In spots like these it sounds like you sort of throw the idea of defending alpha of your bluff catchers on the river out the window and prioritize only calling with hands that have positive removal effects. (It seems like a ton of your river range will be the AsQc handclass by the river, so if you wanted to hit 1-a you'd have to call with some). Is this coming from the assumption that Isildur is going to be relatively balanced in this spot and as a result you can just fold those hands with neutral or negative removal effects since they're 0EV or worse?
I've been doing some work with trying to get close to alpha defends on the river and I've noticed my River Call Efficiency in both NL and PLO is very bad (like 1.0-0.9). So it seems like either A) People are under bluffing and I should fold hands with neutral removal, B) People are balanced but I'm calling with more hands than I should be calling theoretically, C) I'm calling with too many hands that have poor removal effects, or D) It's a sample size issue and I'm just running poorly.
Just curious to hear how you approach this situation in general.
First, I think we can defend somewhat less often than 1-A in this situation because of removal effects for overbluffing and because all combos have substantial EV checking turn on this texture.
Second, I agree with your points partially and I think I overstated how picky I can be defending here. I will not hit a reasonable defense frequency only calling spade+PR hands here, and I think I made a comment that suggested that. I think I want to unblock the region 4-Q here and having a spade is also good. AQ+Qs is kind of marginal on these criteria but is probably a fine call with some frequency.
11:20 3b pot Q9ss you check 988r and say it's a board where you have to do some checking you can't just bet your entire 3b range on the flop.
Just really quickly, what are the general guidelines for situations where you can play the 'auto bet flop with range' strategy? Is it just a function of range equity advantage and nut advantage or do you break it down into specific hand classes and think about how each one of those could max their EV?
For instance on a board like AK7r in a 3b pot, the PF 3bettor certainly has the equity and nut advantage on this flop, but there's still this fairly large subset of Kx-7x pair hands that might not be maxing their EV by betting so we prefer to play a checking range and have to balance that with some stronger holdings.
I think it's important to try to maximize each hand class's EV as much as feasible. On a board like AK7r we might have higher raw equity than on say 223r but the protection on 223r will shift the equilibrium to an auto bet while on AK7r we have some obvious slowplays like AA and some obvious Xing hands like QQ.
Last hand- If you are really 3betting around 38% at this stack depth (~ 80bb) then 4bet shoving A10o is really positive EV and I would guess much better than flatting. Especially because he 3x the button and then you made it 9x. You even said you were folding 55 to a shove so imo he made a great play just unlucky you had 99.
I think it's a bad to assume a 3b stat that's far higher than the population averages at these stakes and is compiled over varying stacksizes is accurate.
I agree but I watched all three videos and you were certainly 3betting a really high amount. Even if you are in the mid 20% then a 4bet shove with A10o will still be really good
Edit: I ran the numbers and I was incorrect for 25%. 4bet jamming A10o here is slightly losing vs 25%
Maybe I'm missing something, but I'm not getting the same numbers as you.
Vs 3/8.5/80, where I'm 3betting 25% (linear for simplicity), I'm getting a slightly positive EV for IP jamming ATo+, 87s+, A5s, A7s+, 22+, T8s+, K7s+. It seems like all of these weakest combos besides perhaps the small pairs are winning more by calling than by shoving though so there's no danger from indifference.
If I'm 3betting 30%, I'll need to call roughly 55+, AJo+, ATs+.
Anyways, it seems like for 25% 3b I need to call something like 77+, maybe some freq of AJo/ATs, and AQo+, AJs+. Adding pure calls with AJo/ATs and either mix or pure call with 66 as I increase towards 28-30% 3b. I don't think I was too far off these numbers.
In any case, ATo isn't at indifference between 3b and call unless I'm way overfolding or if I'm 3betting like 35%+ and only calling 55+, A9s+ to jam. I guess I still just think Isildur was a bad Bayesian here and made an inefficient shove.
