We'll just get the solver solution out of the way. Solver is pure jamming this over the raise (lol wtf) and snap calling turn.

Hand History driven straight to this forum with DriveHUD 2 Poker Tracking Software

NL Holdem 2(BB)
HERO ($203.50) [VPIP: 28.9% | PFR: 24.1% | AGG: 36.7% | Flop Agg: 41.7% | Turn Agg: 33.8% | 3Bet: 11.5% | 4Bet: 14.2% | Fold to 4Bet: 56.9% | Hands: 313415]
BB ($200) [VPIP: 48% | PFR: 40% | AGG: 41.2% | Flop Agg: 33.3% | Turn Agg: 50% | 3Bet: 8.3% | Fold to 3Bet: 50% | 4Bet: 0% | Hands: 26]
HJ ($340.06) [VPIP: 26.9% | PFR: 15.4% | AGG: 0% | Hands: 26]
CO ($265) [VPIP: 24% | PFR: 16% | AGG: 37.5% | Hands: 26]
BTN ($215.65) [VPIP: 40% | PFR: 0% | AGG: 66.7% | Hands: 5]

Dealt to Hero: A K

HJ Folds, CO Folds, BTN Folds, HERO Raises To $6, BB Raises To $16, HERO Raises To $40, BB Calls $24

Hero SPR on Flop: [2 effective]
Flop ($80): 7 2 7
HERO Bets $19.25 (Rem. Stack: $144.25), BB Raises To $64.50 (Rem. Stack: $95.50), HERO Calls $45.25 (Rem. Stack: $99)

Turn ($209): 7 2 7 7
HERO Checks, BB Bets $95.50 (allin), HERO ?

X/C flop. What does the c-bet in a 4-bet paired flop accomplish? You're not getting any folds, so you're just increasing the pot size in a spot you're hoping to get to SD or turn a card. As played, you're better off jamming turn than x/cing of course because your opponent is almost never jamming worse. So once you check, it's a fold. Solvers are fun, right?

Well the one thing I am 100% sure about is the flop cbet haha.

We get called from so many worse hands if BB is defending correctly.

Here is the OOP solution (range bet)



Look at what BB has to defend vs 1/4 cbet to NOT overfold. A solver is only folding 12.7%! Our cbet prints!

You need 25% to call turn, you already have 12% vs TT so if he shows up with random hand from time to time you have odds to call off.

You can 3bet smaller BvB.

Jam makes sense being OOP, it's mostly an equity protection thing. We're ahead of all the bluffs that have to fold their equity, and we have decent equity vs the TT/99 that calls

Makes even more sense in practice if villain is raising wider than the very slight amount the solver is

Yeah I think you are right with the call it off.

I don't think you are right on the bolded part, have you seen GTO Wizard Sims for 3--->9bb? They go bigger not smaller.

GTO is raising 15%-20% and population is raising 10%-13% so don't think that's a great argument. The equity protection thing makes sense to me though.

I just meant to say that it makes even more sense to jam if a villain were to be raising more often than the solver in this spot, no clue if they actually would be

Didn't it say that it's raising flop here 6.3%? Or do you just mean flop raises for 4bps in general for these positions

Oh yeah definitely agree with that. I was looking at overall frequencies so you're right that wasn't the best stat. Let me check SBvsBB population

I don't know what GTO is in these positions but based on the overall GTO data vs population we can probably make an educated guess that they are under raising.



4bet pots are by far the worst part of my game, it's interesting that the solver only raises 6.3% here, I thought it would be way more than the aggregate on a low-mid paired board.

Totes... I'm sure you guys are seeing tons of peeps also defending correctly by calling 4bets w/ K5s-K7s, K9s, T8, A7s, Q9s, K9s, 55-77, and so on.

So in this hand are we trying to GTO solve, or MDA solve?

I'd do the flop cbet and call the turn jam... don't mind x/c over jam

idk either, I think that's a safe assumption tho. But based on your data it seems like fish specifically will likely significantly overraise vs small sizings if we range bet, so maybe jam is better vs fish but worse vs regs

There are some spots in 4bps where IP will have no flop raising range whatsoever, but this is basically only versus range bets at 10% sizings from what I've seen. Not really sure why, definitely not very intuitive to me

I'm trying to play the hand in the most profitable way possible, usually that's a mix of GTO/MDA.

You're right they do overfold preflop vs the 4bet but there's a few flop MDA reasons why we want to cbet instead of x/c.

1. Solver says it's higher EV to flop cbet (all thing's being equal just play GTO by default if you don't know anything else)

2. BB still slightly overfolds to flop cbets even though they under defend preflop.

3. BB probably under raises relative to GTO frequencies (I only have overall not position specific raising frequencies but it's well under GTO)

I do think I made a mistake here in folding OTT, curious to see what IP had here and to get some more data on the spot.

Vs Fish you just call them down with a hand like this. You are basically ranging small to induce and then call them down with Ace high.

If you look at fish frequencies for R-B OTT. Fish are at 80% and GTO is in the mid 60%s. Fish are also way wider preflop so we just bet small to induce and then print vs their bluffs.

I get that the standard play by everyone for quite some time is 1/4 c-bet in 4-bet pot. Everyone is following what the solver says. But...

1) That's because the range the solver is defending with is way out of tune w/ what's actually happening. So of course, if you can get your opponent to call w/ worse w/ that sizing, it's going to be +EV.

2) Really? I can't make any sense of that... they defend 4-bets perfectly, but they don't understand they shouldn't be folding ATC to that sizing? And if they do over fold, that's a stronger case for X/cing, because you want your opponent to be betting w/ worse, not folding worse.

