Sizing a 3-bet

[angrystoc]

3 bets are kind of new to me. 90% of my experience comes from a house game plus about 2.5k games of micro over the last week.

From what I understand, I'm making a 3-bet with strong hands (AA KK AKs, possibly QQ / AKo depending on the range I give them) in order to:

1. Get a better read on someone raising into me.
2. Take away implied odds from small pair (?)
3. Get better value with them (?)

Is #2 correct? Should I mostly be betting up to a 9th or 10th of the effective stack?

I'm not sure I understand #3. Is it simply meaning that when someone does call the pot will be bigger and otherwise I'm taking down more raised pots that could flop badly?

[Erpel]

I don't think information should ever be the only reason for putting a lot of money in the middle.

The implied odds observation has merit and applies wider than just pocket pairs. Suited connectors are also implied odds hands. It's along the lines of making the opponent make a mistake.

3betting to a 9th or 10th of the effective stack sets up a favourable SPR for TPTK type hands - but if the 3bet size (headsup) is more than 3.5 times the previous raise a lot of the weaker hands in the opponents range will start folding - meaning the calling range you run into becomes very strong. A normal 3bet size is between 3 and 3.5 times the PFR. Some go up to 4 especially OOP. It is possible to manipulate it for SPR as you seem to want to, but you need to be aware what it does to the calling range - he'll fold the weaker part of his range.

Getting value is the #1 reason imo for 3-betting. One way of considering value is by comparing money going in with hand equity on that street.
Let's say you get 10% of your stack in pre-flop with AcAd vs KhKs - now I'd say that you have 8.2% of your stack in good and 1.8% of your stack in bad.
The flop comes Kc7d9d and you put another 10% of your stack in - now I'd say you put 1.2% of your stack in good and 8.8% of your stack in bad.
The turn comes 3d giving you a flush draw and you put in 20% of your stack - now I'd say you put in 4.5% of your stack in good and 15.5% of your stack in bad.
The river comes Ah and you put the remaining 60% of your stack in - now I'd say you get 60% of your stack in good.

In total you got 73.9% of your stack in good and 26.1% of your stack in bad. If the percentage of your stack that went in pre-flop or on the river goes up so does the percentage of the stack that went in good - if the percentage of your stack that went in on the flop or turn goes up the percentage of your stack that went in bad goes up. The quality of your decisions on every street will be evident in the long run.

As an example, if you'd managed to get all in on pre-flop you would have gotten 82% of your stack in good and 18% of your stack in bad. If limped PF and checked through to the river you would have gotten 99.8% of your stack in good and 0.2% of your stack in bad.

This method is critically flawed in some key areas which I don't need to go into, but it illustrates some underlying principles of getting your money in good on every street that I believe are worth knowing and which I think addresses your question. With a hand preflop that you think is stronger than the hand the opponent is holding you want as much money to go in as possible - so you put as much of your stack in good as possible. That's why you 3bet.

I'm only just scratching the surface of any of these topics, but I think the above will start you looking at it the right way and asking the right questions. It's also very important to think about 3betting as a bluff and how bet sizes manipulate ranges but it requires quite a lot of independent thought and analysis to sink in.

[a500lbgorilla]

You raise 88 in the CO to 2 american dollars, I reraise you to 7.5 american dollars from the BTN. What do you do with 88?

You raise 88 in the CO to 2 american dollars, I reraise you to 7.5 american dollars from the BTN. What do you do with 88?

shove because you're full of shit

[Illfavor]

You raise 88 in the CO to 2 american dollars, I reraise you to 7.5 american dollars from the BTN. What do you do with 88?

Set mine, assuming you have well over a full buy-in let behind, and fold otherwise?

[Sup3rM4N]

You raise 88 in the CO to 2 american dollars, I reraise you to 7.5 american dollars from the BTN. What do you do with 88?

Set mine, assuming you have well over a full buy-in let behind, and fold otherwise?

If its 50NL and u guys have 100BB stacks u arent gettin odds to set mine...also his range cud be polarized so even if u do hit ur set, doesnt mean u are going to get paid

[Illfavor]

If its 50NL and u guys have 100BB stacks u arent gettin odds to set mine...also his range cud be polarized so even if u do hit ur set, doesnt mean u are going to get paid

That's what I meant by "well over a buy-in left behind" being a requirement. "Well over" 100BBs. Yea his range could be polarized, but I generally set mine here when his stack size warrants it. *shrug*

[Robb]

From what I understand, I'm making a 3-bet with strong hands (AA KK AKs, possibly QQ / AKo depending on the range I give them) in order to:

1. Get a better read on someone raising into me.
2. Take away implied odds from small pair (?)
3. Get better value with them (?)

