Called 3bet preflop question
Posted by e-mal
Posted by
e-mal
posted in
Low Stakes
Called 3bet preflop question
Something I've noticed is that I have a hard time navigating my way through pots where I have called 3bets preflop. I was taking a look at my database by position and realized I'm down quite a bit in this kind of spot. I was wondering if that's a position where I should be down for the most part overall, and if so, what is a reasonable bb/100 to be down here? Here's the stats on it, sample size is kind of small.
Loading 16 Comments...
I think I saw somewhere that between -45bb to -50bb is a good number to shoot for when filtering "call PF 3b". Not sure though. It came from something Ryan Fee did.
Considering you open 2x-3x , 131 is not so bad, its better than folding
Hi,
I put a filter for when I call a 3bet BTN v SB and my overall winrate is -98 EV bb/100. When I look at the different hand classes in my calling range, basically everything is better than folding apart from 22-55 (which might be a small sample or me making some mistakes). At present, I'm mixing fold and call with what I thought was the more marginal stuff, but doesn't this mean I should keep calling the more marginal hands until their bb/100 is just better than folding?
Would this then mean that my winrate when calling a 3bet will go down, but my winrate when facing a 3bet will go up?
At first get big database to analyze
That sounds right, if it is better bb/100 than folding you should always call all hands until they start to drop. Essentially some of those will drop in bb/100 and gonna be breakeven(fold/call) if you play smart regs or higher limits.
Would this then mean that my winrate when calling a 3bet will go down, but my winrate when facing a 3bet will go up?
If I understand correctly, your winrate when calling 3bet will go up(still gonna be negative), and overall winrate will go up too. However keep in mind playing OOP in 3bet pot can't make you a lot of money(and sometimes you can lose more than fold) +++ rake is huge issue to be defending wide
Number of samples isn't big enough to conclude anything useful. Averaging over all hands just is killing what you actually want to know and that is are your calls with marginal hands profitable or not. When you e.g. would call 3 bets also with AA then your average will shoot up a lot but that has nothing to do with the profitability of calling with marginal hands.
Also OOP vs IP is a big difference so your CO stats does not say if your IP or OOP when you call the 3 bet.
Yeah I was purposely trying to capture the stat as a whole and not narrow it down. Otherwise I would have filtered for OOP vs IP.
I wish I could be more helpful, but the standard deviation in bb/100 is very large (think ~1000bb/100), so you'll need tens of thousands of hands for the winrate here to converge to something useful.
As others have said, you have to compare the winrate to folding, not 0 BB/100. This number will mostly be negative, just like you mostly lose money from the blinds. The aim is to make it less negative.
Also, as Tyler Forrester said above, the std deviation is very high, so you shouldn't be using this stat to make drastic changes.
So what's the best course of action for analysing our call v 3bet range given the samples needed for a particular spot are so huge? Maybe if our calls seem to be doing well, slowly defend a little wider and then see how we are doing after a big sample? I still can't get my head around how adding in more slightly winning calls won't reduce the win rate of the calling range.
For example, if I 2.5x the BTN and my call v 3bet range v SB is exclusively AQs, which has a win rate of -50bb/100, then the bb/100 of my calling range would be -50bb/100.
If I then add in 86s, which has a win rate of -200bb/100, and I only call these two hands, the win rate of the new calling range would be -125bb/100. This is worse than the bb/100 of the previous calling range, but calling 86s is clearly a profitable play.
Before, when facing a 3bet with these two hands and only calling AQs, the bb/100 would have been -150bb/100 I think. Now, it is -125bb/100. But I can also imagine a situation where us defending more versus 3bets results in us getting 3bet less, in which case, the win rate gains would show in our BTN RFI bb/100? Is that how it would work?
Hope OP doesn't mind me rambling in his thread, but I was looking at this myself and despite playing a reasonable number of hands, I still feel pretty stumped when looking into these numbers in particular. The comments in this thread have been super useful though :)
You are correct that making some gains in one particular stat could affect some other stat altogether. However, in this case, your particular concern isn't a problem.
Suppose, you have only one objective: to maximize winrate in 3-bet call spots. To do this, you add in more and more marginal hands.
Now, suppose that you're correct in that because you defend a lot against 3-bets, you get 3-bet less. This is a good thing in itself: the less you get 3-bet, the more equity you can realize postflop.
So, you extremely narrow objective leads to direct upside and indirect upside.
Of course, this story is too simple. There can be other adjustments which people can make: as you call more and more 3-bets, your opponents will tend to 3-bet you with a linear range rather than a polarized range. So, in the process of getting a better winrate for this stat, you'll have to study on how to deal with your opponent's different range.
Finally, keep in mind that just because calling 3-bets is with a hand is profitable doesn't mean that it's the most profitable play. 4-betting could be more profitable still.
Hi belrio. Thanks for your detailed comment, it is very much appreciated.
I'm still pretty confused by the idea of maximizing the win rate of our calling range v a 3bet. In your comment you state:
Suppose, you have only one objective: to maximize winrate in 3-bet call spots. To do this, you add in more and more marginal hands.
But doesn't the AQs/86s example I gave above show that adding in marginal hands decreases the win rate of the calling range (assuming I worked it out correctly)? For example, my win rate calling 3bets BTN v SB is -98bb/100. If I add in more marginal (but +EV) hands, I would expect my win rate when calling to go more negative, but my overall BTN win rate to increase?
Maybe an analogous situation is defending the BB? A very tight player who only calls strong hands would increase their overall win rate from the BB by calling more marginal hands; but this would likely result in the win rate of their calling range decreasing. Isn't defending our open when we RFI on the BTN a bit like defending the BB?
The point I'm trying to get at is, is it even really worthwhile to look too much at the overall win rate when I call a 3bet in a certain spot (provided it isn't extremely low) since it isn't really giving a the full picture of what's going on? Is is better to just try to maximize the win rate of every hand in my calling range, but be pretty happy if I look and see that the worst hands in my calling range are about breakeven since I can be somewhat confident I'm not folding hands that are profitable calls?
Thanks again for your time :)
Sorry, you are correct that the winrate in "3-bet/call" pots will decrease, but your overall winrate in "3-bet" pots will increase. The latter is the one which matters.
Ah right. So I'll try not to focus too much on this stat for now as long as all my calling hands seem to be doing well. Thanks!
Not at all. This is the kind of conversation I'm trying to get out of my RIO subscription. I really appreciate it. I have all the same concerns you do.
I usually like empirical models for these type problems, because they are more attuned to player pool tendencies. But since the observed data here is so shoddy, I'd suggest using solver simulations for these ranges (with extra EV if you can find some flaw in the player that allows you to call wider).
That makes sense. Thanks Tyler!! I'm a big fan of yours :)
Be the first to add a comment