5/10/20 deep stack against good player

Haven't read home thread, but have you seen him cold 4b really light? Like with A5s? And follow through with big bluffs? Then I think I lean towards sigh calling river.

Most suss part of villain’s strat is preflop. If your preflop ranges are reasonable, then anytime you somehow show up on this river without at least top pair, it’s reasonable to just bluff with it.

But the chances of you having a reasonable range is pretty low if you’re showing up with 88 here. You can sprinkle in a tiny bit of board coverage hands here but with ranges this narrow you’d literally have to roll like a 99 or 100 on a 100 sided die to go for it in theory. And any sim I’ve seen prefers either SCs, Axs or Kxs to hit random cards.

That last part might be one thing that changes this deep because obviously I’ve never looked at a sim where you’re deep enough that smashing a set actually matters in a 4bp…

Thanks for sharing.

I’d have to play with it myself or at least see the ranges but it SEEMS to me the lower frequency is due to larger sizing and more polarization rather than an aversion to aggression. IME, if you lock in one size and eliminate calling as an option, the range will remain around 7-8%. It’s just that other options have emerged that are (perhaps marginally) higher EV for your whole range.

The fact that we go with more polar strats the deeper we get does suggest I’ve misunderstood (or wrongly assumed) something about the theory here:

I thought we play more linearly OOP because we HAVE to—ie: we can’t get away with making super light 3bs OOP, (which is something that would only become MORE true the deeper we get).

But your research suggests maybe we play more linearly OOP 100-200bbs deep because we CAN get away with pushing more marginal equity to put us in more frequent AIPF (or at least low SPR) spots where we can realize our equity well. It’s always be clear to me that this is the case for 4bs because OOP 4bs more than IP, but I just thought that’s because it was SO close to AIPF that IP’s only aggressive line is to shove, which nullifies positional advantage altogether. But your research suggests it’s true of 3bs too!

Obviously as stacks approach infinity, OOP can no longer realize more equity by forcing shorter effective stacks with merged hands, so it makes sense OOP would give up on this and just start doing 2x PSRs with polar ranges.

I mean sure. Its a bunch of low and mid stakes players guessing at a spot that is over our head and that literally nobody knows the answer for. And we are dealing with such narrow ranges that if one combo plays it this way occasionally, that is enough. Im not saying he plays JJ this way every time, or even that its the most common. I dont see why youd play Jx like this, it makes no sense to make a river bet that is almost never being called by worse. You need a hand that beats Jx to jam river imo. But again, i have no idea nor does anyone else on planet earth including V id think.

Also as far as all this preflop solving ****, I actually have a fast computer with piosolvers preflop solver on but I cant give an actual solve because after buying it and screwing around with it i determined im too stupid to use it. If one of you two wants to tell me how the hell to run it, i can run it to see who wins this dick measuring contest lol.

I hope the dick measuring contest you’re referring to is between submersible and PugDolk, cause this is how the conversation is going, at least in my head:

I’ll hit you up in DMs about helping with the sims.

I think every big pot that gets posted on here at low stakes is “I called with the 3rd nuts only for villain to showdown the nuts, could I have gotten away from my hand” and every big hand that’s posted on here from T/20+ is “I made a big fold only to later find out villain was button clicking, should I have called?”

So if you want to look smart in these forums and get validated by results you can just say “fold” to every low stakes hand and “call” to every mid-to-high stakes hand, but I’m not sure that’s a winning strat at the tables.

Haha very true

It won’t solve a 1000bb hand that’s my entire point. None of the commercially available solvers will. You’d have to buy a private solver and believe whoever made it that it works.

You can run a 300bb sim but it’s not the same thing. This hand is an easy illustration of that. Putting in 5% of your stack with 88 preflop as a raise is not by itself a mistake. Putting in 18% of your stack is.