Hey Ben- Sorry I was incorrect for my estimate of 4bet jamming vs a 25% 3bet. it is slightly losing to jam A10o.
But during the video you were 3betting a lot so my guess is you were playing a style in the low 30's. In that case his jam with A10o was a reasonable play
I really enjoy the series, pretty nice from You to bring up a footage with some legendary high stakes player like Isildur.
HU videos are great because lot of action allow me to learn a lot, while listening to Your thoughts, although I play mostly 6max, but the point is to I can see and hear Your train of thoughts, great ideas, Your thinking process.
So I'm pretty happy any types of videos what You like to make in the future, just slightly prefer cash game variants.
Have you changed the way you play this HU tournament structure from a cash game structure ? As you stated on your ICM intro video, ICM should impact postflop equilibrium. As chips increase in value, the cost of opportunity of losing a pot, either small or large, also increases.
This should make IP (for example 37:36 you decided to bet 33 on Jd9h2d2h for protection/value) more inclined to protection bet turns even when protection isn't as important if OOP plays a high betting frequency with FD/SD OTT. The easiest counter adjustment for OOP is to start calling turn bets more often. If OOP does that, the larger the pot OOP loses, more his tournament EV decreases (assuming IP doesn't value cut himself OTT). How do you see the impact of chip utility on post flop equilibriums ?
ICM has no implication in HU (and winner takes) all tournaments. Obviously, since you have to win the match to step up in the prize, so it's the same as a winner takes all "mini" tournament.
Hey Ben, great series very interesting spots and excellent analysis.
Would like to ask you in what spots you like to develop a donking strat. You did no donkbets at all against isil and im assuming it is because he has a fairly high cb frequency, and donking is usally good against low cbet opponents.
However I've seen some strategies where develop donkbetting into our range even against a high frequency cbettor seem to be successful. What do you think about this?
I'm pretty sure donkbetting is part of optimal strategies but it's low on my to-do list of things to implement. The work I've done suggests it's not contributing much to EV and adds a lot of complexity.
At around 15:00 don't you think it's better for Isildur to be hero calling with Jx (and the like) than 7x? Jx blocks more of your value check raising range. Great video :)
Good point. It might be a mix between these hands depending on how well the second card blocks strong hands (for example KJ might be a call). If villain somehow calls too many Jx over 7x then I have an exploit to XR 7x myself getting value and pushing villain off the chop which is obviously a disaster for villain's EV.
I think he'll prefer to bet hands with better blockers to my river calling region. Always bluffing this river (or bluffing extremely often) is a bad strategy.
great video.
With AQ at 9min, having any club can also block bluffs cant it? Like is having the Kc/Qc even that great bc he might barrel a hand like this (Qc8x Kc7x) and when he bets this size on river, it seems like he almost always will want to have a club in his hand. And since there are a lot more combos of Qc7x-Qc9x and Kc7x-Kc9x than there are Kxcc Qxcc flushes, than arent we maybe blocking more bluffs than value bets? I think having the 7c might be just way better than it actually seems bc ppl dont usually barrel/double barrel 7cX but there are still tons of 7xcc value hands.
I think it's going overboard to assume he'll always have a club in his range to bluff here, solver results typically show a more blended bluffing strategy for wide range bluff spots like this. Generally there's a lot of blocker blending on both sides so as not to bleed info to a clairvoyant opponent.
How does PT4 calculate the stat PFR? if you look at the numbers for both you and isildur (at time 4:57) the numbers are 61 and 55 respectively.. but you guys have had a strictly open raise strat (no limping) so far when on the button, and you've certainly been playing more than 61% and 55% of hands from the button... so what's up with that..?
Loading 24 Comments...
9:45 AsQc facing a river bet.
In spots like these it sounds like you sort of throw the idea of defending alpha of your bluff catchers on the river out the window and prioritize only calling with hands that have positive removal effects. (It seems like a ton of your river range will be the AsQc handclass by the river, so if you wanted to hit 1-a you'd have to call with some). Is this coming from the assumption that Isildur is going to be relatively balanced in this spot and as a result you can just fold those hands with neutral or negative removal effects since they're 0EV or worse?
