great read but are you saying go ahead and 3 bet with j2-j6s q2-q6s and k2-k6s in the right spots?

I must admit, I was one of those NVG and BBV tards, mainly lurking, until I read this thread. All I've been reading now is Poker Theory and there is seriously some great stuff in here.

This has got to be the single best thread I have read on any poker forum. Great work and thanks. Now I just need to apply it to my HU game.

In the case were villain never calls but either 4bets or fold than, yes!

This is the part of the OP that I'm definitely not on board with. The 3 required features identified for the light part of the 3-bet range are:
1. Hands that are not +EV to call with.
2. Hands that are easy to count.
3. Hands that have such low equity that folding to a 4-bet is very clearly correct.

#1 is important, but I don't think it is justified to use such crappy hands just to be 100% certain that we are nowhere near anything that is a +EV call. If you want all low connectors in your calling range, use low 1-gappers for light 3-bets; if you want to call with even 1-gappers (and offsuit connectors, I suppose), move down to 2-gappers for light 3-bets. I want to say "and so on" but that's already plenty far enough for 99% of situations, and yet we are far from resorting to such dredges as J2s. Even if villain is known to never call, someone in the blinds will on occasion call with JJ and you'll be forced to play a flop; when that happens, you're clearly better off holding T8s or something like that.

#2:
Further discussion in this thread appears to conclude that if a 3-bet is profitable by the folds it gets alone, then everything that is a -EV call should be 3-bet. But hold on, we don't want to have 80% 3-bet because villain will adjust too fast and other people will start exploiting us! So the concensus is to abuse people a bit without triggering adjustments that would kill the golden goose. That means there is no mathematically derived perfect balance point that we aim to reach, it's more a question of what you subjectively think you can get away with, and that is going to be very fluid and doesn't require exact counting of hands to achieve a precise nuts:air ratio.

#3:
This is more debatable. Some villains will 4-bet so small that reraise-folding 87s in position starts to be somewhat painful. I don't know the best response to that situation, but it's probably not to call too many 4-bets (that reminds me of a hand where I 3-bet 86s and called a tiny 4-bet in position with deep stacks; we somehow check it all the way down and I win with a flopped 8 against 33... a win is a win but damn was that uncomfortable ).

Well, lets say you 3bet someting like J2s and SB coldcalls. Depending on playertype, game flow, metagame and so on you can make pretty good decisions postflop. Not to mention that you got position... Having said that, if someone coldcalls from the blind and they got a tight range, there is no problem just giving up unless you hit. You got a read then go with it. You are only in with 10bb or so...

Now, Lets say you have been an active 3bettor and you notice that people are starting to adjusting, well, then you have adjust accordingly. If people starts to 4bet you alot then you should continue with beeing polorized but perhaps tighten up a bit and flat there 4bet with good hands. If they starts to flat your 3bets then you should start 3bet a depolorized range... As long as you know when and how to adjust its ok to play liike a (percived)maniac..
Aslong as someone is never calling your 3bets you can be totally unbalanced with you 3betting.

Hopefully thats bring som more light over you problems. Hope its ok with OP that i answerd this and feel free to correct my post if somthing doesent seems right. I guess you got better knowledge than me.. Im just trying to help..

I would really like to here comment or response from op about Petey's live post, simple agree/disagree mostly....I really enjoyed/learned from the post

Now I pose a scenario that comes up a lot live in my game, villian open raises a ton, almost never limps from UTG to BTN, if I 3bet him with 24ss IP and I know he is never 4betting without the stones and we take a flop of AXX no help to my hand and we cbet and he flats to a broadway turn card and it goes check check and we are almost always folding to a river donk bet....

Have we talked about the then PERCIEVED range that the villian now has on us....and in live poker he will mostly just think "hero was trying to make a move on me preflop"

Is this more reason for us to then pound his open range with the top of our 3bet range in hopes that he 4bets or donk bets in later hands?

At what point are we throwing away money, how many 56ss 3bets before we should stop?

Some of that is totally not cohesive but maybe you get my drift, live player here so not really in tune with HUD aspects

It's probably so we can still get value when villain folds and still be able to fold when he 4bet. If he 4bet we can easily fold when we have thrash and ship when we got the nuts. Therefor he points out it's only in spots where the villain is either 4betting or folding.

Probably been said a million times already but this is a absolutely brilliantly written article.

10/10

Awesome thread, thank you very much OP and everyone else who contributed.

I'm still puzzled by something though and I hope someone will be able to help me.

If I understood the OP correctly, when the villain is 3betting with a merged range we should adjust by 4betting more. I didn't really understand with what kind of a range we're suppose to be doing that to combat his strategy?

