Hey guys,

A few people asked that I write an article on what I know about the 3bet/4bet game theory, so I thought I would just throw out an actual post. This is all, obviously, in my very humble opinion. If anyone would like to add to this or question something I say, feel perfectly free. My ego isn't exactly wrapped into this post, and the best thing I can hope for is a ton of feedback and an evolving guide to HU 3bet/4bet game.

I'd also like to stress that I'm unsure of whether or not this should be put into one place. If there is an overwhelming wish for me to stfu and force people to do this research on their own, I will willingly just edit my post to contain nothing but a pic of Alyssa Milano mostly naked. I do understand that many people out there probably don't want information this accessible, I just don't know what portion of the HU SnG 2+2 community this is. So, again, I'm just some guy, if you'd like to er...put me in my place, which I quite clearly already know. I just want it to be perfectly clear that I don't know everything about everything, don't think that I do, nor do I take personal interest in the existence of this post, and would be more than willing to relinquish the right to making this information extremely easy to access if that was asked of me by the larger community. As a coach, this information is my product, and the more easily accessible it is, the weaker my product becomes. As a player, my bias is again to keep as much to myself as I can.

But hey, it's the holiday season. And this is geared for HU SnGs, where, personally, I don't even engage in a 3bet/4bet dynamic in 95% of my games...so...here...

There are a lot of threads that read: "what should I do against a villain with a 25% 3bet%" or "should I 4bet or flat with ATs?" etc. There is generally a very large misunderstanding of the 3bet/4bet game among the poker community, as there is also a misunderstanding of why people have and do engage in a 3bet/4bet game.

The 3bet/4bet game as we know it is the result of what was a very profitable way of abusing people's inability to adjust to rampant 3betting. To my knowledge, it began in 6max games where people's PFR was rather high (in same cases, 25%). Some of the smarter regulars realized that a vast portion of players considered 3betting only when they had the best possible hands (AA, KK, and sometimes AK), and the 3bet was so strictly used for value that it was the most easily abused situation in which to bluff. A fantastic paraphrase from Bart Hansen, "You can't make a play for value that you do not make as a bluff". Create a polarized situation as an example: if I only 3bet with AA, you can play perfectly against me with any holding when I 3bet. If I 3bet with AA and 45s, your KK has 48% equity against my range, and you can't make much a move at all. Pretty simple, but rather fundamental when you think about poker in general, and what all this "balance" and "range analysis" is really about.

So, people began to widen their 3betting ranges. At first, it was a bit absurd. There were winning regulars with 3bet%s upwards of 30% in 6max games, which sounds ridiculous. However, when people are unable to adjust in any way, you can 3bet any two cards profitably. Eventually, people began to adjust, 4bet lighter, flat lighter, etc. This forced these regulars to rethink their 3bet game and actually balance. Obviously, the first level of this game is:
- If someone is either going to 4bet or fold, and going to fold more often than they should, it is profitable to 3bet them.
Or, more generally:
- If someone folds too often, you should bluff more.

Rather obvious, but fundamental.

Now, people started to 4bet lighter. They stopped folding TT and JJ 100bbs deep. No longer are people folding too often, so you have to do more than just bluff very often. Now, you have to create a balanced range. It was rather revolutionary when people began to 3bet with a polarized range (the nuts or a bluff), but they were doing it incorrectly. They had the theory correct, but applied it incorrectly. The theory is:
- If someone is either going to 4bet or fold, you are getting your equity from their folds and all-in situations. Thus, if you don't plan on calling a 4bet or 5betting, and your opponent is rarely ever flatting you, the hands you 3bet-fold are largely irrelevant.

People began 3betting SCs and SgCs. However, what they did was turn a hand that already makes them money into a hand that makes them slightly more money. What they should have done was take a hand that makes them no money into a hand that makes them money, and allow SCs to remain SCs. For example:
- You are 100bbs deep
- You have position on Joe Donk
- Joe Donk's PFR is too high, and thus his range is relatively weak.
- Joe Donk does not flat 3bets, and when he does, plays fit or fold (ie., a Cbet will often take it down).
- Joe Donk raises preflop, and you decide you want to start abusing him.

