How to play middling top pair facing a 33% pot river donk on a wet card?
Live $1000 buy-in tournament. Four players remain (of about 50) and we are in the money.
PREFLOP
Folds to hero (27bbs) on the button who has Q♦9♥ and opens to 2bbs. Villain (69bbs and chip leader, clearly a very strong player) calls in the SB. BB folds.
FLOP (6bbs)
6♣5♥3♥
Villain takes his time and checks, hero bets 2bbs, villain calls quite quickly.
TURN (8bbs)
6♣5♥3♥Q♣
Phew. Villain checks, hero bets 6.5bbs. Villain again takes his time and calls.
RIVER (21bbs)
6♣5♥3♥Q♣7♥
Villain tanks and bets 7bbs. Hero...?
What are the other two stack sizes?
Not as important with 4 left and ITM as it would be earlier at a final table (assuming you're ITM and each elimination is a pay jump), but against the chip leader I think we really want to be playing pot control here. If we're shortest it doesn't matter as much, but I suspect that's not the case.
What are the other two stack sizes?
Not as important with 4 left and ITM as it would be earlier at a final table (assuming you're ITM and each elimination is a pay jump), but against the chip leader I think we really want to be playing pot control here. If we're shortest it doesn't matter as much, but I suspect that's not the case.
Cutoff had 33bbs, BB had 23bbs.
Your hand is now a bluff catcher. He will have the flush or a 4 a decent amount of time as you filtered him two street. Bet sizes on flop and turn is also too large. tpwk in this board is just a middling hand.
I'd check flop and bet more like b33 on the turn. We don't want to play big pots against this opponent unless we have a hand we're happy to stack off.
The SB calling range is going to be pretty specific and not as wide as a BB defense range. I don't think your flop bet is really getting the big chip leader to fold any of it.
I decided to go in depth and look at this hand closely in solvers.
Some key takeaways:
Preflop: You have to tighten up significantly due to ICM. Q9o is a break even open and is mixing folds in this configuration, with the big stack in the SB.
Flop: At equilibrium the big stack is actually leading range after calling in the SB, mixing between small and large sizing. That's despite their range being weak and them being OOP. Even with less than 1% of their range checking, we are supposed to check back 82% of the time with our entire range.
If I node lock for SB to check range, we respond by checking range 100% ourselves.
Turn: Both as played and also after checking back flop, Q9o is pure checking back the turn. When we do bet our range favors mostly betting with a small sizing.
River: As played our exact combo is calling with the 9h, presumably because of the relevant flush blocker. Pocket kings without a heart is mixing folds so the call is not purely due to our hand strength.
The above was based on a HRC preflop sim and a GTO+ postflop ICM sim, which I based on a standard top 4 payout structure from another tournament.
The postflop ICM sim is pretty eye-opening. No one is playing these spots anywhere near optimally.
The postflop ICM sim is pretty eye-opening. No one is playing these spots anywhere near optimally.
Yeah, I'm sure I'm not either, but even what I've learned from ICM studies at final tables, we really strongly favor keeping the pot small when we're covered, unless we flop something like TPTK on a dangerous board and want to shut down the action. We push even small edges when we're playing chip EV, but it's the exact opposite at a final table; when we're covered (and presuming we aren't short enough to just get all-in preflop or on the flop), we really don't want to push the action without a substantial equity advantage.
I'm also not surprised Q9o mixes folds here. Not just because we can really run into trouble when the SB decides to play the pot, which seems like the most obvious reason, but also because, when we don't play this pot, we give the BB, currently the shortest at the table, a chance to lose more chips to the big stack.
The postflop ICM sim is pretty eye-opening. No one is playing these spots anywhere near optimally.
Yeah, I'm sure I'm not either, but even what I've learned from ICM studies at final tables, we really strongly favor keeping the pot small when we're covered, unless we flop something like TPTK on a dangerous board and want to shut down the action. We push even small edges when we'
Good point about the potential advantage of folding pre and allowing for the big stack and big blind to collide, possibly resulting in an opportunity to ladder with no risk.
Also regarding post flop, in practice your suggested way of playing the hand (checking back flop and B33 turn) may still be the best approach. The solver playing as the big stack is extremely aggro, and will check raise aggressively even after checking twice (to crank up the ICM pressure).
