8 posts / 0 new
Last post
mersenneary's picture
Sizing your 3-bets based on hand, opponent, and stack depth

 

Sizing your 3-bets based on hand, opponent, and stack depth

by mersenneary

 

Throughout this series of articles, a common theme has been to exploit your opponent's tendencies and the characteristics of your hand and the situation, and not excessively worry about balance or playing an unexploitable strategy. If you're playing unexploitably, you're not exploiting your opponent. That axiom holds true when sizing 3-bets, as well. To maximize expectation, it is important to vary your 3-bet sizing based on the situation.

 

1. The specific hand you have matters.

 

How well does your hand play postflop against marginal middling hands like J8o, Q9o, and 75s? Hands like AK and ATs play poorly against these type of holdings out of position – the ace especially is a very overt part of your 3betting range and not a big portion of your opponent's flat-a-3bet range, so middling hands manage to play pretty well against your holding. Hands like KQ, KJ, and premium pairs play much better with smaller 3-bets – you'll get a lot of credit on ace high flops, dominate a lot of your opponent's calling range, and just generally make it so that your opponent can less profitably call in position.

For fanatics of unexploitability, I still question the approach, but it's important to notice that just because we're taking advantage of the qualities of our hand, it doesn't mean that our ranges are unbalanced relative to preflop strength. We can make bigger 3-bets with big aces, 88-TT, and some bluff hands, which means that we have plenty of hands that are willing to get it in. Similarly, with our smaller 3bet sizing, we have some hands that aren't willing to get it in, but also premium pairs along with giving our opponent a worse risk/reward price on a 4-bet jam. You can argue that we become unbalanced on certain flops, but for the most part, there will be strong and weak holdings in both ranges, while still taking advantage of the properties of our hand.

 

2. Your opponent's tendencies matter.

 

If your opponent is calling a lot of 3-bets, experiment in making your sizings bigger, and see what happens. Even if you make it bigger than pot with AK, and your opponent is now calling in position with insufficient expectation with hands like J5s, it's often worth it to see if they'll call bigger sizings as well, and do even better than profitable (which again goes back to our big theme – don't just be profitable, be the most profitable). If your opponent is very tight against aggression, smaller 3-bet sizings often work better.

That is fairly basic, but when you start to get a little deeper into the specifics, it becomes clear that there's a lot more to it. If your opponent opens wide but rarely 4-bet bluffs (a common default tendency of many regular players), your opponent's range contains a ton of stuff that is folding against a 3-bet. Against a 100% minraising range to t60, and assuming a 3-bet to t150 will fold out 32o-A2o, 43o-A3o, 54o-K4o, 65o-K5o, 76o-Q6o, 32s-K2s, 43s-K3s, 84s-Q4s, and T5s-J5s, this smallish 3-bet with any two cards is already close to preferable to folding, even if we were to check/fold every single time postflop! Since clearly we do much better than this after the flop, it shows not just the value of 3-betting wide against a player with these tendencies, but also of making it on the smaller side, especially with bluff hands.

There are a lot of other tendencies to think about, so this is just to get you started. How often your opponent will 4-bet is often very important in determining 3-bet sizing, which leads well into the third part of this article.

 

3. Effective stack size matters.

 

The two important aspects of stack size are how often your opponent will 4-bet, and how well your hand will play postflop if you are flat called. At shorter stacks (less than 25bb deep), hands like big aces tend to just be correct to jam over a minraise, rather than make a non all-in 3-bet. This changes only when you learn your opponent is willing to 4bet wide, particularly with weak aces and suite connectors upon suspecting that your 3-betting range has enough bluffs in it. At this point, the value from making a smaller 3-bet with AK goes way up, as those A3o type hands may fold to a jam but ship it in over a 3bet. Before that, it's generally a mistake to make non all-in 3bets less than 25bb deep with AK, and particularly a mistake when the sizing is small, allowing tons of middling hands to correctly come into the pot in position against your holding.

If you need to have a balanced 3-betting range against an opponent with bluffs in it when less than 25bb deep, it is generally best either to leave yourself with a potsized bet to go all-in on the flop, or a ratio that allows you to c-bet/fold on the flop if necessary. At deeper stacks, your 3-bet size mostly determines your opponent's 4bet jamming options – whether they can 4bet/fold, and how comfortable they can 4bet jam all-in. Again, this is why a size of 150 can work really well over a t60 open against a wide button opener – it can be really awkward both for your opponent to jam for 1350 more, or to make a non-allin 4bet with bluff hands.

jackoneill's picture
Great article :-) I must

Great article :-)I must admit, I have been very, very lazy in this regard recently and mostly used a "standard" 3-betting size.


