5/10/20 deep stack against good player

Hand is played fine. I just fold river in practice. What is someone realistically betting 400bbs with OTR? Turning AQs into a bluff? Potting single pair for value in the biggest pot either of you have played in your life?

Might even be a fold in theory. Ranges are extremely narrow and we’re on the outside looking in at a boat party.

I would agree that V can size his 4bet down but H can also size his 3bet down and if this is just the game dynamic where 3bets/4bets go on the larger size then it's not very notable.
there is an undiscussed element in the hand which is why H is 3betting HJ so large to begin with. if it's because HJ calls too wide for big size and H is trying to isolate HJ that influences the dynamic, if it's because HJ folds too often and H is sizing up to maximize fold equity and steal the pot that influences the dynamic. if this is a standard open/3bet/4bet sizing for this game then that's just the standard game dynamic.
in a 3 blind game there's more value to stealing preflop than in a 2 blind game. overall ranges might tighten but they're also played more aggressively. 88 can easily slip from cold call to 4bet range here because the 3 blind dynamic influences the aggression. it's certainly never a fold out of the straddle.
3 blinds + H's move to isolate/steal vs HJ who was still involved in this hand but kind of dropped out as being relevant in any discussion of it can incentivize V to select from the wider 4bet candidates to try to resteal. V's history vs H also means he's going to be a lot more comfortable in this hand playing a large pot in position vs H who hero folds.

its hard to talk with you because you don't really have anything to justify why you think things. your proof for things is just you think that they're right. i dont think straddle is meant to cold call much because hj is going to be able to 4b ~15% of the time and also he's going to end up sandwiched postflop quite a bit. it just doesn't make much sense for him to vpip from an ev / pot odds standpoint, and the burden of mdf isn't on him. if sb is a whale (possibly given this thread), by all means but for the most part i haven't really seen evidence to suggest coldcalling here is part of an actual strategy. the whole conjecture and philosophizing of why people are doing whatever isn't really all that relevant and is generally what people do when they don't know how to come up with an answer. i'm not sure what u mean by more value, when u raise pre to steal the blinds u are just laying odds that u take them down. the open size usually goes up and there's another player whos going to vpip ~10% of the time so idk man. the actual ev if it changes is likely like 1/100 of a bb

the thing is when people aggress the biggest pots of their life, they either always have it or always dont. most have it, some dont

the issue too is i think most peoples range as op is {KK}

your proof for things is running a 1000 straddle 3 blind hand through a 300bb 2 blind sim and seeing that the solver said V made a sizing mistake pf so deciding the entire hand is him torching money.

This logic only makes sense if you know the board is going to be QJJT. Otherwise, why would you want to fold out hands like KJ and AJ? Why would you be so concerned about running TPTK into a set?

Stack depth makes us more inclined to have a flat range, not less. If we were 100bbs deep, we’d mostly just ship whatever hands we continue with. You can’t be that merged this deep.

We might not ever need to call here because of their size, though…

I don’t disagree here. I think the problem here is how most people are playing the rest of their range.

*I* feel comfortable folding here because *I* know *I* can easily have a boat here.

Anyone who feels married to their hand cause they think they’re top of range is a skill issue 😃

Against a normal size, 20/20/60 Raise/Call/Fold is probably a good starting place, though optimal is probably closer to 50% fold.

So a little less than 2% total, but it shouldn’t be perfectly linear.

Against this size specifically, should definitely strive to 5! at least 25% and flat as little as possible.

because 1) we probably have the best hand vs this V and putting in more money when we're most confident our hand is best is fundamentally sound and 2) playing this deep we want to define our opponent's range as much as we can, as soon as we cede control of the aggression we lose that opportunity.

it can also be fine to play this hand as a trap, which I said originally, H basically played this hand as a trap and turned broadway in a spot where V probably isn't crediting him many combos of AK so river should be a call as played. I don't like playing AKs as a trap though, especially not super deep.

It's not that I'm "so concerned" about running TPTK into a set it's that playing AKs as a 5bet simplifies a lot of different flop scenarios considerably and removes opportunities for us to get "coolered" or get put in the blender like in this hand. I also just don't believe we're running into a villain in a casino 1 out of 10,000 times who is going to 6bet bluff so let's just take the hand to the 5bet level preflop.

I think I've only seen a 6 bet bluff once in my life and it was Chino Rheem playing with someone else's money.

Just saw results and lol, the addage “if no range makes sense facing a mid stakes live pro, then there’s enough weird spew to make a call worth it” wins the day!

I was trying to be disciplined and not a station for once, but the part of OP where you said “he plays exploitative against you” should have been the clue to blow this case wide open.

