Hand Analysis

PokerStars No-Limit Hold'em, $0.02 BB (5 handed) - PokerStars Converter Tool from http://poker-tools.flopturnriver.com/Hand-Converter.php

BB ($4.50)
UTG ($3.07)
MP ($1.33)
Hero (Button) ($2.91)
SB ($5.79)

Preflop: Hero is Button with 8, 8
UTG raises to $0.06, 1 fold, Hero calls $0.06, 2 folds

Flop: ($0.15) 6, 9, 2 (2 players)
UTG bets $0.06, Hero calls $0.06

Turn: ($0.27) 3 (2 players)
UTG bets $0.12, Hero calls $0.12

River: ($0.51) 2 (2 players)
UTG bets $0.48,

Villain was 48/16 over 32 hands. Was getting a bit suspicious that he was bluffing me I always seemed to have a hand that was in my checking range vs him and it'd check down to the river and he'd take a stab and I kept folding. In an earlier hand I was BB he was BTN and I check raised him on a 98x tt flop with 67s he called flop. Turn comes an 8 checked through. Then I bluffed an ace river and he called with A3o, so putting him on a range was difficult.

I did a bit of a HH analysis:

My preflop range: JJ-22,AQs-A2s,KTs+,QTs+,JTs,T9s,98s,87s,76s,65s,AQo-AJo

Villain's preflop range: 22+,A9s+,KTs+,QTs+,JTs,T9s,ATo+,KTo+,QTo+,JTo

I decided to play my range like this:

Flop:

Raise: 78s

Call: 22, 66-JJ, 89s, T9s,

Fold: everything else

If villain's cbet range is: {99+,66,22,AQs+,A9s,JTs,T9s,AQo+} Then 88 has 39.3% on the flop and my best calling hand is JJ with 52.4% and my calling range as a whole has 55.47%

88 seems like a shitty hand to be calling with with only 39% equity when villain has 36 combos of missed broadways. I didn't think villain would be attacking my calling range since he had checked down to the river a few times before before he bet. So I thought he would check most of his unimproved broadways on the turn which is why I called.

Turn:

Raise: 66, 99, 22

Call: 88, TT-JJ, T9s, 98s

Fold: everything else

Villain's turn betting range: {99+,66,22,A9s,T9s,AsKs,AsQs,JsTs,AhKd,AsKd,AsKh,A cKd,AcKh,AcKs} A 9 or better all turned FD's and AKo half the time.

88 only has 16%, JJ has 40%, and my calling range as a whole has 39.17%

Calling the turn looks terrible now.


River:

Raise: 66, 99, 22

Call: JJ

Fold: everything else

JJ only has 45.16% vs villain's betting range which I estimated to be the same as it is on the turn.

Bet/(bet+pot) = 0.49% JJ isn't even a +EV call with that range when it beats 18 combos draws with 1 and is beat by 19

I'm confused, I don't know the best way to tackle this guy. He called a check raise with Ahigh before so bluff raising on the flop seems like a bad idea, but if I narrow my calling rang to make it stronger I'll probably be folding too much to his bluffs. Actually check-raising and raising his cbets with a wider range for value would be a good idea if he's calling them with rags, is that the best strategy?

I dunno, if you think we have 18% on the turn vs his range, I suspect you're giving him too tight a range. For example, why is he barreling all his AK but not all his AQ? I would have thought we have over 40% vs his range by turn. These bets of his are weak and I'm leaving the majority of his Ax in his range, plus some of his broadway. I'd say we're probably ahead of his range on the turn based on his sizing. Yeah ok his river sizing makes me reconsider, but he's been winning pots at river by stabbing and I'm looking him up here to refine my notes if anything.

He called a check raise with Ahigh before so bluff raising on the flop seems like a bad idea

Indeed, bluffing flop would be bad. How about raising for value?

Originally Posted by OngBonga

Yeah ok his river sizing makes me reconsider, but he's been winning pots at river by stabbing and I'm looking him up here to refine my notes if anything.

Indeed, bluffing flop would be bad. How about raising for value?

Paying 24bb roughly 1/4 of a stack for information seems like a bad idea and bad reasoning, I'm sure it would have a significant impact on your win rate over the long term. However I think calling 6bb on the turn with the hope of getting the same information via a free showdown on the river is ok, but I think a fold on the turn is best.

