This is going to be a generalization thread.

Villain cbets too much (as most players do) but also double barrels too much (less common). We do not know about their PF or river tendencies. How should we adjust?

My thoughts:

Say we are OOP.
1) On the turn, we xr wider for value and with more bluffs. This is obvious to me.
2) *Never* xr flop. Probably not that controversial. On turn cards where we think V will check back more often, we can form a donk range.
3) On flop, fold more (speculative) hands. This is the most debatable. My thinking is that our most speculative hands benefit most from getting a free turn card. Since we can't count on that, we have to plan to turn many of these hands into turn bluffs. If we don't feel comfortable doing that or already have too many bluffs, we might as well fold them on the flop.

If we are IP, it's even more interesting. I think a viable strategy is to flat turn with a wide range, and punish them on the river with raises or bets if they check back. The more standard strategy would be like the above, by raising on the turn wider with value and bluffs. We can also consider raising flops to get a free card on turn with parts of our range.

Thoughts?

*Bonus: how do I test this out on a sim?

x/c flop, x/r turn

i don't know why you still want a donk range. as far as for a sim, maybe try removing postflop checks from villain's gametree and see how it compares to the normal tree.

I think if villain cbets flop and turn with a high frequency IP, their bluff ratio (as well as hands that are neither bluff or values) is so high, we can just click check/call 2 streets and profit

with speculative bd type hands, im conflicted since those are potential candidates to bucket into the turn k/r range. however if that flush draw doesn't come then it's hard to realize its equity as villain will barrel turn again. i guess those hands should probably thus k/f or k/r flop.

as for bluffs on the turn, i guess flush or straight draws make the most sense. i also like the idea of delaying a couple more top pair type hands. i'm not sure if this punishes those hands that are neither bluffs or values. i also wouldn't over do it since villain for the most part has a hand with showdown.

do we really need to have these bd draws tho? like, we need to balance those with value, but we're giving up a lot with the value side of our range by not delaying until turn. i'd just accept the fact that we can't continue as light and know that we're gonna make up for it by pwning him on the turn with our legit hands and draws.

There is no way to lock a turn action in a lhe hand on pio afaik

Can you not force OOP check and IP bet in pio? Am I misunderstanding it?

I don't know about pio but you can force a % action in monker for stuff like this.

forcing anything is bad in lhe. before people would say I want to play gto my action would make an indifferent to their action. A better way of thinking is that whenever I play gto the other going to leak money in each part of the game tree. Instead of focusing on exploited a spot that will cause more confusion to your overall strategy focus on what solvers are telling you and look at the Ev of each actions that is + or -. Both program let you force a specify leak but it min bet the combos are gigantic and by switching up one spot you are messing up your overall range strategy. There a reason why pre flop it choose to 3b a wider range( when ev are about the same to flat) then expect so that post flop it can use the lesser range to do certain things with them.

Also another thing people don't realize is that when you force an action in solvers what it does is that it recalculate the gto of the action from the force mistake. So now you going to do even more guessing game of what you think your opponent are going to react so that you can force more action in solver to make it accurate. Maybe if you playing nl where a spot can be time X the min bet of a mistake but why do it for min bet lol

Obv by playing straight GTO, we will profit. But when their mistakes are so extreme, it makes sense to adjust to profit even more.

Are you not making adjustments to obvious bad flop tendencies? You should definitely alter your default strategy if Villian is completely passive on the flop vs competent; if they cbet 100% vs cbet 70%. This idea should translate to obvious turn tendencies. Someone who cbets 100% and dbarrels 50% requires a different strategy than someone who cbets 100% and dbarrels 90%. These are very common player profiles.

I don't mind deviating from Nash strategy and being exposed to counter-strategies. Most villains will not be on that level.

The adjustments you’ve suggested in the OP are correct. A friend of mine figured out the EV of delaying to the turn OOP vs habitual double barrelers and it’s the correct adjustment. You’re on the right track.

My adjustment to any type of player is to make sure I am playing the correct mix %. That already is super hard to remember. And in a random match regs will make quick adjustment to my randomize mix seeing that rng happen to hit 15% c/r and will all the sudden try to "exploit" that and make gigantic mistake . The bad players are leaking so much already that they don't last long for me to worry. FYI there a huge discussion in the solver chat about this adjustment like "forcing" which isn't the correct way to do it. However if you really insist in knowing how to do this I can show you exact how to do it step by step with either pio solver or monker for a X amount

Also my guess is that people aren't even playing optimal to any flop cbet % in the first place because the optimal strategy require you going to get something like c/r 80% with certain hands and people will go well it close to 100% so I am always going to c/r this. Or a more common one would be something like this ranges should be c/r 20% and people will never c/r it. If you can't see why that isn't good and you want to focus on "exploited" spot which takes even more time to learn.

Can you ask your friend if he like to play some head up match if he isn't azrarn or yoghii and kpr16

DonJuan you guys are talking past each other a little bit. You are describing how to play against a smart adjusting player and not to overdo it (which I agree with), phunk and others are talking about how to adjust to bad live players with huge leaks and static strategies, and saying that understanding how to optimally adjust to them is hugely +EV (which I also agree with). Solvers are very good at this sort of thing, so I think his question is natural.

I can certainly relay the message to him.

Also, this.

Btw no idea if your friend or good or not I am just bored sitting waiting for games. I would offer this to anyone that reading this thread that not the same player I wouldn't play.

Right. Once the opponent makes the mistake of betting the turn too often, we're no longer playing the strict Nash Equilibrium; After the mistake happens, then there are new comaximally exploitive strategies. This presumably will cause an ev shift in our favor(if the deviation did not cause such a shift, then it would not necessarily be a mistake).