I should be clear that I 5bet AKs at least 50% of the time here. But so long as we have a flatting range (again, thats maybe possibly not the case when we face a 1.5p raise OOP) AKs is a fine hand to include at some frequency.

my proof is looking at solver and deriving things from similar situations.

without looking at a solver, i think villains' river play is a combo that shouldn't be bluffed (almost certainly wants an ace) and is probably losing a good amount with this sizing assuming op plays the spot well. its also iffy if he can use large sizings when he is capped more than op

re akss. come on. at 1k bb its a bluff and we are praying the guy folds lol

its really difficult if you won't look at a solver to see what ranges actually look like here. i see bb cold 4bing 2.2% of hands, you aren't getting value from that range, this spot is much tighter preflop from sb and straddle than u think it is

im fine with looking at anything you want to see how to approach the situation, but it's like we're doing a geometry proof and you can't actually show why what you believe to be true is true. your intuition and anecdotal heuristics (we should 5b so it doesn't come k85r and we get stacked!!!) are worth literally nothing when it comes to analyzing the spot. you're just starting from a conclusion and telling me the conclusion is the reason your assumptions are true, but you have no actual reason to believe the conclusion is valid beyond you don't study and it's what you feel. that can be fine in soft games or maybe it works for you, but i don't think it actually helps anyone get better / learn the spot

i do think its a reasonable question to start asking at some point is there a stack depth we just don't want to 3b good people oop

Insane hand because of how deep you are.

I’d 5bet preflop if I thought he was 4betting too wide.

As played you just need to think about what his bluffs are which goes back to preflop.

Individual reads are the most important part of the hand and if you cover him and he has 20k then it means you have been playing for a long time. Only you can answer the question to your post because vs an unknown it’s a really easy fold as this is almost always QQ/JJ/TT. It doesn’t matter if it’s only 5 combos as it is such a tight range spot preflop.

you're being disingenuous because solvers don't support 1000bb play. what do you want me to do, force monkersolver to solve for no limit and then wait 2 weeks for it to chew through a decision tree? tell you I bought a private solver from a siberian bot farm and I inputted the hand and this is the results I got?

inputting the factually incorrect information into a sim and then autistically refusing to consider any input from somebody actually taking into consideration the hand that was actually played, not a hand that kind of sort of looks similar in a 6max 300bb sim is worth literally nothing.

No.

I’m gonna say something that might seem crazy, but stack depth doesn’t matter THAT much.

Like, obviously it matters when stacks get shallow enough relative to the pot that it changes how the hand is played on a functional level (eg: OOP can realize all their equity by 5b shoving pre, SPR is low enough you don’t need to bet all 3 streets to get it in, etc). And obviously once players have put 500bbs+ into a pot that’s going to put us in unique situations with extremely narrow ranges with fragile strategies.

But the difference between being 500bbs deep or a million bbs deep is not going to have a notable impact on our RFI or 3! strats. It might sound naive to just be like “I’ll worry about what to do facing a 10,000bb river shove once I actually get there,” but I mean yeah, that kinda is how you should approach it.

(Unless villain is a whale, or course, then you should constantly be gunning for as much of their stack as you can.)

i don't know why you're getting angry at me. i'm pretty simple, i dunno who anyone on the forum is and i dont really care. i'm willing to look at anything that can actually be proven or verified, and usually when people type paragraphs that have nothing to do with math / ranges etc its generally pretty useless in terms of application. but your random thoughts and name dropping people you played with 15 years ago isn't really enough of a credential for me to abandon wanting some shred of proof of what you're saying.

unsure about this when it comes down to spots we're 3betting < 5% of hands and going to be oop a trillion bb deep. my guess is equity realization decreases (non linearly) the deeper we get if villain is good / better than us. at some point that has to be a problem. if you take a look at ev's in gtow in the research mode, the only hand i see a substantial ev change between flat and raise as sb 300bb deep facing 2.5x from mp is AA (granted its a large difference) but the rest of our 3betting range is indifferent between the 2 options

what this means idk really. i got the idea because i think matthew moss? had a blog on here when he was grinding macau games and talked about if you wanted to have a 3b range bb vs btn vs a good player, and that always stuck out in my mind. have no idea really practically speaking and i would just 3bet but i can see a world where if you're playing vs actually really good people you're going to end up giving back whatever ev u gain preflop, post and can potentially make back some of the ev by having a stronger flatting range than they are used to dealing with. obviously would still squeeze and 3b vs later positions blah blah

also i think stack sizes matter alot between like 20bb 50bb 100bb and then around 2-300 it becomes a somewhat different (even) more polarized game. i havent looked at any 1kbb sims tbh but the actual hand is just spr 8.5 with very slightly tighter pre ranges then cold4b at 300 if i had to guess. am gonna get conned into renting a server or aisolve for gtow for this hand, i see this coming from a mile away

ok cool I bought a private solver and inputted this as a hand and it told me H should be 5betting AKs here at 85% frequency and that when played as a call vs this line V's value range is QQ, JJ, also AKs, and the solver likes 99-88 as bluffs. It's running on a machine with 1 TB of DDR5 RAM and thinks faster than you so you can't argue with it.