I've been doing some work with trying to get close to alpha defends on the river and I've noticed my River Call Efficiency in both NL and PLO is very bad (like 1.0-0.9). So it seems like either A) People are under bluffing and I should fold hands with neutral removal, B) People are balanced but I'm calling with more hands than I should be calling theoretically, C) I'm calling with too many hands that have poor removal effects, or D) It's a sample size issue and I'm just running poorly.
Just curious to hear how you approach this situation in general.
Great comment.
First, I think we can defend somewhat less often than 1-A in this situation because of removal effects for overbluffing and because all combos have substantial EV checking turn on this texture.
Second, I agree with your points partially and I think I overstated how picky I can be defending here. I will not hit a reasonable defense frequency only calling spade+PR hands here, and I think I made a comment that suggested that. I think I want to unblock the region 4-Q here and having a spade is also good. AQ+Qs is kind of marginal on these criteria but is probably a fine call with some frequency.
11:20 3b pot Q9ss you check 988r and say it's a board where you have to do some checking you can't just bet your entire 3b range on the flop.
Just really quickly, what are the general guidelines for situations where you can play the 'auto bet flop with range' strategy? Is it just a function of range equity advantage and nut advantage or do you break it down into specific hand classes and think about how each one of those could max their EV?
For instance on a board like AK7r in a 3b pot, the PF 3bettor certainly has the equity and nut advantage on this flop, but there's still this fairly large subset of Kx-7x pair hands that might not be maxing their EV by betting so we prefer to play a checking range and have to balance that with some stronger holdings.
Jd,
I think it's important to try to maximize each hand class's EV as much as feasible. On a board like AK7r we might have higher raw equity than on say 223r but the protection on 223r will shift the equilibrium to an auto bet while on AK7r we have some obvious slowplays like AA and some obvious Xing hands like QQ.
Last hand- If you are really 3betting around 38% at this stack depth (~ 80bb) then 4bet shoving A10o is really positive EV and I would guess much better than flatting. Especially because he 3x the button and then you made it 9x. You even said you were folding 55 to a shove so imo he made a great play just unlucky you had 99.
I think it's a bad to assume a 3b stat that's far higher than the population averages at these stakes and is compiled over varying stacksizes is accurate.
I agree but I watched all three videos and you were certainly 3betting a really high amount. Even if you are in the mid 20% then a 4bet shove with A10o will still be really good
Edit: I ran the numbers and I was incorrect for 25%. 4bet jamming A10o here is slightly losing vs 25%
Maybe I'm missing something, but I'm not getting the same numbers as you.
Vs 3/8.5/80, where I'm 3betting 25% (linear for simplicity), I'm getting a slightly positive EV for IP jamming ATo+, 87s+, A5s, A7s+, 22+, T8s+, K7s+. It seems like all of these weakest combos besides perhaps the small pairs are winning more by calling than by shoving though so there's no danger from indifference.
If I'm 3betting 30%, I'll need to call roughly 55+, AJo+, ATs+.
Anyways, it seems like for 25% 3b I need to call something like 77+, maybe some freq of AJo/ATs, and AQo+, AJs+. Adding pure calls with AJo/ATs and either mix or pure call with 66 as I increase towards 28-30% 3b. I don't think I was too far off these numbers.
In any case, ATo isn't at indifference between 3b and call unless I'm way overfolding or if I'm 3betting like 35%+ and only calling 55+, A9s+ to jam. I guess I still just think Isildur was a bad Bayesian here and made an inefficient shove.
Hey Ben- Sorry I was incorrect for my estimate of 4bet jamming vs a 25% 3bet. it is slightly losing to jam A10o.
