Thanks again for all the great comments on this week’s What’s Your Plan?. I was a little worried since the question was more open-ended than usual, but once again we had a lot of really though-provoking contributions. I actually write about this hand in some detail in the new issue of Two Plus Two Magazine, so I’m mostly going to refer you to that for my thoughts and address a few of the comments, starting with Prabhat:
“Putting this person on a range is very difficult as a result of his unusual stats. I can’t help but think that if this person really plays 18/17 over an adequate sample size, its virtually impossible that he flat-calls a lot of suited connectors here. If he calls T9s, he also calls JTs and 78s etc, and this already makes his flatting range much wider than 1%. I find it tough to believe that he doesn’t flat anything except from the CO and Button. Accordingly, I will slightly discount the suited connector portion of his range. ”
It may be difficult, but I’d say you nailed it! And really, shouldn’t it be easier to enumerate a narrow range than a wider one? I think you’re right that a narrow calling range will consist almost exclusive of set mines. As Dana points out, his 1% cold calling range is an aggregate statistic that is almost certainly higher when he’s BTN vs UTG+1, but how wide can it be? Even {QQ-66,AQs+,KQs,QJs,JTs,T9s} is 5% of hands and probably a little too wide, so it’s tough to put many non-pairs into his range.
For those of you who can’t or won’t read the article, the short version is that I advocate check-calling all the way because it’s extremely difficult for Villain to have a draw. It’s rarely correct to slowplay on a board this wet or to focus on inducing bluffs with a hand this strong, but in this case so much of Villain’s range is so weak that I think it’s correct just to try to induce bluffs or ideally get him to hit a lower set.
Jonny argues that, “If he expects us to check AA or KK on this board then he is presumably less likely to bluff. ” He’s less likely to one-barrel, that’s for sure. Particularly if the board gets scary, it would be very reasonable for him to think that AA/KK will fold by the river. Meanwhile that first barrel can get me to fold the bottom of my range, like AK, Ad Jd, and small pairs that I’m just check-giving up on this flop but that have good equity against Villain’s small pair.
As Prabhat and others point out, checking probably results in us stacking 88 less often. For the same reasons, though, we probably get stacked less often by QQ and T9. Granted 88 should be a larger part of his range, since we’re not certain he shows up with QQ or T9 at all, but on balance we’re actually not a huge favorite if Villain is willing to get all-in on the flop. Even being generous about the number of combo draws he could have:
Board: 8c Qs Jc
Dead:
equity win tie pots won pots tied
Hand 0: 52.599% 52.05% 00.55% 7729 82.00 { JdJs }
Hand 1: 47.401% 46.85% 00.55% 6957 82.00 { QQ, 88, AcKc, AcQc, AcTc, KcQc, QJs, T9s }
A lot hinges on whether he can have T9 or QQ, since subtracting T9 bumps us to 60% equity, and subtracting both gets us to 80%. Then again, if has neither, he probably doesn’t have those other suited connectors much either, meaning that we rarely get action at all when we bet. I think we make nickels on the dollar getting all-in on the flop, whereas we have near 100% equity against underpairs, so inducing them to bluff should be the higher priority.
Indeed, as Shawn says, a bet on this board will look very strong: “I think another thing to consider is our perceived hand range also to Villain. If we bet the flop I think he can take out AA, KK, TT, 99, AQ, KQ as we don’t really want to build a huge pot OOP against such a wet board with one pair/gutter hands. If we’re UTG+1 raising pre and betting the flop into such a wet board vs a button caller I think that narrows are range down to nut flushes with overs and sets although maybe not 88.”
This is an instance where Villain probably doesn’t even realize how much information we have about his hand. Rather than bet and give away so much information about our own, I think it’s better to exploit the leg-up we currently enjoy by giving some Villain some rope to represent hands he’ll rarely if ever have.
I know I described Villain as “not very tricky or creative,” but I don’t think that precludes him from taking a few stabs at what looks like a very weak line. I do think it makes him less likely to float or bluff-raise the flop, so in this case checking seems like the best way to induce bluffs from his very weak range:
PokerStars No-Limit Hold’em, $4.00 BB (9 handed) – PokerStars Converter Tool from FlopTurnRiver.com
CO ($685.70)
Button ($449)
SB ($402)
BB ($527)
UTG ($400)
Hero (UTG+1) ($434.35)
MP1 ($400)
MP2 ($412)
MP3 ($206.95)
Preflop: Hero is UTG+1 with J♠, J♦
1 fold, Hero bets $16, 4 folds, Button calls $16, 2 folds
Flop: ($38) 8♣, Q♠, J♣ (2 players)
Hero checks, Button bets $25.25, Hero calls $25.25
Turn: ($88.50) 6♣ (2 players)
Hero checks, Button bets $59.85, Hero calls $59.85
River: ($208.20) A♦ (2 players)
Hero checks, Button checks
Total pot: $208.20 | Rake: $3
Results:
Button mucked 4♠, 4♦ (one pair, fours).
Hero had J♠, J♦ (three of a kind, Jacks).