It’s also a three blind hand. Three blind incentivizes players to play more aggressively, with a (slightly) tighter overall range yes but more aggressive ranges meaning some of the better hands in a calling range slip up to the weaker hands in a raising range. The blinds are incentivized to 3bet to iso more because the straddle gets laid such a good price and also has position on them. This is like a very obvious alteration to the game I don’t understand why I have to explain further. When you look at standard blind vs HJ 300bb stack 3bb open ranges they’re not taking into consideration with the calling range and 3bet range that you have another person on your left that only has to call $40 to play for a pot of $165/$170 if you call.

This also incentivizes straddle to play a more aggressive 4bet range. Because the range is more aggressive overall some bluffs also move up into it for balance. A5s doesn’t give you enough bluff combos anymore. 88/99 is probably a wash in terms of EV because 99 blocks more relevant straights postflop but 88 unblocks hands like T9 that make the RFI range weaker and just fold to a bluff. It’s very minor difference either way.

You can disagree and that’s fine but claiming it’s objectively, mathematically provable to just “simplify” the hand by running a sim with a third of the stacks, 2/3rds of the players, and missing a blind bet is ludicrous it’s not the same hand, it’s not even close to the same hand, it’s not even the same game.

The fact that players are playing more aggressively just means they’re more closely approximating the sims. A lot of these assume practically raise or fold strats for both SB facing a raise and BB/straddle facing a 3b cold.

The problem isn’t that it’s not accounting for enough aggression. The problem is the narrowness of the ranges (which as you say, will be even narrower in a 3-blind game). Straddle topping out at 2.5% 4b at most, and we hold a top 1.25% hand, so we’re in the middle to bottom half of their range.

(Doesn’t mean it’s bad to 5b here; I think I do 75/25 raise/flat, but just that we should expect them to very often continue and end up in an even deeper pot still OOP against hands that even dominate or are racing against you.)

Now another question is whether in practice people might “defend” their straddle too light, even if that means barging into a pot cold with a 56bb raise lmaooooo

I already said my piece on stack sizes, but gotta say pointing out that we’re deeper here than in the sim just seems like sowing plausible deniability into the conversation. The fact that being deeper makes this more of a mandatory 5b defies all logic and theory I can dream up.

Yeah in extremely deep stack situations AKs is doing a lot more flatting vs 3bets and 4bets in sims I've seen, like 300bb+. At a certain point AA is kind of your only value hand and you might be flatting that sometimes too. If anything AKo is more likely to do a mix of 5betting and folding facing a cold 4 and low frequency suited AXs like A5s.

I'd be wary of thinking villain is actually playing a strategy close to solvers. Solver is going to fold 88 almost always, probably folding AKo and JJ a lot too. The fact that he shows up with 88 here probably means he is cold 4betting way more than he should in theory. Unlikely he is only 4betting because he somehow rolled a 95 in live poker.

idk this was pretty interesting one i think

if you get over the whole "i would never call w worse than a jack" and u look at how much equity Jx has on this river in a cold 4b pot, i think it's going to be a clear value bet from an equity standpoint. you beat KK / AK / slowplayed AA / who knows what else op is doing. i think you would need monker not pio to run this because its multiway

again my thing w my why i think river if its a value hand its jx (and why i thought it was possibly a bluff) is he chooses a large sizing that doesn't actually maximize from unfoldable coolers while making it likely he doesnt get raised by jx / tx. like let's be real, op is rec vs guy he thinks is gto wizard in 1000bb cold 4b pot, do you think he is ever folding Jx / Tx otr to any bet? probably not.

i don't understand your arguments. because we don't have access to 1000bb solves, we shouldn't look at anything and instead take your word as gospel. how does that actually help anyone improve? realistically, getting the right answers to the hands posted matters exactly 0 in the scheme of things beyond some kind of ego validation. they're mostly just gateaways to conversation and avenues to getting better at poker (in theory, in practice of course its just submersible arguing with people).

you say this stuff is obvious game alterations, but how much of a difference do you think it makes in things like rfi ranges and sb (in the straddle equivalent bb) and bb (straddle) 3betting ranges. you're making it seem like it's a completely different game when its probably like a 1% of hands shift.