So, you'll create a polarized 3bet range. You decide that Joe Donk will 4bet shove a range of approximately:
JJ+
AQ+
Because, well, he's a nit. That's 4.2% of his possible hands.

So, you want to stack off with a pretty similar range of hands. Let's say the exact same range of hands. This means that your preflop ranges begin as:
- 3bet: top 4.2% of hands
- Flat: 22-TT, SCs, SGCs, generally +implied odds hands that want to see a flop but cannot stackoff against his range.

Now, this isn't optimal, but say we wanted to exactly mirror our top 4.2% range of stackoff hands with hands that we 3bet with as a bluff. This way, when he folds preflop, he is going to be making a mistake 1/2 the time. When he 4bets, he's going to make a mistake 1/2 the time. Our range matches itself in the sense that we are not more likely to show up in a 3bet scenario with a stackoff hand than we are the nuts, and vice-versa. Well, since we're either going to force a fold or fold ourselves, there's no reason to use any of those +implied odds hands. Take the hands that you can't flat with profitably and turn them into a bluff. What this does is weaken your 3betting range, but make hands that otherwise were NOT profitable into hands that are very profitable. So, we now 3bet with:
- JJ+
- AQ+
- K2s-K9s
- Q2s-Q7s

Now, logic is probably telling a lot of people "why would I want to 3bet Q2s instead of AJs?" The answer is rather simple:
- What happens when your opponent shoves?
- What happens when your opponent calls, then Check-Raises all-in on an A25r board?
- What happens when your opponent folds?

In all of these situations, you aren't particularly happy. If your opponent shoves, you have a hand that doesn't have equity against his range. You can call and deal with the variance involved in getting it in behind because you have committed so much money that folding would mean folding equity against a stronger range, or you can fold and lose 9 or 10bbs.
If your opponent folds, it doesn't matter what you had.
If your opponent calls, the situation really isn't that different than hand 1. You do not dominate your opponents stack-off range, so when he wants to stack-off, you don't have equity against him. He's probably going to fold to a Cbet, but again, your hand doesn't matter. Sure, you can flop 2pair or a flush, but you can do that with Q2s too.

So, flat with AJs. Now your opponent's range has dominated hands when you have AJs, so your equity shoots way up. You don't isolate his range to only include hands that beat you.

A quick note: why hands like K2s? What's the difference between K2s and 72o in this situation?
Answer: In theory, not much. But poker isn't black-and-white. Sometimes, your opponent WILL flat, despite all evidence to the contrary. In that situation, you'll have a high PSR and a hand with the ability to flop TPWK and flush draws. Though there isn't equity in your hand to flat (in this situation--there are situations in which flatting K2s is +EV), the range I chose here is meant to give you the most equity from a hand that did not otherwise have enough to flat, and also not good enough to stack-off with (so we do not commit ourselves into stacking off with what was meant to be our 3bet-fold range).

But, now we have a new evolution in the 3bet/4bet game. We all realize that people are 3betting with air and the nuts, and it is really hard to play very strong post-flop poker with air. So, what happens when villain's start flatting our 3bets too often?

Well, now you can create bloated pots in situations where your opponent's hand strength is weak comparative to the strength his range would normally be isolated to. While your opponent might only 3bet with AA, he'll call a 3bet and play an 18+bb pot with something like KJo. Maybe hands as weak as A2 or Q8. When your opponent starts to flat too often (or if they begin this way, as is often the case in the micro-mid stakes) you should depolarize your range. Now, you aren't dealing with the logic that it doesn't matter what your hand is. On the contrary, you are now working on the other end of the general fundamental poker theory spectrum:
- When your opponent doesn't fold enough, your range should shift to value.

If your opponent starts flatting 3bets with a whopping 20% of hands, you can now 3bet that AJs very confidently. When that A25 flop comes down, you aren't nearly as worried about getting it in. The PSR is very large so the value of top pair goes up, and you'll often have a mandatory all-in. The difference now is that your opponent's Ax range isn't AK, AQ or AJ, it's A2+. That makes an enormous difference in your equity share of a stack-off situation. The same will hold true for much weaker holdings (ie., pretty much any two broadway cards).