I just added AQ+ to his flop c-bet range to represent the fact that he'd be bluffing the flop sometimes, but not a disproportionate amount of the time and missed overs are his best bluffing hands on such a dry flop, if my range is accurate that is.

Adding more hands into my raise for value range was probably the correct adjustment to make after I saw him call a check-raise on the flop with rags. Pitty I never thought of that in game.

i'm with ongie on this . WHy not raise flop for value. you called pre so low flop is gonna look like you either hit tP on flop with 89s or T9s or a 78s giving an open ended straight draw. your 22,66 and 99 hit their sets. if you raise and he reraises then an easy fold. if you raise he probably folds. you don't really want to see any overcards coming on later streets to hit his range so why not try and take it down now?

Paying 24bb roughly 1/4 of a stack for information seems like a bad idea and bad reasoning

It's not costing me 24bb for information unless I always lose. And it's not just for information, it could be for value, I can't know this right now. I don't think it's possible to determine if a river call is +ev or not unless we have some idea of what his river bets mean, and we're not gonna find this information out by folding. Calling will be hugely +ev if he's bluffing often enough. What I meant by that comment was that I'm prepared to make a potentially -ev river call to get some idea of his tendancies, which will help me to determine if later spots vs him are +ev or not. If it's -ev, it's marginal enough that in my opinion it's worth the information.

Couple questions for op:

How often does villain need you to fold the flop with this sizing for cbetting any two to be profitable?

What percentage of the time are you folding the flop with the range you detail, once card removal is taken into account?

Originally Posted by The Bean Counter

Couple questions for op:

How often does villain need you to fold the flop with this sizing for cbetting any two to be profitable?

What percentage of the time are you folding the flop with the range you detail, once card removal is taken into account?

He needs me to fold 44% and I'm folding 77.5%.

I don't think this is the whole picture though. Whenever flop and turn checked through when I was OOP he would stab river, this is the first time he'd taken a bet-bet-bet line. There was no reason to think he was bluffing a lot here, although with his sizing I felt compelled to call in game.

Regarding river:

With his sizing to be bluffing optimally he needs to be bluffing pretty much 1/3 of the time. I don't think he is bluffing any where near 1/3 of the time in this spot with that sizing with a bet-bet-bet line.

If he was bluffing optimally I would need to call 0.48% to avoid being exploited.

My river range in order of strength:

22 (1)
99 (3)
66 (3)
JJ (6)
TT (6)
T9s (3)
98s (2)
88 (1)

For a total of 25 combinations. I would have to call my top 12 combos then to avoid being exploited by optimal bluffing. So I would have to call with 22, 99, 66 all of the time and JJ around 83% of the time to avoid being exploited by optimal bluffing.

However, I think he's weighted more towards value on the river so I would need to adjust by tightening my calling range.

He would have to be bluffing significantly more than 1/3 of the time for me to add 88 into my calling range.

Originally Posted by seven-deuce

He needs me to fold 44% and I'm folding 77.5%.

I don't think this is the whole picture though. Whenever flop and turn checked through when I was OOP he would stab river, this is the first time he'd taken a bet-bet-bet line. There was no reason to think he was bluffing a lot here, although with his sizing I felt compelled to call in game.

Regarding river:

With his sizing to be bluffing optimally he needs to be bluffing pretty much 1/3 of the time. I don't think he is bluffing any where near 1/3 of the time in this spot with that sizing with a bet-bet-bet line.

If he was bluffing optimally I would need to call 0.48% to avoid being exploited.

My river range in order of strength:

22 (1)
99 (3)
66 (3)
JJ (6)
TT (6)
T9s (3)
98s (2)
88 (1)

For a total of 25 combinations. I would have to call my top 12 combos then to avoid being exploited by optimal bluffing. So I would have to call with 22, 99, 66 all of the time and JJ around 83% of the time to avoid being exploited by optimal bluffing.

However, I think he's weighted more towards value on the river so I would need to adjust by tightening my calling range.

He would have to be bluffing significantly more than 1/3 of the time for me to add 88 into my calling range.

This looks good to me. You might have to discount some sets if you're raising flop ever, and maybe even some overpairs if you're raising the flop ever. In which case, your river call down might have to be slightly weaker, but yah 88 should probably be a fold either way.

One way see if your 3 street call down range is close to GTO, is to look at what your cold call % would be in this spot. Then based on his flop and turn sizing as well, see what % of combos you need to be defending on flop, turn and river. See what this number works out to.