3) Probably a safe assumption.

I've thought a lot about this spot when people started posting hands a few years back on solver sizing in missed 3 and 4-bet pots, and I just couldn't bring myself to resolve the difference between the ideal and what actually occurs. And looking at my whiffed AK in 3 and 4-bet pots, it's hard to deviate from what's working. That said, I want to remain open to better reasoning.

WRT to #2. No they don't defend 4bets perfectly, they overfold preflop. But they also slightly overfold to a cbet as well.

I think one of the main problems with the x/c line is that population as 4BP IP PFC under stab relative to GTO. I have BB stabbing at 44% over large samples and GTO is mid 50%s. So when we X as SB here we just end up giving BB too much equity realization with worse hands that X back.

I'm open to deviations from a solver but I don't think the flop is the spot to do it.

If they overfold hands like TJ that is good for our hand because those hands have odds to call vs AK.

Betting is bad if you are getting raised a lot. Check can be good if he bluffs a lot but most opponents won't blas of on A or a K, so you just end up haveing tricky spots when you miss and he keeps barreling.

On a site like Pokerstars, when you don't have data, betting an indifferent amount is fine. But when you're on an anonymous site where you know the population tendencies, we should be exploiting the mistakes to yield the highest EV. Is c-betting the flop in a tunnel +EV given what we know? Yes, is it the highest EV line? Probably not, but it depends on some data about the population.

We already established they over fold to 4-bets. I'm surprised the 4-bet stabbing % isn't higher than GTO honestly. If that is the case, then sure, it may be higher to just c-bet 1/4... my data doesn't say that, and I have some from 200nl+.

Really we should be looking at this as a two street solution right? Based on that, what line yields the highest exploitative EV based on what we know? And what are we trying to exploit to yield the highest EV? That our opponents are over folding, over calling, or over bluffing.

Our lines:
X/C-X/C
XRAI
B-B
B-XC


You post this because you wanted to understand this spot better, because it's common enough that a deeper dive is going to be helpful, correct? Which of those does your population bluff more? And conversely, which one do they overfold/overcall and by how much?

I'd node lock some of those ranges based on that, and see how our EV changes.... or not.

ez call, could probably even stack off flop for value
don't over think it

not even close. I just looked at MDA data for this spot and vs 15% cbet they are 15 weak and for 30% sized cbet they are 22 weak.

Also throw this hand in there as well.

Results

Hand History driven straight to this forum with DriveHUD 2 Poker HUD and Database Software

NL Holdem 2(BB)
HERO ($203.50) [VPIP: 29% | PFR: 24.1% | AGG: 36.7% | Flop Agg: 41.7% | Turn Agg: 33.8% | 3Bet: 11.5% | 4Bet: 14.2% | Fold to 4Bet: 56.9% | Hands: 313675]
BB ($200) [VPIP: 48% | PFR: 40% | AGG: 41.2% | Flop Agg: 33.3% | Turn Agg: 50% | 3Bet: 8.3% | Fold to 3Bet: 50% | 4Bet: 0% | Hands: 26]
HJ ($340.06) [VPIP: 26.9% | PFR: 15.4% | AGG: 0% | Hands: 26]
CO ($265) [VPIP: 24% | PFR: 16% | AGG: 37.5% | Hands: 26]
BTN ($215.65) [VPIP: 40% | PFR: 0% | AGG: 66.7% | Hands: 5]

Dealt to Hero: A K

HJ Folds, CO Folds, BTN Folds, HERO Raises To $6, BB Raises To $16, HERO Raises To $40, BB Calls $24

Hero SPR on Flop: [2 effective]
Flop ($80): 7 2 7
HERO Bets $19.25 (Rem. Stack: $144.25), BB Raises To $64.50 (Rem. Stack: $95.50), HERO Calls $45.25 (Rem. Stack: $99)

Turn ($209): 7 2 7 7
HERO Checks, BB Bets $95.50 (allin), HERO Folds

i don't care about MDA, this is a +50bb call, you would need a crazy deviation for it to shift to -EV
I guess go to solver and nodelock slowly removing bluffs and see at what point AK folds

Where are you getting a +50bb call from?

MDA show's how people actually play poker, a solver is just a complex toy game. You shouldn't care what a solver says if population is playing nothing like that.

Calling AK here is going to be a big losing play if they are only 15%-22% weak when they jam turn. We need them to be at least 25%weak to breakeven.

That's assuming that AK has 0% equity vs strong and 100% vs weak... AK has 12% vs REALLY strong, 50% equity vs same hand which is still strong here, and around 90-95% equity vs weak
With a very pessimistinc MDA based equity calculation you will see how this is worst case scenario a decent call best case scenario an absolutely printing call
It's an absurd fold, and you should definitely not throw solver out the window without justifying it, and a 50bb EV loss play is basically not justifiable through MDA, it could only be possible to make such a play with an insanely specific read

So you are saying if we know for fact that BB only has 15% bluffs here, we should still call with AK because we probably have 6 outs to the nuts?

Where is the 50bb EV loss coming from again?

50bb EV loss comes from the fact that you have a value beater, he doesn't have 15% bluffs here from the very MDA that you showed. You used 25% size which is 2/3 of the way closer to 30% than 15%, so if anything you should estimate weak to be around 20. Open flopzilla or do some equity calcs with a rather pessimistic range and you'll see.

Also what's even the sample on your MDA? this is going to be a stupidly low sample spot. I don't even agree with your approach, but even using your approach I wouldn't fold