Outcome: Hero won $205.20
Edit: Whoops, didn’t mean to publish this yet, but I guess the cat’s out of the bag now. The 2+2 Magazine article should be up later today, I’ll add a link when it’s live.
Edit 2: Link is now live.
Wow!
Just blown away by the quality of your plan! I think I just realized the extent of flaws in my thought process when playing monster hands. There is no way I would have extracted two streets of value from the bottom of a competent player’s range on this board. I learned a lot from this, so thank you!
Thanks, Prabhat, but don’t read too much into this. It is almost always correct to fastplay your monster hands and to target the top rather than the bottom of your opponent’s range. Most importantly, though, you need to think about what his range is. In this case I’m not sure his range has a top, and if it does, we’re not in great shape against it, which is why I played the hand so strangely.
Thanks Andrew, I have been reading through your blog recently, but a concrete example like this was really good for reminding me of the dangers of being too robotic or automatically making standard decisions.
I have some issues with this:
1) I think you are relying too strongly on stats here, and not necessarily correct stats either.
What’s his call raise from BTN %?
What’s his bet vs missed c-bet %?
If he is seriously 18/17 and literally only has a 1% flat-call range, then yeah, he won’t have suited connectors in his range.
However, I can almost guarantee his call raise OTB% is higher than 1%. I’ll say he’s bad/nitty and folds 78s or lower, but pretty much everyone is calling T9s-AQs pre here. I think excluding them from his range based on his overall pf stats is a mistake.
Enumerating that range (doing this on the fly so it might not be 100% accurate)
-AQs 3 combos
-KQs 3 combos
-JTs 1 combo
-QJs – 1 combo
-ATs – 4 combos
-T9s – 4 combos
Also counting hands he likely has in his calling range pre-flop, that are continuing on this flop:
-TT – 6 combos
-99 – 6 combos
-88 – 3 combos
-AQo – say 4 combos to be nice
So overall we’re looking at ~35 hand combos (give or take 5) that are continuing on this board. Granted some of them are betting when checked to, but many turn cards can kill our action vs these hands, whereas they may continue to call if we bet.
Now to enumerate his bluffing range
-22-77: 30 combos
-AKo: ~4 combos
We’re basically looking at a 50/50 split in terms of hands that can call a bet vs hands that are likely folding if we bet.
So, do you think you make more on avg by extracting maybe 1-2 streets vs a bluff than you do vs 50% of his range that is calling at least 1 street?
I would argue that you do not.
True, you do make money from both his bluffs and valuebets by checking, but I think you lose more on avg by not betting vs the top of his range, esp since he’s RARELY 3 barreling here with absolutely no equity.
I know you know all this, but I think your assumptions about his PF range given his overall stats, and not stats by position, are over-reaching. I know you’re an advocate of normally fastplaying on a wet board, but given this exact texture and the PF positions, I don’t see a reason vs this villain to deviate.
Basically, I disagree his range is as ‘weak’ as you describe it.
All of that being said i’m actually very impressed w/ your line given your justifications. It’s very creative.
Just a point on counting combos for betting vs trapping:
99 isn’t calling a value bet on the flop because even the Tens are likely not clean outs, though we do have to factor giving a free card.
T9s has us beat so its kind of irrelevant from a getting-value perspective.
It is much more easy and objective to say who is rigth about the range.
You just need large database with 2/4 6 max games.
You execute 2 simple sql queries:
@sum int
SELECT @sum =Count(*)/100 Where ButtonPocket IS NOT NULL AND PREFLOPACTION_BuTTON=”Call” AND V=18/17 AND PREFLOPACTION_UTG+1=”Bet”
SELECT Count(*)/@sum as percent,ButtonPocket Where ButtonPocket IS NOT NULL AND PREFLOPACTIONBUTTON=”Call” AND V=18/17 AND PREFLOPACTION_UTG+1=”Bet” GROUP BY ButtonPocket
The result of this query will give you range.
This is very rude query which operates on agregate 18/17.It is much better to play not with agregates but actual ratio for Button vs UTG+1.
If you have very large database you can include more parameters like Andrew agregates.
Wow me too. The hand was just too complex for me I knew from the beginnig.I knew that the plan will be even more complex to construct.
It was the reason I wished to close the deal on flop and I expected that my opposition will do the same for the same reason with reasonable equity.
I just read your article in 2+2.
Your wrote about opponent’s pre-flop range how “your perception is immediately colored by the flop texture”.
100% agree.
But I see another psychological bias which color out perception before we see flop.
The fith element is our hole card. And I do not mean exclusion our pocket from his range which has statistical bases.
The hole card is able to polarized our read.
The tendency to put our opponent on the scariest hand.
Or wishfully thinking that our opponent holding a just notch weaker hand.
The cumulative analysis of both factors could give even more insights.
“It’s best, if possible, to think about your opponent’s pre-flop range before you see the flop”.
My point was that maybe better before hole cards so you will give better focus to decipher 18/17 and and implication for preflop-range.
Maybe this is it. Those successful (2/4)bots on Ongame,IPoker excellently and exactly did month ago.