We bet for a few main reasons ANY time:

1. Value.
2. To get better hands to fold.
3. To get worse hands to "chase" at poor odds.

1. AA, KK, QQ and AK 3bets are for value. They play well in big pots when they are overpairs or TPTK hands.

2. AQ, JJ, and TT 3bets (and some other hands) allow Hero to represent the "value bet" and take down lots of pots where's he behind or a coin flip to win. There is also good value in these hands as well.

3. As rilla was pointing out, anyone with half a brain doesn't have odds to set mine your ass with pocket 3's. But a lot of them will call. They get pot-stuck, and the payoff they get when they hit their set isn't enough to make up for the times they just have to let it go.

As I understand, you can bet for information, but it's not really needed (Spoony says) until mid-stakes.

A really cool thing about 3betting is that when you get flat called by the PFR, he has either a really narrow range OR a range that is totally dominated by Hero's range. Think about it, if he's 4betting AA and KK and ditching pp's < 99, he's got a range on the flop of TT - QQ, AK and possibly some AQ or ATs+ added in. That's all good for Hero who knows what he has and can attack a lot of flops even when he misses.

Next, think about villain having a wider 3bet calling range. That's even better - every additional hand he calls with is either dominated by Hero's range or a coin flip.

That leaves you, the 3bettor, with a stronger range than villain, with solid value from premium hands, a way to get extra value from hands like AQ, JJ and TT, a way to charge the set miners for the ride, and a much stronger position on the flop.

I'll just mention one final thing. All these great things get totally f**k'd if the action is multiway on the flop, and at the micros you pretty much have to hit the flop to be able to continue if it's rr'd preflop.

[Ragnar4]

You raise 88 in the CO to 2 american dollars, I reraise you to 7.5 american dollars from the BTN. What do you do with 88?

Set mine, assuming you have well over a full buy-in let behind, and fold otherwise?

I seriuosly don't like this answer. seriuosly. no for reals.

The trick to setmining is your opponent has to be willing to give you 15-20 times the amount of the raise. The bigger your stack is in comparison to the stakes you play, the less likely you should be willing to commit your whole stack to AA.

Your opponents range for a 3-bet is much, much smaller than his range for a simple raise, or calling your raise, meaning he's less likely to pay you off, or give you odds for your setmining adventure.

Fold is waaaay better than re-raise which is waaaaaay better than call here. The only way you re-raise is if you soulread something like AK or AQ as his range, or if you have solid evidence that he'd do this with 22+.

[Sup3rM4N]

88 OOP 3bet pot is no fun.
Its disgusting if your plan is to C/F a non 8 flop.

[Stacks]

You raise 88 in the CO to 2 american dollars, I reraise you to 7.5 american dollars from the BTN. What do you do with 88?

Set mine, assuming you have well over a full buy-in let behind, and fold otherwise?

I seriuosly don't like this answer. seriuosly. no for reals.

The trick to setmining is your opponent has to be willing to give you 15-20 times the amount of the raise. The bigger your stack is in comparison to the stakes you play, the less likely you should be willing to commit your whole stack to AA.

Your opponents range for a 3-bet is much, much smaller than his range for a simple raise, or calling your raise, meaning he's less likely to pay you off, or give you odds for your setmining adventure.

Fold is waaaay better than re-raise which is waaaaaay better than call here. The only way you re-raise is if you soulread something like AK or AQ as his range, or if you have solid evidence that he'd do this with 22+.

You have some explaining to do. Because I don't much like your response here. Could be that I just don't quite understand the point you are getting at? Or what your saying is wrong? Or I'm wrong?

But anyways... You say the trick to setmining is being sure your opponent will pay you off enough to make up for the times you miss. That is you must make sure you have enough implied odds to make the call profitable. And you also state that the larger the effective stack is the less likely you should be to commit your stack to AA? While the last part of that statement I definitely do not understand, it seems (from my understanding of what your saying) that your logic is flawed.

I'm not certain of the exact math (although I know where to find it on FTR), but you hit a set somewhere around 7.5:1 (or 11.7%ish). That means if you are certain villain will stack off when you hit, then you can take less odds than the 15-20:1 that is advised for setmining (because he doesn't always stack off in a raised pot). As you state yourself, the range villain is 3betting with is rather tight in comparison with his standard opening range (in most instances). However, you state that because of this he is less likely to pay you off when you hit, when the case should be he is going to be generally more likely to pay you off. Therefore, imo you can stand to make a call given less than the 15-20:1 implied odds if you are certain villain is 3betting a range that will pay you off.

