations happened like that the result would more than likely be as you predicted. Odds are it will not though.

Easiest way to proof this is to take hands you would consider then run them multiway vs typical multiway limped hands through Equilab and see what kind of equity you have.

I've called 74s on the button before in a multiway raised pot just because the size of the pot and flopped 3 5 6 rainbow and got paid off vs UTG's AA (also got called every Donk, Fish name in the book in the chat box lol).

I think it literally takes a situation that drastic to make it work. Problem with the limped multiway pot is unless it is UTG with the plan of limp-raising AA or KK that backfires on him, no one is holding a strong enough hand to give action when you do hit a hand.

For it to be profitable you would need to consider post flop fold equity with air and weak pairs.

I think you're overestimating how often flopped bottom two is going to win in a million way limped pot. Even just top pair has decent equity against you and the board is generally only going to get worse for your hand.

That said, at live 1|2 there are players who call flop bets in 7-way limped pots with A-high, under pocket pairs, naked overs, etc, so given play that bad completing any two might actually be justifiable.

Trying to win pots because you have “good odds” is like chasing a carrot on the end of a fishing line. You always have just enough equity to call but your almost always at a range disadvantage and/or never end up making your hand. In the end your just leaking chips.

The goal of the sb is to soften to loss you’re already gonna take. Adding marginal spots to help this might just make it worse.