Min raises are the norm in my metagame. It's very unusual to see the button open for other than 2x on ignition. I'm very used to flatting a lot vs that size and 3 betting a linear range because I'm going to be called so often.
When I see a player opening for 3x it throws me off a bit.
I remember that 3 betting used to be gold and then people started opening smaller and calling more 3 bets. That makes me want to start 3 betting more vs a 3x open. But I don't know where to get these extra hands from. I used to 3 bet polar vs 3x opens back in the day but I'm pretty sure it's better to save that for playing in position or to use a polarized range once you know your opponent will be playing very 4 bet or fold.
If we don't have reads one way or the other, we only know villain likes to 3x open his button;
1) is it correct to 3 bet more
And
2) should I expand my linear range or should I bring in some bluffs from the hands I would have flatted vs a minraise but now I don't quite want to call given the open size? Maybe something like 63s, J2s, K5o, and 75o?

That's what I've been doing, sticking w my typical linear value heavy 3 betting range and bringing in a few hands that i might use for light 3 bets in a polarized range. So I'm betting a depolarized range w just a few light bluff hands w some blocker and/or post flop playability and board coverage.

Once I see how villain responds to 3 bets I can go full linear or totally polar.

But I'm not at all sure this is the right idea (especially the hands I'm bringing in and also the idea of adding some bluffs to an otherwise linear range?

This linear range you're calling is probably pretty bad if it includes stuff like JTo and KTo for no reason. A good default is a polarized 3-betting range because it gives you bluffing options and keeps your value hands clearly for value. Bluffs, especially high implied odds semi bluffs, will perform well (be higher or the same EV when played as 3 bets vs flatting) vs the population as a default. In NLHE, it's almost always better to start building pots with premium hands. Preflop, this effect will be very small, even for hands like AQo, but it will still be perceptible. It will be clearly better to 3 bet hands like TT-AA. For hands like KTo, it is probably better to flat as a default. I say probably only because I'm not 100% sure. I'm about 99% sure though. Beyond that, when to vary 3 betting ranges and how gets complicated.

Just remember, polarized 15-20% is a good default. Value should be clear value, no middling hands that will likely play better as flats, at least not without a clear reason.

Why would you want to 3 bet more frequently vs 3x if they're calling more*? Or you mean they were calling more when 2xing as a result of responding to the increased 3 betting when they were 3xing? If they're calling you significantly more when you 3x, you would want to 3 bet less frequently, much less frequently. Or did you mean more, as in a larger sizing?

Yes, part of the idea of polarized 3 betting is to have bluffs, but the population slightly overfolds their opens vs 3 bets, so its a good default. Also, keep in mind that vs population 4 betting ranges, which tend to be heavily weighted towards value, hands as strong as T9s probably are going to be folds, in terms of pure EV. This makes your more polarized hands very easy to play well vs. 4 bets, just fold, further adding to their value.

1)No.**

2)All of the hands you just mentioned, to a lesser extent 63s, are terrible 3 betting defaults. In fact I would never 3 bet k5o vs anyone and would only bring in a hand like 75o (at 50% weight) vs people who are severely overfolding and show no sign of adjusting to 30%+ 3 betting frequencies. If you're 3 betting J2s, K5o and 75o, or any of those hands individually, and you don't have a means of strictly controlling your 3 bet frequency, you'll be 3 betting 40%+, maybe 80%+ when including K5o. Some fish will react poorly to that. But at those frequencies, most regs and many fish will start calling more and at a minimum, start exasperation 4-betting you, quickly suturing one of the more lucrative leaks.

The question then becomes, how do I subtly exploit a subtle leak? That's trickier said than done in game and requires a fair amount of homework. What you don't do is start 3 betting some reg 90% because he's folding to your pf pops 55% of the time.

To give you some kind of framework, here's what a value bet is OOP PF vs someone who 4 bets 10% (the population): any hand with > ≈ 55% equity vs villain's not folding range. Does that mean that it's better to 3 bet every hand that meets the minimum requirements for value? No. Hands like KTo will play fine as 3 bets, but they make even more money as flats.

