Set mining deep stacked

Great question, and the answer is largely no, stack depth is severely overrated as a factor for set mining for exactly the reasons you're alluding to.

People will use a sort of "line goes up" logic because as you go from 10bbs to 25bbs to 70bbs deep, the value of set mining shoots up. Unfortunately, it tapers off after that and baby PPs eventually become reverse implied odds hands sufficiently deep.

Now, caveats abound. Some situations where it does matter:

1) As I just suggested, stack depth is enormously important for set mining up until the point you become sufficiently deep stacked to consider set mining to begin with.

2) I also specified baby PPs above, because having TT or even AA is a great way to turn a set over set situation into a "Call an ambulance...but not for me" scenario.

3) If you're against a total whale who will stack off 460bbs deep with the type of hands mere mortals would stack off 100bbs deep with, then by all means you can salivate sizing up their whole stack. (Though even this depends on there being some mechanism for getting that much money in, like a massively multiway pot and/or 3b/4b pot and/or they're especially aggressive and/or there's a history of huge overbets between you, etc.)

4) Being deeper stacked puts pressure on your opponent to raise sufficiently large to keep you from just flatting all your PPs. So you'll mix folds facing a 3b as small as R75 IP 100bbs deep, but they better go R133+ if you're 200bbs deep. However, if you math that out, you'll find this ain't exactly the 10:1 rule or whatever.

So the caveats might add up to a simple majority of scenarios where stack sizes will have a significant effect on your decision, which maybe would mean my first paragraph is overstating the case, but I'm guessing this post will serve as a counterpoint to what most other people will say.

Mostly what he said ... but with less words, and caveats:

0. In general don't stack off 30x flop SPR with top two or bottom set vs. competent people.

1. Some 2-5 players are just 1-2 players with more money, so lots of "bad" things can be fine. If you are in great games, some might mean lots.

2. The math works out differently based on open sizes, so a competent player might be opening 4x-6x at 1-2 but only 3x-4x at 2-5. You'll also (often) get less random callers ... this means you can't just look at stack BBs and/or can't easily take experience from 1-2 in general (apart from rule #1). Dito. tournaments where an open size might be 2x.

3. People will squeeze more at 2-5, so it doesn't matter if your call is fine "if you see a flop" when you don't see a flop.

4. A lot of mid or worse 2-5 players will auto call 22-77 for set value, so you can go a bit lower than TT (77 would be my floor) as overset value hands if it goes multiway (but again make sure it's not likely someone will squeeze behind).

5. Because of #4 there's still value in not auto stacking off postflop when you've put 25% of effective stacks in, but sometimes what you gonna do.

Unless they're huge fishes or super aggro
we are probably never getting their whole stack for 400+bb even when we hit our set.

Of course it all depends on spr. In single raised pot, I'd assume they aren't going to stack off without top two or better, combo draws, etc. Sometimes we are even faced with better sets when the money do go in.

You're a decent poster (better than I am) and I usually learn from your posts. Which is why I'm asking, what's your authority for the above? A bunch of sims, experience, Galfond said it, something else?

Absolutely agree with the idea (if that's what you meant) that big hands that aren't the nuts lose a lot of value the deeper we get. And small pockets which, even when they hit, don't make the nuts usually, certainly qualify.

I'd add to OP that IO needs to be 15x, not 10x for happy set mining. My query was about the idea that 40x might not be a good idea either for the 22-77 set.

Question on #5, (and completely agree on #0. Deep stacked is a lot like Omaha in that, if all the money's going in, at most tables, it's the nuts or air. 2nd, but usually third or so, nuts is often wondering why they lost.) But anyway, on #5: Is it 25% now where we shrug, sigh "Pot committed..." and go with it, versus 33-35%?

(obviously what RA said, but I'll add my 2 cents too)

First, 10x is way too thin to setmine. We're just barely getting over the 8:1 of flopping our set, but KK/etc. isn't going to stack off on Axx/etc. flops, a whiffed AK/etc. isn't going to stack off, and we'll sometimes lose our whole stack with our set.

In my opinion, setmining requires (a) a multiway pot so that our immediate odds get us ~halfway there plus we don't limit ourselves to a single target, (b) position to help prevent streets checking thru / turning the strength of our hand face up and (c) ~inbetweenish stacks which are large enough to make setmining profitable but not too large to tip to RIO if some non-insane person is willing to get in very large stacks.

Here's a little situation to think about regarding how useless stack sizes alone are when considering whether setmining is profitable:

Let's say it folds to the Button who raises to $20 with an $80 effective stack. We have 22 in the BB. Seems like an ez fold due to no IO.

Now let's say the dood has a $200 stack. Getting closer, but still not close enough.

Ok, now let's say dood has $600. Wow! We're getting 30:1 IO! EZ call, right?

Now let's say the Button is [insert our fave poker pro]. Is that dood, in position to boot, going to give us his stack when we flop a set? Highly unlikely. Course, we're not playing our fave poker pro. But while the worst player in our pool might lose his stack, most others won't (especially if we have anything remotely close to a tight / nitty image).

And on top of all that, now consider the Button just open raised when folded to them. How often are they actually going to have a hand willing to consider stacking off?

The point is that stack sizes are only one consideration for set mining. IMO, profitable setmining spots are actually fairly rare and it's an overrated concept.

GcluelesssetminingnoobG

Grunch: 10x is not enough, imo, even when not deep stacked, as they don't always pay you off, and sometime you flop a set but lose anyway. I want 15x.

As for deep, you don't necessarily need to get thier full-stack to have enough IOs, but in order to get the SPR ratio small enough to get a stack off, you'll usually have to open PPs and call a 3-bet (assuming you are OOP) and sometimes 3-bet yourself (which opens you up to a 4-bet, and is thus less preferred).

I did not do a lot of math/work to come up with a new lower bound ... I just picked a "random" number roughly between 10% and 35% where I don't try to shrug stack off, but also try not to hate myself when I do.

It's inferred from:

- Preflop charts make it clear that (in theory) the direct odds you're getting to win the pot unimproved with PPs are more important than the odds on someone's stack when you hit a set. For example, solvers call pocket pairs more facing a 3x 3b IP 100bbs deep than they do facing a 4x 3b 200bbs deep, even though you're getting better odds on players' stacks in the latter case. Another example: you basically shouldn't fold a PP facing a 25bb 4b over a 10bb 3b 100bbs deep, even though you're not getting odds to set mine.

Plenty of interesting things happen with pocket pairs at different stack depths. IMO they're lowkey the most interesting preflop hands in NLHE. But most of that isn't important to this sub in particular.

- Postflop math: You can calculate the point at which a hand gains implied hands from being able to hit certain hands (like bottom set) and then at what point it starts to suffer from reverse implied odds.

- All my lived experience (and other people's HHs) corroborates that the theory applies in practice. By the time both players are happily pouring money into the pot after a flop 3bet (or whatever), myself and every other thinking player I know starts to feel like they're on the worst end of it. (Caveats notwithstanding.)

Obviously theory isn't everything and cooky situations abound in live poker. Like, if there's a 4bb raise and everyone calls, and you're in the SB with ducks with a passive BB, then the fact that you're 250bbs deep is gonna make this a juicy spot to pure set mine even though it's obviously trash in theory. But *for the most part* the way that these games diverge from theory aren't usually that people get super big stacks in super light.