When you depolarize your range, you'll probably want to choose hands that play well postflop. While A2s is a marginally better holding than KJ, I would rather 3bet KJ in this situation as you will flop top pair more often. You want to simplify your decisions in these spots. Your opponents will often react in very different ways to bloated pots (some become very fit-or-fold, others will spazz relentlessly). With hands like KJ, you'll flop top pair, middle pair, flushes and straight draws very often. You'll be able to play your hand better, in general, than you would with 22, A2, etc.

This covers a general outline of how the 3bet/4bet game developed, but obviously isn't as deep as it goes. And further, as HU SnG players, we are very quickly limited in our abilities to 3bet with air. To create a polarized example for the simplicity of learning, even if your opponent is only 4bet shoving with only 20% of hands and opens 100%:
- Blinds, 50/100
- Effective stack size: t1000
Villain raises to 200
Hero re-raises to 500
Villain raises to 1000, and is All-In
You have 500 left with a pot of 1,500 chips, so you're getting 3-1 on your money. You only need to win the hand 25% of the time to break even. Say you have K2s and your opponent's range is 20%, you have 38.5% equity. If you fold, you're flushing money down the toilet.

Not that getting it in with K2s is a mistake, but you aren't 3betting with it with the intention of created a polarized range. You can't, because you can't fold anything in your preconceived 3bet range. Since you cannot 3bet-fold with almost any holding and make it a profitable play, you now lose this part of your arsenal. Every 3bet has to be a 3bet-shove or a 3bet-call. When the blinds get to the point where they are too high to 3bet-fold, you have the same question to think about:
- Is my opponent folding too much, or not enough?
If your opponent isn't getting it in light enough, it's pretty simple math to figure out how often you should be shoving. The question here is how to balance your flatting range versus your shoving range. This is more of an end-game strategy thought-process, and not the 3bet/4bet dynamic. It's just important to know that HU SnGs are very limited in how long you can maintain a 3bet/4bet dynamic, and how it affects the late-game. I personally don't engage in much 3betting, as I prefer to keep as small a PSR as possible in the early game to abuse my postflop edge. What people fail to realize is that people aren't folding too much anymore, and this entire dynamic is often NOT the most profitable way to make money. SnG players are not used to playing hands on the turn and river, and so taking players to those streets as often as possible will force many more mistakes than you can force preflop. Though engaging in these dynamics is profitable, it usually isn't where most villains have their biggest leaks.

Now that you know the limitations of the 3bet/4bet game, and understand the development of thought that brought poker to where it is today, you can take that information and apply it to your opponents. Say you sit across from me, and I'm feeling spunky that day. My 3bet% is about 20%. How do you figure out if I am polarized or 3betting with a wide value range?
- What do I do with A2-AT? Do you see me flatting with them oop?
- What do I do with my 2broadway hands? Do you see me flatting them oop?
If they aren't in my flatting range, they're likely in my 3betting range.

- What is my VPIP oop?
The beauty of a polarized 3bet% is that it makes MORE hands profitable. When I depolarize, my flatting range slims considerably. If I'm 3betting all the broadway hands that I would otherwise flat with, my flatting range drops considerably. For example, if I am polarized, my range might look like:
3bet: top 10%
Flat: second 30%
3bet: bottom 10%
VPIP: 50%, 3bet%: 20%

If I'm 3betting a wide value range, say the exact same amount, my range drops a full 1/5 of the hands I was playing before:
3bet: top 20%
Flat: second 20%
VPIP: 40%, 3bet%: 20%

Now, think about what I am trying to accomplish with what I am doing. If I have a polarized 3bet range, that means I'm counting on you either 4bet or folding. So, your adjustment is pretty obvious. Just flat me a lot. Maybe just never 4bet.

If my range includes a wide value range, you have other adjustments. You might not realize this, but it also weakens my OOP flatting range, and does crazy things to the way I have to play pots in which I flat.
- Put a LOT of pressure on my when I flat. If the flop is K23r, I can't have TPGK. If I have top pair, it's weak. As a matter of fact, there probably isn't even a K in my range for flatting, unless it's K9 and the occasional K8. Three barrel the crap out of me.
- Don't flat. Just stop doing it. I'm wasting all my value hands to earn 2-3bbs, and only playing pots oop with mediocre holdings that you can now abuse. If we were deeper, you could create a polarized 4bet range, with a similar range of hands to your 3bet range. However, this is a HU Cash concept, not a HU SnG concept. For example, when I play a cash game reg, it might go:
Hero: Raise to 3bb
Villain: Raise to 10bb
Hero: Raise to 25bb

And I'll raise to 25bb with that range of JJ+, AQ+, Q2s-Q7s, K2s-K9s. You can't really flat that 4bet, or you're in for a world of hurt. Your KTs is the best hand 1/2 the time, and crushed the other 1/2, so shipping it for another 75bbs is wild. You have no profitable option.