Now the validity of the assumption that a villain will pay you off often enough to make a call +ev (given less favorable implied odds like 10:1), depends on villains 3betting range. If they are only 3betting a strong range, then it could very well be a profitable call if the effective stack is large enough. If they are 3bet bluffing often enough then the call becomes less +ev as a setmining endeavor because they won't pay you the full amount as often as before (however, with some hands it then becomes profitable to call or 4b because you are ahead of his range).

As a general rule I don't think it's +ev to call a 3b OOP (under hardly any circumstances), and it becomes even worse when your hand is a small pp and you aren't getting good implied odds. However, I think it's incorrect to say that it's absolutely horrible if you are getting good implied odds (as it appears you suggested by talking about the larger your stack the less willing you should be to commit your whole stack). But as stated, I don't encourage calling 3bets OOP, and you generally don't have the implied odds for a call if the effective stack at the beginning of the hand was 100bb.

This is of course IMO, and is subject to criticism, debate, or flaming.

[Ragnar4]

The biggest thing I'm getting at is this: The larger and larger the effective stacks become, the more and more incorrect that stacking off with a single pair becomes. If your opponent understands this, you'll never get paid enough to justify 7.50 3bet.

Fnord writes a pretty good article in the digest discussing why we as players want a little more leeway and should expand the amount we want to get for stacking off with a set and why it should be in the 15bb range instead of the previous 10bb range. (Set over Set, your opp drawing out, getting it in as marginal coinflips etc etc)

Lets run this through the grinder shall we?

a 2 dollar standard raise, means we're playing 50nl. a re-raise to 7.50 means that typically the standard effective stack will be about 8x the raise... therefore a fold is the correct play.

The deeper you get, the less correct, stacking off is with a single pair. So if you both have 75 bucks (your bare minimum for a set-stackoff here) your opponents going to figgure out you're trying to bloat the pot and as they say in the poker biz, "Big pots for big hands" His hand is on the very bottom of the "big hand range" if he has AA. If he's a good player alarms will be going off in his head if he's deep, and he won't want to commit any more fundage to the endeavor. 15-20x the 7.50 3-bet is 100-150. In a 50nl game, would you be willing to risk 300bb on AA post flop? Ever?

[Miffed22001]

You raise 88 in the CO to 2 american dollars, I reraise you to 7.5 american dollars from the BTN. What do you do with 88?

4 bet obviously for metagame value

[Stacks]

That's a much better explanation Ragnar. And I definately agree with alot of this. But given the lack of information that rilla's little example presented left alot open to debate. And when OP said that calling is okay if there is enough implied odds, I felt that it isn't horrible under certain circumstances.

A few things. First, the reasoning for wanting more implied odds is also attributed to villain not always stacking off. That's why I said if you know they are 3betting only a strong range for value (or they are crazy and will stack off incredibly light), then you can begin to take less implied odds situations (such as 12:1 or so).

Generally at 100nl on Stars the standard 3b size of a 4xbb open is to 12xbb. So if the starting stacks are 100bb you are getting 2:1 immediate pot odds, with 11:1 implied odds should you hit. So right here you are getting 13:1 on your call. If you are rather certain that villain is willing to stack off postflop with his strong range, then you are getting closer to a +ev call. If the effective stack becomes larger then you are getting closer to the appropriate implied odds you want to make the call.

I just want to say that I don't technically endorse calling 3bets OOP. I'm just trying to state that if the conditions are right that it might be plausible that it could be +ev to do so.

[d0zer]

In a 50nl game, would you be willing to risk 300bb on AA post flop? Ever?

Sure. Safe flop, preflop 4bet resulting in low SPR, mediocre villain.

[Ragnar4]

In a 50nl game, would you be willing to risk 300bb on AA post flop? Ever?

Sure. Safe flop, preflop 4bet resulting in low SPR, mediocre villain.

lawl....

But am I totally off with my analysis there though concerning calling 7.5 bucks to set mine?

[Stacks]

No your not. That's the thing. I'm not saying to call. I'm arguing that I believe that given the fact that we didn't have information regarding stack sizes, reads/stats, etc we can't make an informed decision, and that I believe there are circumstances when you could call and it would be +ev to do so. Under standard circumstances I don't think a call is the most profitable play (100bb, standard villain).