----

Also if counter exploitation by the opponent is only a minor concern, then my adjustments will be dependent on a few things: the composition of the opponents range, what he will call down with, what he will 3 bet with, and what he will fold.

If counter exploitation is a major concern, then my adjustments will be done on the river, or not at all. The reason I'll save my adjustments for the river vs some players is because the option to end the hand closing the action is exactly where your opponent has the hardest time counter exploiting your deviation.

yeah i think it's hard to realize the equity with those speculative hands (like ~30% equity with very very little showdown value that can improve on the turn to some sorta draw, eg. QTs on Kxx) and i see merit to needing values to balance with flop k/r. maybe the correct adjustment is to just fold them. but then again, why should we balance against players who are incapable of adjusting. the way i see it is this part of our range is a function of how showdown bound villain is. i guess another way to look at it is since these hands are mostly used to balance our value hands and we have no value hands on the flop, the correct strategy is just to fold them. i'm not sure which thinking is more correct and actually in tune with the game since im really at this point just trying to mimic a bot but not understand the reasonings behind the things it does (maybe there are no human reasonings and it is just pure calculation).

----

i'm going to assume a hypothetical optimal is somewhere around cbet 80% flop and cbet 80% turn (64% of his continuing range by turn). if villain cbets 100% flop and cbets 90% turn (90% of his continuing range by turn), the % in combo difference between the former and ladder is 26%. If villain opens 400 combos, he'll have 256 combos if he cbets optimal and 360 combos if he's a double barreling machine.

there is a 104 combo difference and i'd guess the vast majority of this are hands in the complete air to having 50ish% equity range. i'm not looking at a solver for reference so i'm pulling numbers out of my ass, but let's assume the board is 8642. Villain's gonna double barrel almost all Ax, all air, all draws, all gutshots, all pairs with 5 outs to improve, and almost all pairs with 2 outs to improve.

some advantage to villain's strategy is it can get hero to mistakingly overfold his own Ax type hands as well as deny the equity of those speculative hands. villain also get to make more with his stronger hands since none of those are checking back the flop. Ax type hands are going to make up the largest part of hero's range so it makes sense to figure out how to correctly play these hands first. i don't know how really. my guess is just k/c down more and gamble more since the majority of villain's range is also Ax type hands (as well as more bluffs). It just depends who's x is bigger.

for villain, the # of air is probably a dozen or so combos higher than what it should be, while those Ax type hands are probably going to make up for the majority of the double barreler's range.

by delaying 100% (i'm skeptical on this but i wouldn't be surprised if this is the max exploit strat against a habitual double barreler who wouldn't adjust), i assume we're aiming mostly to punish those ace high type hands on a bigger street. i don't see any point in punishing their bluff portion as i think such a player is just going to triple barrel bluff close upto a dozen more combos than they should on river (altho there are times they spike their 6 outer so it's not completely dead ev).

i rather actually just deviate very little from what i think optimal is since i believe most villains are not goign to keep habitually double barreling. in fact i believe most villains incorrectly over adjust once they see a few hands because that is just human nature. but maybe there are some with a hardwired strategy that they've been using for years and cannot and will not adjust. maybe against those, it is correct to delay 100%??

is what i said logically sound? i haven't really done any work or formed any conclusion. i just want to figure out why
1) delaying 100% is sound, and
2) why folding speculative hands is forced

with raw human thinking.

2468 is a bad (although extreme example since double barreling is a huge deviation from optimal). A better example would probably be J952

Delaying way more to turn seems very standard. Really more interested in discussion about (3) and IP delay to river strat.

I nerd out on this stuff... so here's my 2c ^_^

e.g. "Minimally Exploitative Strategy" or MES (you see this in PIO accuracy) and is the name of the solution provided after nodelocking

e.g. "Maximally Exploitative Strategy" most explo regs learn intuitive poker w\ this style of exploitative play sprinkled into spots they learn to recognize opps are making large fundamental mistakes in.
1) This style teaches opponents what they do wrong
2) This style leaves us wide open to counter exploitative play

Example: You're playing any game... lets use RPS... and someone has a bad strategy. 20/20/60 rock/paper/scissors w\ a static strategy.

Max Exploit = throw rock every time
--- 1) He quickly notices you are obliterating his strategy + most will look to adjust. (You're developing metagame via this approach)
--- 2) He can easily start throwing paper and your strategy is getting crushed. Especially if he adopts a MES vs your approach and you can't tell it's going on

MES = way better ^_^ still hard counters the opps mistakes... but he never catches on + we're not anywhere near as vulnerable to counter-exploitative play, adjustments, or metagame progression.

Because there is massive edge to the correct execution of MES vs a known opponent frequency deviation from the correct GTO %s! - even in structured betting games

While at the same time it's super important to understand what GTO looks like in all spots for both players....
1) It's how we identify what people are doing wrong
2) It's how we can specifically attack the spots they are making the biggest mistakes in (and take them more often to those spots!) -- If you knew someone sucked against a specific chess opening, you would clearly take them into that line more often.
3) It's how we can understand application of MES to explo opportunities we were able to recognize in the first place due to our understanding of GTO!

This is HUGE in big bet games... and still has a massive impact on limit structures.

The idea that pure GTO is > GTO+MES is fundamentally ludicrous. The real skill is knowing how to correctly oscillate between the two as metagame develops imo!

cbet turn 80% vs 90% shouldn't really be all that noticeable esp live, so i think your assumption's wrong in freq deviation. i'm not saying it should be delay 100% either. for instance, i prefer to c/r coordinated or boardlocked boards. so i'd just raise the threshold of coordination or boardlockness for c/r'ing, call down lighter. i mean, i don't even know if my normal strat or freq's are even good, so i try not to make any radical adjustments from what is already just a mediocre wr.
like, if you're folding all bdd's and delaying 100%, it's only a matter of time before you're being exploited.