Originally Posted by seven-deuce

He needs me to fold 44% and I'm folding 77.5%.

44% doesn't look right (although currently drunk). Shouldn't this be 1-bet/(bet+pot)?

I do think your wide pre-flop calling range gives you a problem here tho if you want to get close to being unexploitable.

Originally Posted by seven-deuce

I don't think this is the whole picture though. Whenever flop and turn checked through when I was OOP he would stab river, this is the first time he'd taken a bet-bet-bet line. There was no reason to think he was bluffing a lot here, although with his sizing I felt compelled to call in game.

Yeah, this is one of those things I don't quite get with gto. Griff/sav/other gto inclined can probably answer this, but presumably if we "know" villain Cbets x% on a certain flop texture, we should be weighting our defence accordingly vs a given Cbet sizing? If so, we would probably see our unexploitable call down range as being very few combos by the river.

On the river it should be, his bet needs to work 48/(48+51) = 48.4%, so you need to continue with 51.5%.

If we're adjusting our call down ranges, based on villains cbet tendencies, this is fine but we're obviously exploiting villain in this case and not trying to play GTO.

Originally Posted by The Bean Counter

Shouldn't this be 1-bet/(bet+pot)?

I do think your wide pre-flop calling range gives you a problem here tho if you want to get close to being unexploitable.

Yeah, this is one of those things I don't quite get with gto. Griff/sav/other gto inclined can probably answer this, but presumably if we "know" villain Cbets x% on a certain flop texture, we should be weighting our defence accordingly vs a given Cbet sizing? If so, we would probably see our unexploitable call down range as being very few combos by the river.

He needs me to fold more than 28.5% of the time to have a profitable bluff with atc. I don't know where I got the 44 out of.

I've never thought about the breadth of my cold calling rang pre-flop affecting how exploitable I'd be post-flop. I was calling so wide because he was playing passively IP vs me up until the river so I assumed he would be just as passive OOP so I could win a lot of pots simply by having position on him.

Regarding using the alpha value on the flop to work out your ranges. Is this correct? I don't know anything about working out optimal ranges on the flop or turn but it seems to be a misapplication of the alpha value using it for that purpose. Since you could be folding more than the alpha value amount on the flop but when you continue villain keeps putting money in vs your strong range so defending the required amount according to the alpha value would actually reduce the overall ev of your cold calling range.

The common theme with all of these examples of the alpha value being applied incorrectly is that implied odds, reverse implied odds and future bluffing opportunities usually change our total expected value a lot more than what’s happening on the present betting street. Because the alpha value is only a shortcut for basic, single-street situations under specific limitations, it cannot be applied as anything more than an extremely rough and borderline-useless guideline in the majority of other situations.

http://www.flopturnriver.com/poker-s...ha-value-17369

Note: I didn't read your dissertation, I'm only commenting on the hand.


Pre: Vs someone loose and donkish like this there's a good amount of value in 3-betting preflop IP. You get some good isolation and you build the pot with a hand that has nice equity vs his calling range. You'll probably have to take a lot of pot control lines post flop when you flop an underpair, but you'll still be in a lot of profitable situations. Anyway 99% of people call there and its fine, I'm just saying look at alternatives.


Postflop: The flop and turn play themselves, you're getting excellent odds to see the next street and he's probably a more aggressive than average donk. River is the blankiest river ever so just call. His entire line is suspicious (small bet small bet big bet on superblank) so expect to see total air or a busted AK/AQ a lot. And if he has 9x or better, oh well.

Originally Posted by griffey24

If we're adjusting our call down ranges, based on villains cbet tendencies, this is fine but we're obviously exploiting villain in this case and not trying to play GTO.

This makes sense. I was wondering if we could combine the two as a rule of thumb though e.g.

- Let's say we flat 100 combos
- Villain cbets half pot, so GTO would tell us to defend 67%
- But we estimate villain will only cbet 50% of the time on a given flop texture

Could we then crudely use:

100*67%*50% = 33 combos

as a rough guide for our defence?

Obviously the assumption about cbetting frequency is pretty unreliable..

Even if cbet is for sure before any decision you still need to consider his position, flop texture, chance he fires turn again on different cards, equity to river vs the cbetting range, etc. My opinion is that a defending range vs cbeys is more complex then just cnet stats.