Note that if villain never 4 bets the definition of a value bet would change to >=50% equity vs his not folding range. Also, note that the equity of a hand like KTo pf can be significantly affected by villain's opening range as well as his calling range. But the biggest factor that can consistently push marginal value hands into becoming higher EV when played as 3 bets than flats is if villain never 4 bets. Keep in mind, a 4 bet stat of zero probably won't have even marginal significance until at least 1,000 hands into the match, that's if you 3 bet 20%.




*Proportional to your 3 betting size

**3xing as a binary 3x open/open fold IP opening strategy is bad for a number of reasons, mostly because it would only allow the opener to safely open around 40-50% of hands vs someone defending near optimally. Another reason is that it magnifies errors, the gravest of which is overfolding, where, at any given rate of overfolding, it is significantly worse to be 3xing than 2xing vs a perceptive opponent who is able to exploit overfolding.

TOTALLY AGREE! THIS IS THE CRUX OF WHAT I AM GETTING AT RIGHT HERE!!

Ok, so, we see a player who is just 3xing or open folding every button. We know we want to exploit that tendency but we don't know yet how he is going to respond to our 3 bets.

Should we; Polarize our 3 bet range right off the rip? Push those QJ, A8s, KT, hands right on back into our flat calling range and 3 bet stuff like TT+, AQ+, and KQ for value and balance with some hands from the bottom of our call/top of our folding range?

OR: Should we; continue using a merged 3 bet range and just go tighter and more value heavy that we would against a 2x opener? Suspecting that, just because he is opening 3xbb does not mean he will fold more than he ought to or even as much as he ought to, it could just mean he is a fish who is going to be a station preflop so we want to have strong hands when we play bloated pots out of position?

OR: Should we; Expand our 3 bet range to take advantage of the extra folds and the additional dead money in the middle when he does fold but keep the bulk of our 3 bet range strong in case we just end up going post flop a lot and then adjust as we go. And if we just want to expand the range should we do so by splashing in a number of bluff combos like the ones I mentioned or some other hands?

This is what I was saying, not to 3 bet a merged range AND all kinds of bluffs. But to take that same value heavy, thin value full, merged range

And so, if I use this type of range purely as a default until I know what villain is doing, then I will not be 3 betting a polarized range, I will not be 3 betting a linear range, I will be 3 betting a value heavy depolarized range. Mostly value, but with a few specific bluffs to give me some mix of post flop playability and blocker effects with my few bluffs and actually keep some board coverage as well by chosing a number of hands from across the spectrum BUT NOT ALL THE HANDS LIKE THEM OR ALL THE HANDS IN BETWEEN THEM


This is really just a thought I was playing around with.

I'd love to hear some insight into this dilemma.

Very basically:

How would you structure your 3 bet range in HU NLH vs an unknown player who sits down and starts 3xing his button opens?
And why?

Yes, this should be our default 3 betting range and we should keep using it until we have a reason not to. It's possible that people who 3x open fold proportionately less frequently (I have some moderate Bayesian priors suggesting they don't fold more, but that's all they are) to 3 bets, I have no data on that. But I wouldn't assume that's the case until I did.

Polarized 15-20% should be your default 3 bet strat. You should never change your default strat until given a significant reason to do so.

I would go tighter against someone who has proven that he is either calling too much or clearly is unwilling to fold too much. In game, this would roughly translate to someone with a statistically significant fold to my 3x 3bet stat of <45% (his sizing doesn't matter)*. If he fits into that category, I would also start betting bigger, but on a slow, graduated scale. Folding 45% to 4x is a notable mistake. Folding 45% to 5x vs a AQo+ range is a devastating mistake.

BUT, we don't want him to pick up on the fact that we're exploiting him. It's safe to assume that he is capable of making the right adjustments, when presented with overtly exploitative play, until he proves otherwise.