An example:
hero holds JJ or TT in CO
villain 3bets a merged range OTB
hero?

Does hero 4bet/call here? Does he 4bet/fold? 4bet shove? Does hero fold to the 3bet?
What if hero holds As4s? Does he 4bet/call bluff? 4bet/fold bluff? 4bet shove bluff?

I mean I know JJ is ahead of the villains range but I have no clue what to do if his 3betting range is merged. And neither do I understand how to bluff without risking of either bluffing to often or risking getting shoved by something that wouldn't call a 4bet shove? Is 4bet folding ever ok 100bb deep?

I would really appreciate these questions answered!

I believe if villain 3-bets a merged range we should be 4-bet or folding more often and, further, we should be 4-betting a strong but polarized range. With the hands you're discussing what we do is dependent on stack sizes. Deep stacked, against an opponent with a looser range, we should call. Small stacked, against an opponent with a tighter range, we should fold.

If villain is 3-betting a merged range you also really can't be so certain we're ahead of his range either. A merged 3-betting range is somewhat villain dependent... but I would say a merged 3-bet range would most often contain [TT+, AQ+]. We are actually pretty far behind this range. Even if we add 9s (which is pretty wide, I think.) we still are not showing great equity against this range

I'm sorry. Apparently I failed miserably in specifying exactly the parameters.

All my questions are strictly meant with 100bb effective stacks vs a villain's merged range 3bet.

I feel my reasoning still holds, most merged 3-bet ranges have JJ in a good deal of trouble. Depending on 3-bet size you may be able to set mine profitably given villain's post flop tendencies.

Soo if our opponet is 3-betting OTB with amerged range ( as said above TT+ AQ+) we should be more inclined to 4 bet small/medium suited connectors right? since they have the best equity against that specific range.

In the case where we have jacks, they're not actually in THAT much trouble. Checked with Pokerstove, they have about 46% equity. In my opinion, if youre gonna play the jacks you have to 4-bet and take back the initiative or fold, otherwise you will be getting outplayed by your opponent far too often OOP. Lets say we flat the 3 bet, the SPR will probably be around 4-6, still far to high to commit with jacks in the rare case we flop an overpair to the board (27%)

I think you talked about some narrow 4bet ranges. Could they be optimal? Let's say for 100 BB stacks an optimal button open is 32% and the optimal big blind 3bet is 8%. I would guess the optimal 4bet range is 2% because 32/8 = 8/2.

If 2% 4bet on the button really is 'right', then it's even smaller in earlier positions. Our strategy would look something like this (on average, it will vary depending on which position is 3betting us):

button: 32% open, 9% call 3bet, 2% 4bet
cutoff: 20% open, 5.6% call 3bet, 1.3% 4bet
utg+1: 14% open, 4.0% call 3bet, .9% 4bet
utg: 12% open, 3.0% call 3bet, .7% 4bet

It gets to the point quickly where we only 4bet AA for value. But it could serve better as a flat to protect our increasingly face up calling ranges, which shrink down to just 3-5%. Maybe this is behind why you've found it works better to flat-only 3bets when OOP.

Both your estimates for optimal button open and bb 3b vs btn open are too tight.
And this just doesn't make any sense at all.

Vs 8% 3b you 4b JJ+,AK which is 3%, but say you slowplay those 25% of the time so you have a 2.25% value 4b range, but about 33% of your 4bets can be bluffs so you end up with a 3.4% 4b range.

Ok. 3.4%, or 4% without slowplays, is still pretty small. The cutoff's 4bet size would be 2-3%, like KK-AA, AK (or QQ-AA) and a offsuit's worth of bluffs. Then middle position KK would be the bottom of value.

Starting with 36% open and 11% would have got it closer to 3-4%.

But I would think JJ+ (or TT+), AK is the value range of 8% 3bet, it would be weird to 4bet them with the same value range?

What when our opponent has a polarized 3bet range but also is capable to 5bet bluff over our 4bet. Do we add some medium and low PP to our 4-bet bluff range with the intention to use them as bluff catchers against 5 bets?

It is AWESOME!

Thanks a lot for sharing

nice writeup

- Can't find the post you're referencing.

And live poker follows the same set of logic; it's just that the situations you are presented are generally very different than online. People too often overlook their 3bet game when playing live poker professionally, which is kind of silly.

I don't have enough information in the situation presented, but what we do know is that villain is opening a very wide range and not 4betting much at all (maybe 1.5% out of the 20+% of his opens, or 3-8% of his opens?)
There's still one very important piece of information we need, and that's how often he folds VS how often he flats the 3bet.