A general adjustment you may have to make to someone who 3bets a high percentage, regardless of whether it's polarized or depolarized, is to lower your PFR. While you might have started the match with a PFR of 90%, drop it to 50%. My 3bet game also hinges on the theory that you raise an exploitable percentage of the time. To adjust, you can strengthen your BTN raising range and thus fold less often. Again, we return to:
- When your opponent folds too often, you should bluff.
- When your opponent doesn't fold enough, you should play for value.

So simple, but so incredibly important. I was 3betting you because you were folding too often to my 3bets. By strengthening your range, you are no longer folding too often. If your PFR is only 50% and you can flat 20% of hands, you now fold less than 30% of the time I 3bet.

Many people will also create a limping range...but I don't really want to delve into this too much. It's avoiding the 3bet/4bet game, which can be good. However, most people's 3bet/4bet game isn't very strong, and thus can be abused. You can small ball rather effectively this way, but it's the exact opposite of this post. Benefits and drawbacks:
- You get to see flops in position.
- If your opponent is a super-donkey/station, you'll get to play postflop poker and make 10bbs whenever you hit.

- You lose initiative.
- Your range for both raising and limping is unbalanced and exploitable.

I would rather just fold than pay a SB to see a flop with a hand that I can't raise, with rare exception. There are some villains that will still play rather large pots with air whenever you flop top pair, even if it's a limped pot. However, these villains are rather rare, and most people err on the side of folding too much in limped pots. This is also exploitable, but for 1bb at a time. Further, if they are raising their BTN, the pot will, on average, be larger when you are oop than IP. You have to have a monstrous postflop edge to counteract that. This is actually a large reason I don't 3bet much in HU SnG land, I minraise preflop. I don't want to play 4bb pots ip and 20bb pots oop. It's a recipe for disaster. In most situations, I prefer to keep the pots small and 3bet a wide value range in the end-game, abusing the image I've created by 3betting a small % of the time. If you 3bet 20% of the time at the 10/20 level and that shows up on your villain's HUD at the 50/100 level, he's getting it in with KJ when you 3bet with what is still a 20% range. If your opponent sees that your average 3bet% is 0-5%, and your actual range in the late game (when the fold equity matters most) is 25-30%, your opponent is going to make an extremely exploitable fold. While the equity is there in the early game, it's much larger in the late game. However, understanding why and how you should be 3betting will help you exploit people doing it incorrectly, and adjust to those doing it correctly.

If this is all a bit overwhelming, think about it this way. No matter how deep you go, poker boils down to two theories. These two theories are applied in every individual situation against every individual opponent (ie., dry boards, crazy villains, wet boards, nitty villains, river NASH spots, etc.):
- When someone folds too often, you should bluff more often.
- When someone doesn't fold often enough, you should play for value.

Playing for value means both bluffing less (or not at all) and widening your value range. In all spots. If you see Isildur shove against Ivey on an 789T board with 6x, it's because he probably shoved with air the few times. At first, Ivey would only call with Jx. Then, Ivey adjusted to Isildur's polarized range (the nuts or nothing) but calling with two pair and straights. Now, when Isildur shoves, he shoves with a "merged" (wide) value range that includes no air, but instead a lot of hands with equity against Ivey's calling range.

When I raise the turn with nothing but a straight draw, it's because I think you're folding too often because you believe my range to be weighted heavily toward monsters.

When I raise the turn with TPTK, it's because I think you're calling with worse hands, and I should just call with my draws.

This continues into every facet of poker, as it defines your 3bet/4bet adjustments. The reason your structure your ranges the way you do is to find the most value across all your ranges.