There's no reason to believe there will be extra pot-proportional folds, as stated above**. Related to this, there's no reason to assume that there will be extra dead money in the middle. However, when he DOES fold too frequently, for any given raw fold percentage, the mistake will be amplified with the increased open sizing, and dead money will increase.


How you expand the range is complicated. There are infinite answers. But the guiding principles are these.

1)When villain is clearly overfolding, you want to add more bluffs. An interesting question is, "above what villain folding percentage, for any given pf 3-bet sizing/open sizing pair, does 3 betting 72o become profitable, assuming it is not profitable to flat?" This question's answer tells us at what point we can safely 3 bet ATC that we cannot profitably flat. It also suggests which hands we might be tempted to first pad out our 3-betting range with, assuming it won't induce a spastic correction towards the optimum in villain's strategy.

2)By extension, if villain is clearly incapable of adjusting, we want to pad our 3-betting range with as much uncallable trash as we can get away with.

3)Freakish things brand the mind. Therefore, we might alternately want to first pad our 3-bet bluffing range with trashier but plausible hands like 45o, so as not to alert villain to our own villainy.

4)Versus someone who folds too infrequently, we should not expand our range. But if they are insufficiently 4 betting, we can get close to that 20% top-end-optimal figure, subject to the constraints in my first post.

5)As villain folds more frequently, all things being equal, more value hands will become preferable to play as flats. In the extreme, we may wish to 3-bet 72o while flatting AA.

Playability for semi-bluff hole cards can be defined by how often they make hands by the river that you want to get stacks in at 100bb deep. 76s, 65s, 54s are the most playable under that definition and most others that have been devised. Blocker effects, at risk of starting a row in 2014, should just be ignored.

As for choosing hands, I would just get stupid about it. Use a simple, linear range that you'll remember. Is that "value heavy"? Is it "depolarized"? I don't know. But if it's generating a 15-20% 3-bet frequency, it's pretty hard to exploit.




*I use a 3x 3-bet sizing as a default, so everything I say is based on that. Vs that sizing, villain should be folding pf in the 45-50% range at 2x open. That fold percentage slightly increases as his sizing increases to 3x, but in game these are just rounding errors. There are good arguments that something more like 3.5-4x is a better sizing. If that is the sizing you use, you have to translate the optimal villain fold percentages accordingly, keeping in mind that substantial equity is in play and his optimal fold frequency will be significantly lower than the naive freq.

**Other players may have better informed opinions on this, and I could certainly be wrong. But in the absence of convincing evidence or a respected player noting to the contrary, I would stand by my statement that there is no reason to believe 3xing leads to more folds. In fact, there are Bayesian priors that lead to the opposite conclusion.

Good posts, JudgeHoldem1848.

But one thing I don't get:
You say that TT+, AQ+, KQ (4.7% range) is your default value range, but you also say you 3bet 15-20%.
That means that your range will consist of 2/3 - 3/4 bluffs.
Why wouldn't I 4bet you every time?

Big aces play well, or ambiguously, as 3 bets, right down to ATo. 3 betting A9o is probably ever so slightly worse than flatting it vs population. QJo is pretty borderline and probably plays better as a flat vs pop. But if you're concerned about getting to a 20% 3bet - without too many bluffs - which doesn't seem like much but makes your strat notably more effective than 15%, you can throw some of those hands in as needed. A value to semi-bluff ratio of 1:1 pf is probably reasonable, especially considering hands like 98s have big implied odds, with high equity for semi-bluffs.

Hands that start to clearly become better as flats vs the pop are KTo, KJo, J9o, probably JTo etc. That's primarily due to the risk of being 4 bet and not being able to call. Although a hand like A9o clearly shouldn't be called vs the pop 4 betting range either. But it still has an equity and what people here call ex-showdown (that means realizable showdown equity) advantage over a hand like KTo.

Anyway, you don't need very many offsuit combos to fill out a 20% range at about 1:1 bluff:value.