Against most human players we can likely get away with making a small turn bet after checking back flop. Once the big stack checks turn following a flop check back, they will likely not check raise as aggressively as a solver.
I guess the counterpoint that would suggest we should be checking back both flop and turn is because if our opponent is not leading appropriately as the big stack then we actually need to play even more cautiously when they check. That's since their checking range will be stronger if they're not leading so many of their strong hands... That sort of effect was demonstrated when I node locked for them to check range on the flop, how we went from checking back 82% of the time to checking back 100%. Anyway just thinking out loud here. These final table ICM spots are complicated and nuanced compared to chip EV spots.
Regardless of the nuances, my conclusion after reviewing the hand is that we should have at least checked back either the flop or the turn. I don't hate the flop stab against many players who likely overfold and will not leverage their big stack appropriately. However if we do bet flop we need to check back turn for pot control. The big bet on the turn after betting flop was likely a pretty big mistake, albeit an easy mistake to make given how the spot plays so much differently compared to how it would without the ICM pressure.
Grunch
Interesting hand
SB range capped but stronger than BB defend, so reasonable to assume neither player has range advantage on 653 flop
2bb cbet isn’t folding any part of his range. Weak part connected and strong part still sees sdv v BTN open. Either check or go larger and much prefer check when we’re in no mans land like this. Oh and icm.
Turn as played yes sure but we haven’t defined SB range any further so it’s risky. At this point I’d rather have checked flop.
River w/e there were better decisions to be made on earlier streets. We have no idea what SB has.
Good point about the potential advantage of folding pre and allowing for the big stack and big blind to collide, possibly resulting in an opportunity to ladder with no risk.
Wanted to mention it because it's one of the hidden advantages here - it's easy to see direct factors like "big stack in SB can put pressure on us" and "we have to raise/fold and that might happen a lot here," but the second-order factors aren't always as noticeable but are still worth keeping in mind.
Also regarding post flop, in practice your suggested way of playing the hand (checking back flop and B33 turn) may still be the best approach. The solver playing as the big stack is extremely aggro, and will check raise aggressively even after checking twice (to crank up the ICM pressure).Against most human players we can likely get away with making a small turn bet after check
Yeah, definitely prefer this vs. a human, who is going to check twice to check-raise a lot less often than a solver would, unless you're playing some of the very best (and while a live 1k isn't nothing, it's not where you find those players). Assuming that's the case, I also like it because we should be able to mostly check back the river (although in this scenario we got the donk lead).
I guess the counterpoint that would suggest we should be checking back both flop and turn is because if our opponent is not leading appropriately as the big stack then we actually need to play even more cautiously when they check. That's since their checking range will be stronger if they're not leading so many of their strong hands... That sort of effect was demonstrated when
Yeah, I like the flop check back for a number of reasons - and SB leading "out of turn" less often than a solver would is one of them, but it's not just that, it's also what oldsilver and I mentioned - between the board, the SB calling range, and the bet size, we're folding out very little, and of course we want to keep this pot small.
Regardless of the nuances, my conclusion after reviewing the hand is that we should have at least checked back either the flop or the turn. I don't hate the flop stab against many players who likely overfold and will not leverage their big stack appropriately. However if we do bet flop we need to check back turn for pot control. The big bet on the turn after betting flop was li
OP described the SB as a strong player, so I think we want to assume they are good at leveraging the big stack, if not necessarily playing solver-perfect.
Seems like an easy river fold everything got there. He would have to be turning something into a bluff to make a call here, and you could easily have a flush/straight
We don't want to play big pots against this opponent unless we have a hand we're happy to stack off.
what I've learned from ICM studies at final tables, we really strongly favor keeping the pot small when we're covered
.
The key preflop is how aggressive the SB and BB are. If they are frequently 3-betting then Q9o is a fold for me. If SB is calling a lot in SB against raises I probably fold as well because of stack sizes. But with 27 bb's against players who won't 3-bet I have no problem raising preflop.
I prefer a check back on the flop against a chip leader who just called preflop OOP. They often do that with small PP's like 22-77.