think we've probably reached the end of our discourse

I should be clear I’m talking more theoretically. If villain has a considerable skill edge on you, then maybe you just minimize damage against them HU. If this is what you meant, then I guess you can stop reading here :P

Not sure about 3betting <5%. I’m 3betting 7%+ here if I’m linear. FWIW, Snowie is too.

Those EVs are assuming you’re playing the same exact strategy with your entire range. When you change your strategy, those EVs change and different actions become highly differentiated between different hands.

This is a common confusion with comparing the EV of different combos, especially on early streets and especially OOP, and it sends a lot of conversation into some sort of poker equivalent of moral relativism where nothing you do matters either way.

In any case, at equilibrium, significantly improving the EV of your nutted range is the main driver of the EV difference in taking an aggressive line in the first place. You’re forcing your opponent to pour money into the pot against your clear value range as often as possible to realize equity, and much of the rest of the range is geared toward indifference.

So it’s not intuitive but also not surprising that one hand alone (which in your example makes up a full 10% of your range) is single-handedly driving the EV difference between two actions.

Poker changes significantly between 20bb, 50bb and 100bb insofar as you frequently run into situations where things are functionally different. When you RFI 20bbs deep, players in the blind can easily jam on you. A 3bp 50bbs deep has a similar SPR to a 4bp 100bbs+ deep, which changes strategies significantly. Etc. Even so, things like initial raising ranges aren’t TOO crazy different. BTN RFI is SLIGHTLY tighter 20bbs deep and UTG’s range is less differentiated from later positions, but I find people who ignore this factor are performing closer to equilibrium than those that treat it like a whole different ballgame.

What makes this spot different from 100bb poker is that we couldn’t just jam over his 4! pre, and the fact that we’re facing 3 big bets in a 4 bet pot in the first place. Everything else is pretty negligible.

i get that re the the nut combos, was just surprised to see how little kk / akss were worth but fair point re moral relativism

the stack size thing my point was i think it hits some arbitrary depth inflection point the sims are going to look mostly similar. i have no idea if its 200 or 300 or 500 but at some point the strategies are going to look functionally identical. is sort of w/e because i feel like the other guy is arguing in bad faith but i thought this actual situation (super deep pre) was interesting

post the snowie out put. pretty much everything i see on gtow > 200bb is mixing flat and 3b from sb. closest i see is 6.2% in 200bb 6m 2.5x open w low rake, but im trying to look at like 4 different pre solves here and trying to kind of combine conclusions from them. it looks like we definitely get tighter from 200 -> 300 but not by an overwhelming amount

in terms of 3b % in gtow in research mode (mixes flat and 3b)

for 2.5x cev (no rake)
100 bb - 6.9%
200 bb - 4.8%
300 bb - 4.4%

for just 200bb general ranges (which uses 2.5x open)
200 bb - 6.2%
100 bb - 7.4% vs mr 8.1 vs 2.5x

am not exactly sure why theres that large of discrepancy between the solves (the research mode solves got added when they bought ruse so id think they are more likely to be correct)

oh. it looks like in the pre only solve at 200 / 300, the 3b size is larger (15.5 / 15bb vs 11). they have research mode for pre where it lets it use few diff sizes and it uses all of them so would think the 15ish size maybe best)

one interesting thing is it seems to prefer larger when 3betting the earlier opens i guess bc more strong / less linear and shows preference for smaller size vs the lp opens from the sb at 300bb. can see this in research mode where it gives it 4 diff sizing options pre and it mixes them vs the ep / mp opens from sb but strongly prefers the smaller size vs co / btn

Sorry to derail thread from sub and polk’s beef lol

I would personally 5bet 1k straddle deep, honestly this deep, 3bets are like opens, 4bets are like 3bets and 5bets are 4bets etc…

As played i think its fine to fold river, before seeing results, i was thinking what if dude just turns 99 into a bluff but then i was like nahh that would be wayyy too overbluffing if he was.

I dont think its a good bluff by V, on 85% of rivers he loses huge (if he triple barrels) and he has no relevant blockers, like 99 would make more sense cuz at least he blocks J9s , T9s etc, 88 is just bad a bad bluff imo.

I honestly wouldn’t think too much about it. The only thing i wouldve done different is 5bet pre. The rest of the hand is fine imo.