ServerBTest002's picture
Similarly, with our smaller

Similarly, with our smaller 3bet sizing, we have some hands that aren't willing to get it in, but also premium pairs along with giving our opponent a worse risk/reward price on a 4-bet jam1. Villain has worse risk/reward price because the pot is smaller and the stacks behind are deeper so... he's going to risk much more than he would win right? (sorry for the stupid questions but as I stated in the July FastTrack program, i want to be sure to understand every word).2. If villain recognize the worse risk/reward price on his 4bet jam, he should shoves a tight range, very tight... so what about 3bet more than the pot with big pairs? Risk/reward ratio is much better and a thinking opponent could shove wider ... what's your thoughts about this? If your opponent is calling a lot of 3-bets, experiment in making your sizings bigger, and see what happens. Even if you make it bigger than pot with AK, and your opponent is now calling in position with insufficient expectation with hands like J5s3. insufficient expectation because J5 is 2:1 dog? So we raise more than pot because we don't wanna get him better the 2:1 to call? Against a 100% minraising range to t60, and assuming a 3-bet to t150 will fold out 32o-A2o, 43o-A3o, 54o-K4o, 65o-K5o, 76o-Q6o, 32s-K2s, 43s-K3s, 84s-Q4s, and T5s-J5s, this smallish 3-bet with any two cards is already preferable to folding, even if we were to check/fold every single time postflop!4. Maybe I make some mistakes, but this range is actually his folding 50% of his 100% raising range... if we check/fold every single times it implies it is +ev preflop... probably I made some mistakes but, he should fold around 62% in order to BE? It feel strange... so 100% I'm wrong, can you post the calcs plz There are a lot of other tendencies to think about, so this is just to get you started5. What about postflop tendencies... if someone is willing to flat a lot 3bets and floats really wide your cbets what are your adjustments? I think our 3bet range should be more toward value range, a range that plays good postflop. But we'll not have these holding very often, so our range is tighter... so a thinking opponent should exploit this by floating flops that doesn't hits our range too often... some thoughts? 

mersenneary's picture
"1. Villain has worse

"1. Villain has worse risk/reward price because the pot is smaller and the stacks behind are deeper so... he's going to risk much more than he would win right? (sorry for the stupid questions but as I stated in the July FastTrack program, i want to be sure to understand every word)."For a 4bet jam, yes, they're risking an extra 1350 to win 300 in the pot, rather than 1320 to win 360 or 1290 to win 420 or whatever it may be."2. If villain recognize the worse risk/reward price on his 4bet jam, he should shoves a tight range, very tight... so what about 3bet more than the pot with big pairs? Risk/reward ratio is much better and a thinking opponent could shove wider ... what's your thoughts about this?"We're getting flatted much less often when we make it a bigger sizing, and that's the big problem with premium pairs, missing out on a ton of value."3. insufficient expectation because J5 is 2:1 dog? So we raise more than pot because we don't wanna get him better the 2:1 to call?"It's not about equity at that point, but expectation. Equity is what percent J5 is against AK if the hands get it all-in. Expectation is what will happen over time when AK plays with the initiative against J5 in a 3-bet pot. Those things are very different."4. Maybe I make some mistakes, but this range is actually his folding 50% of his 100% raising range..."After our opponent minraises, there's t90 out there. We could fold, or we could risk another t90 to win the t90 out there. It needs to work 50% to show a profit immediately."5. What about postflop tendencies... if someone is willing to flat a lot 3bets and floats really wide your cbets what are your adjustments? I think our 3bet range should be more toward value range, a range that plays good postflop. But we'll not have these holding very often, so our range is tighter... so a thinking opponent should exploit this by floating flops that doesn't hits our range too often... some thoughts?"Yep, definitely.

ServerBTest002's picture
"3. insufficient expectation

"3. insufficient expectation because J5 is 2:1 dog? So we raise more than pot because we don't wanna get him better the 2:1 to call?" It's not about equity at that point, but expectation. Equity is what percent J5 is against AK if the hands get it all-in. Expectation is what will happen over time when AK plays with the initiative against J5 in a 3-bet pot. Those things are very different.mmmm does his expectation depends on how we play the hand? First hand, Villain raises to 60, we 3bets to 200, Villain calls... Jxx . If villain know s we'll overplay AK here, he got positive expectation here. He's investing 140 preflop to win part, if not our entire stack... Obv this is an example... but if you can elaborate a little bit should be much clearer  "5. What about postflop tendencies... if someone is willing to flat a lot 3bets and floats really wide your cbets what are your adjustments? I think our 3bet range should be more toward value range, a range that plays good postflop. But we'll not have these holding very often, so our range is tighter... so a thinking opponent should exploit this by floating flops that doesn't hits our range too often... some thoughts?" Yep, definitely.Some ideas to counter it?

mersenneary's picture
Yes, there are favorable

Yes, there are favorable flops for J5, there are also unfavorable flops. A lot of the time on the flop, we'll c-bet and fold out 25% equity. That's why we can't think of it in terms of equity, but rather of expectation over time given that we usually don't even see a showdown. When we raise preflop with 87o, we don't calculate our equity against a calling range, we know that there's going to be a lot of good spots to c-bet and play the hand well in position, etc.As for floating, you really just said it. If he's too active on the flop, we tighten up our preflop 3betting range and get huge value with our big hands. We can also double barrel more often and bet/call lighter on turns, etc.

adastfe's picture
"After our opponent

"After our opponent minraises, there's t90 out there. We could fold, or we could risk another t90 to win the t90 out there. It needs to work 50% to show a profit immediately."I think if we 3bet 60 to 150 we would be risking 120 chips to win 90 chips in the pot, instead of 90 to win 90. Hero BB with 30 chips in the pot, villain SB with 15 chips in the pot. Villain min raises to 60. There's 90 chips in the pot. Hero can flat for 30 chips or 3bet to 150 for 120 chips. So villain needs to fold 120/(90+120)=0.57% of the time. Or did i make a mistake somewhere? 

mersenneary's picture
Nope, you're right. I'll

Nope, you're right. I'll edit.