I don't mind the sizing on the turn at about 65% pot (Nath is right that pot is now 10 bb's). With top pair we will get a lot of folds especially against gutter draws. Also since 2 flush draws are potentially out there we may not be wanting to bet the river. The advantage of checking back the turn is that the board will get there over half the time for a flush and/or straight and there would be a lot of river bluffs for hands that missed so it would be a good calling spot on the river given we would be under repping our hand.
I would pretty likely fold the river unless this guy is an overbluffer, because so much got there (straight draws (4x), flushes, 77/76/75 and 66/55/33 were already there though there might have been a c/r earlier). In theory SB may be betting smallish (at about 30% pot) as sort of a blocking bet with a hand like A6, but the sizing is too big for that purpose. In reality the river donk is most likely trying to get a call from a hand like AQ/KQ/QJ by making the price right (we only have to win ~25% of the time to break even). I'd prefer not to drop below 10 blinds...
ICM to my knowledge is taking a specific spot in a vacuum without consideration for future EV when winning said pot - the bigger our stack the more +ev spots we can create vs a small stack, and ICM I don't believe factors that in at all. Here we turn a card that's great for our range terrible for the caller - typical barrel spot regardless of our holdings, and should have a larger more polar sizing in theory. I think this is even more profitable given our small cbet sizing keeping villains range wider then if we bet larger.
ICM to my knowledge is taking a specific spot in a vacuum without consideration for future EV when winning said pot - the bigger our stack the more +ev spots we can create vs a small stack, and ICM I don't believe factors that in at all.
This is all well and good until you lose the big pot playing like this and lose far more equity than you would have gained from winning it.
Stacks are close enough that you're going to be shortest by a good margin if you lose a big pot here.
I don't know that going from 27 to 35BB if it works out changes that much for us, but dropping under 20BB and to the clear short stack certainly does.
I decided to go in depth and look at this hand closely in solvers. Some key takeaways:Preflop: You have to tighten up significantly due to ICM. Q9o is a break even open and is mixing folds in this configuration, with the big stack in the SB. Flop: At equilibrium the big stack is actually leading range after calling in the SB, mixing between small and large sizing. That's despit
Wow, these are really interesting findings, thanks for doing the analysis.
Cool to see the SB leading on this board, which I guess is very good for his range. Interesting too that hero is meant to mostly check back on this board if SB is nodelocked to check range. I guess that's because there will be a lot of check-raising?
Really unintuitive that KK without a heart is folding to a 33% river donk. I think this almost never happens in "reality" - even strong players are not folding overpairs vs that size (of course).
Well, here's the hand as played:
https://youtu.be/wO_cIY2kNcA?list=PL2bAZ...
It was such an interesting hand that I thought it would be fun to post it as though it was played at much lower stakes and see what y'all thought 🙂
Don't think that's the correct link.
Regarding the turn lead and the IP player mixing folds with KK on the river, I am assuming both of those are due mainly to ICM effects. This was a postflop ICM sim I ran and it was also based on ICM-adjusted preflop ranges. We have a lot of incentive to protect our chips and avoid busting before the other players, so the big stack can push us around.
Maybe if I have some free time this weekend I will run the hand again as a chip EV sim to compare and contrast. I suspect a chip EV sim would include many fewer flop leads from the SB, and the IP player would likely never be folding KK on the river vs the lead. Also the turn barrel on the Q would likely be standard without the ICM effects.
Just looking at limited postflop ICM sims it becomes clear that even many of the super high roller players are not as well studied at this aspect of the game as they are with many other aspects. It actually gives me hope that if I work on my game I may be able to actually compete at those levels one day 😀 Even the best players in the world are playing far from theoretically perfect.
Yeah, the link takes us to a playlist with like 45 videos. If that is the correct link do you have the title of the specific video and a timestamp?
Regarding the turn lead and the IP player mixing folds with KK on the river, I am assuming both of those are due mainly to ICM effects.
Highly unlikely villain donk bets 1/3 pot on the river for value with worse than KK. So blocking the value hands becomes more important than the raw strength difference between Qx and KK. (I guess Qx also blocks two pair, but I don't really see queens up in SB's range on this board.)
Sorry, got the link wrong. This is the correct link:
