Slowplaying, Valuebetting and More GTO Stuff
Posted by Daniel Dvoress
Video Player is loading.
This is a modal window.
The media could not be loaded, either because the server or network failed or because the format is not supported.
Resume Video
Start from Beginning
Watch Video
10
Loading 24 Comments...
35:30. You say because IP has more than enough bluffs to fill his bluffing region, OOP must call at 1-A in order to make those bluffs 0EV otherwise BTN will just jam all of his bluffs and show a profit.
How does OOP's strategy change in situations where IP can't fill up his bluffing region? Does he just respond by folding all of his pure bluff catchers and prioritizes calling hands with positive removal effects or hands that are beating some of IP's value betting region?
Hi jd,
Good question. If the IP player can't "fill" his bluffing range the OOP player just has to fold a bunch. Keep in mind that a GTO solution is just a set of two max exploitative strategies. So in this spot where you are shoving for pot OOP is just going to call whatever has 33.3%+ versus IP's shoving range and fold everything else. IP can't make an adjustment in terms of his range to this because he is already shoving all of the bluffs that he can shove.
This shouldn't happen very often though - maybe only on some very rare equity changing rivers that are really good for the IP player. If it does happen chances are your strategy on previous streets is off somewhere.
jd - Naturally, the answer is that how OOP should adjust his strategy depends on how much IP is underbluffing. In the extreme example where IP is bluffing never, then OOP should only call hands that beat 34%+ (getting 2:1) of IP's value bets and fold 100% of the hands we think of as bluffcatchers. But, using Daniel's CREV tree, if IP is bluffing only 10% (instead of 12.5%) and value betting 25%, OOP will fold some of his worst bluffcatchers. You'll find that blocking IP's value region is a better way of selecting which bluffcatchers do what. If IP is bluffing even close to a good frequency, hands that beat even small parts of his value region become slam dunk calls.
Daniel - Nice video, thanks for the illustration. Lil blooper at 68m: 4c2c makes two pair otr, which is why it's a way better call than 7x. The hand selection stuff you're discussing is still visible in the trees, though - CREV prefers to call the weaker A7 over JT/QT, for all the reasons you mention. Thanks again!
Woops, good catch!
if you are betting this river for value can't you be exploited by someone over folding preflop? You have oversimplified this to a 1 street decision.
Unless you think that some of the hands in the BB's peeling range are -EV and shouldn't be peeled, then no that is not going to be a good adjustment. OOP can't improve the EV of his range by folding hands that would be +EV to call.
If you mean that OOP will have a higher range EV on the river if he plays a tighter range preflop, that is true, however he gets to the river less often.
Hi everyone,
CREV tree for the video can be found here:
https://www.dropbox.com/s/3u6g44rrqqhpgi5/GTO.concepts.tree.stx?dl=0
I think that it's extremely likely that with s marginal hand like this, at equilibrium, a mixed strategy of some bets, some checks is appropriate on the river. If that's true it means that if OOP is playing a GTO balanced strategy, betting and checking are equal in value. To exploit in positions tendency to bet with too high a frequency on the river, OOP can make sure he gets to the river with a stronger range by over folding preflop. Just because a peel is neutral EV vs an equilibrium strategy doesn't mean it's neutral vs a strategy which bets too much on the river.
Keep in mind that if the OOP player chooses to overfold on previous streets he gets to the river less often. Being to a certain spot with a range that is more +EV is not the same as overall EV of range improvement if the frequency with which you got there changed.
Instead of thinking back to preflop, take it just one street back. Let's say BB knows our river strategy so he chooses to continue with only 2nd pair + instead of 3rd pair +. In this scenario a K with a bad kicker would be losing a lot of money by betting the river, but we get to the river less often and BB would be folding the turn more than 2/3 (!) of the time to a half pot bet.
Hi Daniel, great video on a number of levels... I really like a video like this which helps you to simplify the process of determining what to bet with and what to call with when you are considering your range and villain's range... Also, love the demonstration of 3 different ways to choose what to call with and helping to understand how much EV different choices could be costing you... Really enjoyed the demonstration of how you use CREV and some of the latest new additions to it... I did run this scenario on Pio, and one interesting point was that Pio likes a pot size bet on the turn a lot more than the half pot bet size with most hands...what do you make of that? Also Pio seems to like check raising on the flop a lot with hands like KQ and KJ by BB... Would love to see more videos like this...
I think the flop c/r with KJ+ (as well as way weaker "value" hands) is good. I just don't think people really do it so the assumption that BB only check/calls I think is OK.
Re: Piosolver liking a pot sized bet on the turn - I've seen this a lot in similar spots, I'll get back to you once I run the sim.
Hey,
I did not get the same results. I have BU preferring to bet turn half pot compared to pot with most of his range, whether or not BB is allowed to x/r the turn. What kind of parameters did you set for the tree? If you want you can share it and I'll take a look.
Hey Daniel, great video. I have a question you might shed some light on. When you had your strategy of betting 25% value and 12.5% bluffs and you face a villain that starts calling down more from 50% to 65%. Is there any adjustment you should look to make in that spot like value bet even thinner and expand your value range while cutting down on your bluffing range to increase EV or would you let your strategy be as it is and why? I understand that the strategy will not be exploitable by villain, but if he will be calling more than usual, I think you could increase EV by looking to value bet even thinner and would it be something you want to do or is it better to just keep ranges as they are in real scenario?