I'd like to end this by saying this goes deeper, as it always does, and your exact ranges for doing the things you do can be more precise. This is just a general guide for people with little knowledge of what was posted above.

Happy New Year everyone.

Wow, thats a lot of text looks interesting thought, will read.

Can't edit...but someone asked me question which basically went:
"I heard when someone 3bets like 30%, it's okay to 4bet a wide range"

Obviously, true. But this depends on relative stack sizes, his range and sizing.

Again, I stress that in HU SnGs, you are generally going to respond to rampant 3bettors in the late game by tightening your opening range, minraising, and jamming a wide value range. That part is really, really simple.

Best ****ing post of 2011 man. Thanks alot, it's really helpful and I really appreciate it. Should be stickied in my opinion.

So...best quote of...the past 40 hours? lol.

Just kidding. Thanks.

very very nice post. nothing groundbreaking of course, but never before seen it presented so perfectly und crystal clear. thank you very much

Nice post ! Going to read it soon.

woah. id delete this post if i were u. its that good.

very well written and answered a lot of my 3b/4b questions

simply awesome.

+1

Thanks for putting this together.

This is a really awesome post. I took the time to read it all and I like everything you have said. It's a very very useful and informative article. One thing I would have liked to hear your thoughts on for some people would be when they are 3-betting their K5s and they do get flatted. Knowing how they should be responding on say 582 or smth, or AK8 etc.

Nonetheless you did a great job with this. ty!

very informative post, ty.

Will read again (and prob. again after that) when I'm not so tired, but I read most of it and I really like it.

To be honest though, I doubt I will use this a lot in my game, at least right now, since I'm only playing the 10's and 20's, but you do occasionally run into pretty decent (given the stakes) players, but who've been playing the same way at the same stakes for so long that they have no way to adapt. I think this post will be very handy against a lot of those players.

thanks again.

Great post but I dont like info accessible here TBH. I mean most regs at my level get a majority of this but it still is something I prefer people find out for themselves.

Just to clarify im not trying to be a jerk about this or anything. I understand 90% of people who can really put this to good use in the future are micro/low limit stakes players when they move up the ranks.

Then again on another note people can interpret this wrong as it is an advance concept and apply it in the wrong situations.

I guess I'm neutral on keeping it here after thinking about it.

thanks for taking the time to do this.

Although there are a lot of misconceptions in your post and I disagree with a lot of the info provided, I thought there was enough solid info jammed into the OP to make up for it. Also, very coherent and thought provoking post.

Awesome post, thank you.

yay, awesome article on 3b/4b. good work, ronin.

Thanks for all the replies. Appreciated.

Very interested to know what misconceptions Krumb Snatcha is talking about.

And as for people using this, tbh I don't in HU SnGs, as mentioned. I think it's usually a mistake nowadays to engage in the dynamic...I think you'll usually just end up leaving money on the table doing this against poor players.

Well, again, all the replies are appreciated. Happy New Year

Nice post about some 3B/4B concepts in general I liked it.

Detail:
One thing you should def. change is the way you use "PSR"... you use it just the opposite way it is normaly used - this is a bit confusing even though you are coherent throughout your article...

High PSR = you are deep.
Low PSR = you are shallow.

"In that situation, you'll have a high PSR and a hand with the ability to flop TPWK and flush draws", "The PSR is very large so the value of top pair goes up...", "I prefer to keep as small PSR as possible"

First two instances it should be "low PSR" because villain flats your 3bet and you are shallow now. 3rd insance should be "as high as possible" because you talk about being deep is better because you can abuse your postflop edge.

Whoops, I believe you are absolutely right. Will edit as soon as I get home. Thanks for the editor's eye.
Oh, shoot, probably can't edit it anymore.

Very nice read! TY

Btw am i missing something..



PSR=pot-to-stack-ratio= pot-size/eff. stack size

if pot is 10 and eff. stack size is 20 we have PSR of 0,5

If pit is 20 and eff. stack size is 20 we have PSR of 1

so high PSR=shallow right? If so Ronin Talken got it right

I think its because we can also talk about a SPR which is obviously the other way round

Either way, everyone should get it. Fwiw, PSR has a better ring to it IMHO. Lol.

Great post! More like this!!! I'd love to see a follow on about playing 3bet pots post flop.