First, think about direct fold equity. If he folds more than 66% of the time, you're making direct profit regardless of any other information. If he folds less, we don't. So...

- We already know that our stackoff range is pretty much KK+.
- Everything that can flat with value against his opening range, but which isn't KK+, is now a part of our "middling equity range".
- Everything that can't flat profitably is air.

Now, just follow-through with the logic...
- If he's 4betting about 5% of the time, is he flatting about 60% of the time when facing a 3bet? It's doubtful, but possible. If he doesn't flat that often, then 3betting ATC is showing a direct profit. If he does flat that often, 3betting ATC does not show a direct profit and you should not 3bet anything that does not have equity against his flatting range.
- If he IS flatting 60% of the time he's faced with a 3bet, that means his flatting range is very weak, and you can net a huge profit by 3betting all the hands in your middling equity range that have equity against his range. If he doesn't flat 60% of the time, you're going to be bluffing with your air, but you're not going to be 3betting a very large portion of your middling equity hands.

So you're talking about two different players right now:
Let's say a 20% open-raise, 2% 4bet, who DOESN'T fold an exploitable amount of the time.
- Answer: Merge.
He's flatting with around 13% of his hands. That's a lot. My pokerstove is freezing, so I can't tell you the exact range...but I can tell you that you can almost certainly 3bet AT+, KJ+ for value.

Now lets say a 20% open-raise, 2% 4bet, and he folds the other 18% of his opening range (that's 90% fold to 3bet%).
- Answer: Polarize, weighted to air
Since he's opening 20% of his hands, you can flat a very wide range of hands profitably. So I'm not going to 3bet stuff like A2s, because I can probably flat it for a profit when I'm in position. I'm 3betting KK+ for value (about 1%, unless I'm mistaken), but since he's folding a whopping 90% of the time, I'm still going to 3bet a total of like 9% of air, raising my 3bet% to 10%.
What's great about that is...I'm probably flatting a range of like 25%, because I have position...AND I'm 3betting another 10%, so I'm playing about 1/3 of my total hands in that spot, and I can easily point to simple math and say "look, this is how I'm making a profit here".

Hope that helps.
It's possible that the first time you 3bet 45s is complete spew. It's also possible that 45s makes an immediate profit in fold equity. The C-Bet is a completely different equation...but I'd say it's almost certainly +EV. That's just EV stacked on top of EV...don't see how anyone can really tell you that's spew. :P However, if the situation keeps happening, remember that none of the logic is static. Maybe:
- He isn't folding as often as you think:
Answer: Merge. You misjudged your fold equity preflop.
- He does fold exploitably, but has way more Ax in his range than you think:
Answer: 3bet, but don't c-Bet.
- He is peeling the flop wider than you thought.
Answer: 3bet, C-Bet, shove the turn.

Hope that helps!

i'm not sure i understand the word merged. Merging our range means reducing it to only the top of our range or instead of 3beting with thrash or we should now 3bet with more solid hands like KJ (but keeping the 3bet % in the same numbers)?

Sounds like you've got it.
Basically, break your range down into three parts:
1- da nutz
2- stuff you can flat with for a profit, but not stack off with.
3- stuff you can't flat with profitably.

A polarized range is 1 & 3
A merged range is 1 & 2

Ok this explanation of merging ur 3-betting range mostly makes sense to me, but I was under the impression that range merging, involved 3-betting with 1, 2 & 3. Please explain why we don't wanna 3-bet ATC instead?

I have a feeling that, just like the point of range merging while value betting is to go for thin value by value betting with the utter bottom of our value range in the anticipation that someone will make thin calls while trying to bluff-catch, and when applied to 3-betting pre we're essentially going for thin preflop value (in terms of preflop equity of our hand against villain's hand) in anticipation of the medium-strength hands that villain will flat our 3-bet with. Maybe I just answered my own question, but I'd just like some clarification.

Yep, you absolutely did.

A merged range CAN certainly include ATC, and that occurs when our range is so much stronger than our opponent's that we can literally take an aggressive action with ATC and show a profit.

However, for the sake of just making the concept clear--

Generally, you are merging because your opponent calls with too wide a range. When your opponent calls too often, bluffing usually isn't a very good idea. :P

The same is true for why we don't include our thin-value hands when we are polarizing. We polarize when we don't expect our opponent to call much compared to how often he'll raise or fold. So we aren't getting value from our medium strength hands. So we raise our air if our opponent is going to fold an exploitable amount of the time.

Beore i read anymore what is ATC and GTO? But so far very good read, i can already see how this can improve my play, thanks