the biggest thing i think people itt don't get (i don't just mean you) is how tight the sb is supposed to be here re 3betting as he gets deeper w ep

got cut off from posting. but interestingly, still at 300bb, it prefers larger size vs ep / mp opens ~15-17ish and smaller size ~12ish vs lp opens (research mode lets it have 4 different sizes of 3bs). makes sense bc 3b range vs ep is supposed to be so strong i guess.

my hypothesis (based on it shrinking from 200 to 300) is it goes down non linearly as we get deeper until we either get to some % of hands that it maintains forever, or we just get so deep that oop vs a super computer our equity realization goes down far enough that the ev gained by 3betting is no longer there. no idea which and maybe im wrong anyways. i do think the people in this thread looking at like 10+% of hands are wrong though in either case

Some thoughts

1. How deep is HJ? Such a large 4 bet might commit v to HJ shove. So that information is important to figure out his range.

2. I do like a 5 bet pre. Calling the 4 bet with AK is to trap his AQ, AJ, KQ etc. But you can still 5 bet some AK. AKs is a good hand for that. The only question is what do you to facing a 6 bet or 6 bet shove? That depends on dynamics, very v dependent.

3. I also like a flop raise. Basically, versus someone like this ..you have 2 options..either play passively with the intention of calling down all the way OR rebluff him a lot and hope he blinks first. The worst thing you can do against such an aggressive player is build a big pot, out of position, and then overfold river.

4. As played, I would think he's trying to get us off a chop a lot. It's just hard to call down 400bb with your hand ..so I understand the fold. But it's a lesson in understanding people and their motivations.

When you see someone get this out of line, just realize that you can't fold a good hand to him anymore. He's just way out of line too often.

And in my experience, players who have this gene of being way too out of line, will not correct themselves when you adjust. They just can't help it.

the sim is going to be super reflexive to ranges but at least initially it looks like bb cannot do what he did if he is trying to make money.

am using ai on gtow, but turn looks like high frequency lead for oop for a large sizing, and if i lock that to pure x he is supposed to x back pure. his size is not preferred but if he uses it, its like a 2bb mistake in my sim to bet 88 so it will never do it (consequently i can't see river ev loss for what he did), 7 if oop is nodelocked to pure x. facing this action vs a balanced range ak is making 70bb (this will vary in your sim as i just ball parked stack sizes but my river pot is 81 bb and facing 79bb bet w 161 eff stacks) via calling and is mixing raise and call lol. interestingly, ip mostly checking back AK otr (which i think is a function of stack depth and if he goes big he just funnels oop down to mostly str+). punt on his end but would think folding AK is costlier. this is like a really impossible sim to look at ev's for though bc if i don't give oop the option to lead he just isn't supposed to bet the turn. river also will never be pot, itll be all in or x. my guess would be the evs are going to be too finicky to really look at for individual hands bc narrow ranges so will change dramatically based on pre frequencies etc blah blah

it should be somewhat clear, his advantage is mostly in AA and KK and those hands are really not worth that much (vs cold 4 calling range) on this turn. he has ak sure, but so do we and the number of hands he can actually get value from are not that many. i think people itt are conceptualizing sb as going to the flop with like a 3bet calling range or something but im not sure you want to do that

have no idea how accurate this solve is and its one of the few times ive used aisolve on gtow so gl

i think to unsophisticated person they're just like ok T completes all my bluffs so i can blow him off of range, but sb just really really tight here bc pre so if u go all in w ak u find urself vs a range that u dont perform great against (as ip). in practice, i think (as said earlier) i think he just decided to bluff pre and couldn't stop himself post. or who knows maybe hes amazing and op just gave off bunch of live tells or something.

was cool thread though and have learned quite a bit re deep dynamics. i think in most midstakes games, people still simplify sb as 3b or fold at around 9% here even at 2-300bb deep which looks like a large error

HJ is not EP this is just trolling at this point. How the **** does running a sim of a different hand in a different game help anyone think better about the hand they actually played? You trust the solver blindly but don't actually understand how it gets to the results it does or how changing material things about the hand influences those results. You're impossible to have a conversation with because all you ever contribute is just "the solver said this" but you don't actually understand why, so when you're confronted with a situation the solver can't accommodate you just input garbage into the solver and then post garbage.