Keep up the great work, love the videos!
He Everize,
If you know that villain is going to call 65% on the river then the adjustment that will net you the highest EV in that spot would be to have a bluffing frequency of 0% and valuebet any hand that has 50%+ equity versus his 65% calling range.
Hi Daniel
Thanks for the video, very clear :)
at 45:35
On the river we shove for pot but since we are not perfectly polarized you said we can't bet 1 bluff for every 2 valuebets.
why ?
The job the range construction here is to make the weaker bluffcatchers of vilain indifferent between calling and folding. If less than a third of our river shoving range is not bluffs then vilain can increase his ev by changing his own strategy.
Or did i miss something ?
Thanks
Hi Ben,
Nope, you didn't miss anything, what I said was just wrong. I think I was thinking about villain not having to call half of his pure bluffcatchers on the river (since he just has to call half of his range and a part of it is "traps" and for some reason converted that in my head to the opposite/inverse of that. Everything you said is right!
at 48
You say bb is capturing a "pretty bad fraction" of the pot - given that he has 55% equity edge and he gets "only" 43% of the pot on the river before BTN acts.
For such a weak preflop range arriving into this spot I feel like getting more than 30% is actually amazing !
In a perfect NvBC spot, with both ranges having exactly 50%, BTN should be capturing at least 3x as much of the pot iirc.
Why do you think this 43% is bad in this case ?
Or rather, how do you judge if a % of the pot fractured is good or bad for any player on the river ?
Because to me, BB getting 57% of the pot when we delete his turn checkshoving range sounds like it should not happen. It looks like BTN is doing something wrong somewhere for letting the bluffcatcher player capturing such a large fraction of the pot on a blank river...
I think 43% is bad in this case because it is such a common part of the game tree. Unless the BB folds at some point in the hand this bet flop/barrel turn scenario is how we are going to get to the river most often. So it's a combination of realizing a diminished amount of equity and a high frequency of arriving to this spot. If, for example the flop was not c-bet with a 100% frequency I don't think it would be as bad because a) it would be a less frequent in the overall game tree; b) BU would have a stronger (and more polarized!!) range.
Having said that, like you imply my judgement of the % of the pot fractured is subjective so I could be wrong.
Thanks for the answers Daniel
The comments regarding CFs and frequency of the spot is very interesting and new to me.
So the more often a spot happen, the more a player should have a strategy that grants him a fraction of the pot close enough to what his equity would (should he realize 100% of it) ?
What about postflop situations in spots where preflop ranges are quite asymmetric ? Like BB vs MP ? I mean we could have a flop texture like 553r where MP cbets nearly 100% but this doesn't sound like a situation where BB gets a fraction of the pot close to his equity ...
It seems very hard to judge how good/bad a fraction of the pot captured is compared to a player's equity. How can we quantify/determine it more scientifically ?
Hey Daniel,
In your BB preflop defending range you have every suited combo being defended at 25bb including a lot of 2x. Do you think this is accurate in the current state of the game? Would you fold these hands against a very good player? Do you feel that you are going to be in a lot of very difficult spots at this stack depth with these type of hands? Does your 25bb stack have more value being able to open wide, 3b jam w/ fold equity, etc as opposed to playing a pot oop with hands that don't flop well?
I understand that the current state of NL has people defending their bb super wide getting a great price. Is it possible people are starting to defend "too wide"??
Hi,
I think BB defending ranges vary a lot even today, but the range I put in is I think representative of what I would expect good regulars to defend. I wouldn't defend tighter than this even against very good regulars - the odds you are getting are just too good. I do expect to be in difficult spots but I don't see that as much of a discouragement against defending.
Generally I think things like the value of X stack being able to do Y, Z are overestimated. Your stack is currently 25bb, if the EV of a defend >25bb you should defend (early in a tournament, with no ICM considerations etc.) There are "cusp" stacks (which allow you to have FE, allow you to 3b/fold, allow you to 4b, etc...) but normally they require a special situation [where either a) you can be exploitative because of the opposition; b) ICM considerations favor a particular stack] for the EV of a certain stack to be higher than the EV of some other stack. There are also some situations where the table is so soft that just surviving and avoiding small edges with good players has merit, but that normally requires a very soft field.
I have on occasion see people defend too wide, but it's rare. I think it's very hard to defend too wide when there is an ante and you are facing a small raise.
Thanks Daniel. To follow that up, how does the raise size affect your defending range? For example, if people start opening 3x/4x will that significantly tighten up your defending range. Can we exploit people who defend too wide no matter the cost by opening bigger? Is this motivation to open bigger on bb's who are extremely good? If we're opening a tighter range should our preflop opens be bigger? Just trying to counter current trends in the game to figure out how best to exploit. Appreciate the insight.
Hi,
Generally the better (wider) BB will defend the more motivation there is to open bigger. Against someone overdefending you should definitely open to a larger size with a tighter range. It's very hard for people to make much of a mistake against a minraise in an ante game since defending anything at all will be +EV.
Hey Daniel, good stuff. Just one note: I personally would have liked to see some kind of summary of the steps you went through building this cEV tree to caculate how thin to valuebet after you finished the tree. Because it took you quiet some time to build that tree and you did a lot of stuff so some basic summary of the steps at the end would be nice because now I probably just have to watch the whole video again. Maybe you could do that in future videos :)
Anyway, I liked it a lot!
Be the first to add a comment
You must upgrade your account to leave a comment.