But during the video you were 3betting a lot so my guess is you were playing a style in the low 30's. In that case his jam with A10o was a reasonable play
how do you find what is the best stretagy for pre flop play? what tools do you use? what videos can i see to learn by myself?
Thanks for the video Ben!
I really enjoy the series, pretty nice from You to bring up a footage with some legendary high stakes player like Isildur.
HU videos are great because lot of action allow me to learn a lot, while listening to Your thoughts, although I play mostly 6max, but the point is to I can see and hear Your train of thoughts, great ideas, Your thinking process.
So I'm pretty happy any types of videos what You like to make in the future, just slightly prefer cash game variants.
Hey Ben, great series.
Have you changed the way you play this HU tournament structure from a cash game structure ? As you stated on your ICM intro video, ICM should impact postflop equilibrium. As chips increase in value, the cost of opportunity of losing a pot, either small or large, also increases.
This should make IP (for example 37:36 you decided to bet 33 on Jd9h2d2h for protection/value) more inclined to protection bet turns even when protection isn't as important if OOP plays a high betting frequency with FD/SD OTT. The easiest counter adjustment for OOP is to start calling turn bets more often. If OOP does that, the larger the pot OOP loses, more his tournament EV decreases (assuming IP doesn't value cut himself OTT). How do you see the impact of chip utility on post flop equilibriums ?
Thank you.
Why would there be chip utility situations in a HU event?
Was more of a general tournament question that arise from that spot that is quite frequently on MTTs.
ICM has no implication in HU (and winner takes) all tournaments. Obviously, since you have to win the match to step up in the prize, so it's the same as a winner takes all "mini" tournament.
Hey Ben, great series very interesting spots and excellent analysis.
Would like to ask you in what spots you like to develop a donking strat. You did no donkbets at all against isil and im assuming it is because he has a fairly high cb frequency, and donking is usally good against low cbet opponents.
However I've seen some strategies where develop donkbetting into our range even against a high frequency cbettor seem to be successful. What do you think about this?
I'm pretty sure donkbetting is part of optimal strategies but it's low on my to-do list of things to implement. The work I've done suggests it's not contributing much to EV and adds a lot of complexity.
At around 15:00 don't you think it's better for Isildur to be hero calling with Jx (and the like) than 7x? Jx blocks more of your value check raising range. Great video :)
Good point. It might be a mix between these hands depending on how well the second card blocks strong hands (for example KJ might be a call). If villain somehow calls too many Jx over 7x then I have an exploit to XR 7x myself getting value and pushing villain off the chop which is obviously a disaster for villain's EV.
36:10, K7o > J8s, what do you think of your opponents give up on the river? Is this not the best river for him to bet again as a bluff?
I think he'll prefer to bet hands with better blockers to my river calling region. Always bluffing this river (or bluffing extremely often) is a bad strategy.
great video.
With AQ at 9min, having any club can also block bluffs cant it? Like is having the Kc/Qc even that great bc he might barrel a hand like this (Qc8x Kc7x) and when he bets this size on river, it seems like he almost always will want to have a club in his hand. And since there are a lot more combos of Qc7x-Qc9x and Kc7x-Kc9x than there are Kxcc Qxcc flushes, than arent we maybe blocking more bluffs than value bets? I think having the 7c might be just way better than it actually seems bc ppl dont usually barrel/double barrel 7cX but there are still tons of 7xcc value hands.
Fbb,
I think it's going overboard to assume he'll always have a club in his range to bluff here, solver results typically show a more blended bluffing strategy for wide range bluff spots like this. Generally there's a lot of blocker blending on both sides so as not to bleed info to a clairvoyant opponent.
How does PT4 calculate the stat PFR? if you look at the numbers for both you and isildur (at time 4:57) the numbers are 61 and 55 respectively.. but you guys have had a strictly open raise strat (no limping) so far when on the button, and you've certainly been playing more than 61% and 55% of hands from the button... so what's up with that..?
Be the first to add a comment
You must upgrade your account to leave a comment.