That's a slippery slope though. If the aggro maniac is correct in his read that in this game people overfold to aggression, he's not really wrong in playing 88 aggresively. He is not playing agaisnt a table full of pluribi.

So when you say it's a losing play with 88, that's only true if hero is playing perfectly both pre and post

Interesting about the turn lead. I wouldn't think either player has a nut advantage on that turn. Interesting.

I would think SB should continue with all suited aces (cooler potential), any pair that he 3 bet, ak, and a bunch of pure suited connectors from 56s to kqs (may be fold 89s pre). Fold off suit hands worse than ak.

Honestly, I don't know how valuable it is to analyze this spot in a solver.

We know he is not playing GTO. We know the deviation he is making. So for all future spots with him, the response is kind of intuitive.

We’re all shooting in the dark here obviously, but I think facing a small 6b I do something like shove 2 combos of aces (call rest), shove 0.5 combos AKs (call 1 combo, fold 1.5 combos, and I already flatter 1 combo), I’m flatting any other pairs that I happen to show up with (assuming a standard ~2.5x raise) and folding everything else.

Against a 6b jam, I don’t see any reason to call anything but aces. That’s almost 1/3rd of our range at that point and they’re shoveling 6x in the middle.

EDIT: Given the sizings in this hand, it’s really not viable for villain to 6b small. Which is funny that there aren’t even 7bs 1k bbs deep because of lol live sizings. (This makes this a more frequent 5b btw…😉

The above would be my strategy if it were like 60->240->720->2000->5000.

ok so what do you think the pre ranges look like?

have been looking at sb vs mp, wait until u see how tight sb vs ep ranges get!

the issue is sb's range is just so so tight pre. if he starts off 3betting 4-5% of hands, and then he folds 50-60% of that, there just aren't many combos in it. especially on a board like this, he's going to have STR+ often enough that he doesn't need to call anything else very often it at all if large bets go in. granted op fumbled here but it's a cold 4b pot, if you need someone to fold better than AA, you're probably making a poor bluff

i think people just get carried away when they 5b or w/e of "well my range is aces or like 3 combos of bluffs so i should go for it" while overlooking some boards you just cant put infinite in w aa

I liked what someone said earlier in this thread. When you see a very rare action, most players almost always have it and some players almost never have it.

So my heart says if I face a 6 bet shove here, I should lean on my knowledge of v. It is a rare human who would put in 20k without aa at $20 blind level.

I think a small 6 bet is possible. If we 5 bet to 3300. V can go 7k.

My intuition is small 6 bets are very bluffy if v is aggresive enough. Just intuition. Because the kind of person who wants to 6 bet small with aa only is the kind who would flat the 5 bet to begin with (in hopes of getting more money on the flop and not losing their customer preflop)

i think this is too loose for how the spot is *supposed* to get played. again i do get what you're saying re solvers but having 300bb sims pre is the best we or at least i really can do unless we want to let monker run for months so i think its worth looking at them. BB's big issue here is if sb is playing well (honestly, i promise u he isn't given comments in his thread), he is supposed to be super tight pre. like the 300bb solve has bb cold 4bing KK+, AK, a5ss, and like 5% of a few hands for board coverage. i get it's live people are spewing, it's boring, but these super deep spots you have to try to be precise bc you can end up making 400bb errors on later streets really easily (look at OP here).

it just isn't a spot where bb gets to be super loose as HJ is going to have a good hand at some freq (~5%) and sb is going to have a good hand very very often (20?%) of the time. good hand in this context means 5bettable.

i do get i've been antagonistic in this thread so am sorry about that.