1) Between 20-21 min mark I see PIO shows some green boxes and red boxes under certain hands. What do they mean? I tried to pause the video and figure it out, but not sure if it's just representing a huge EV loss for red boxes and EV gain for green boxes? Looks like a massive EV loss for block bet on some of these hands compared to other sizes.
2) Also I noticed at 21:14 time stamp it doesn't look like any AX is donk betting. Does donk bet then folding to a raise with some of the AX hands that would check-fold makes sense? Too weak to bluff catch a hand like A6, but you don't wanna get bluffed either, so a hand like A6 (Top pair, but blocks some bluffs), maybe use a block bet for oop? That way a hand like AJ only calls on the river because you'll have Kx in your range as well. Also mix in some donk bets with 65o where you get some QJ QT J10 hands to fold.
3) 36min - 99 on 752ss hero XR 500 to 1400 and get a turn 752ss3hh. Hero chooses to check. In some recent six max videos, they talk about players floating with 2 overs and bdfd + bdsd and when another low card comes on the turn, they (six max 500z players) have been using a lot of 25-33% turn block bets. Do you think this turn block bet would be a good bet size for 99 on this hand?
4) I am not sure if I missed it or not, but on Hand #1 I didn't see what Kx5d was supposed to do on the river when you used half pot and he went for 6x XR. Also last hand #2, was the A2c just a pure bet (50%) on the turn for IP? Looks like half of the A2s are betting, but not sure what suites. The turn strategy seemed SUPER STRANGE to me with middling pairs betting and higher pairs checking, but bottom pair betting and top pair checking. Everything seems backwards. Would love to see another one of these videos.
This view is pretty confusing I'll admit - I selected 'Strategy + EV' which is simply overlaying two things at once, the Strategy (top half of the box, green, meaning J3s is checking pure) and the EV (bottom half of the box, red, meaning J3s is negative EV). It's not actually saying anything about which options have positive or negative EV, the numbers show that independently (all of which are negative, except checking, which should make sense out of why it checks pure).
You can see this is happening a little bit with A9/AT, but I think it only makes sense with the best kickers so that we actually have a shot at getting called by worse. A7/A4 were counterfeit on this river, so leading a hand that wins vs those combos makes sense. That said, I think we're seeing mostly Kx leading because we need to protect our checking range as well, and Ax is better suited for that. We're already allowed to bluff lead 65 and a good percent of our J-Q high unpaired hands, so it's not too important to lead any more often than this.
Yep, I think if I wasn't so concerned with keeping the pot small, I'd have normally block bet or possibly even bet large for value. I was trying to avoid stacking off at this moment, though, as I mentioned in the comment below.
4a. This might've been confusing since the suits were changed, but the hand class he had (Kx5s in my sim) was check/raising river almost half the time.
4b. The A2s without FD are betting more than half the time, while A2s with FD are pure checking behind. I'm supposed to check/jam pretty often on the turn, and these hands are bet/folding mostly, so I guess with the FD they don't want to face that play.
Looks like it's a sliding scale with red being below 0 and green being the highest, if you click the EV tab it's easiest to visualize. Here's the same node with only the EV tab selected.
Sorry one last question. The last hand with 99 you chose to flat preflop instead of 3 betting because you mentioned it was towards end of the session and you were up a couple of buy-ins. This seems like a bad mindset, but is this legend thing more of a tourney? I would think you would want to optimize your EV every hand whether it's your first hand or last hand of the session. You briefly mentioned ICM, but appears to be a cash game, so not sure how ICM plays a part.
Yes, this was an ICM effect I'm referencing. Each "win" in the early stages of the Legends Showdown was worth some amount of money, and I think in this case I was securing my place in the semi-finals by winning this match, so it was actually worth quite a lot to finish in the green.
The keyboard shortcut for this is shift+ctrl+O, as others pointed out this navigates any sims you have saved in the same folder as the sim you're currently looking at
good video as always! The way I understand the "Create Subtree" function to work, is that it basically uses the ranges that arrived at the selected node from the original tree setup, to then establish a new equilibrium with the new parameters. Wouldn't it be necessary to re-run the whole sim with the new parameters and compare it to the "subtree sim" in order to be certain, that the original bet sizing options didn't influence the ranges in a too drastic way?
An example: if the original sim only had a block probe size for OOP on the turn (after flop went check check) and now you create a subtree with an overbet size for OOP, it seems quite obvious that this would be a high EV option for OOP on most textures, since IP's turn range is not solved for that option and might lack enough hands that can defend properly against the overbet. This effect gets stronger the more limited betsizes on later streets become. If OOP can never bet big on turn and river but is allowed to XR flop, then IP is incentivized to bet less on the flop.
I am not sure if I have expressed my thought clearly. But I was wondering if you have looked closer at this and know, whether my thinking here is wrong or if the effect isn't really worth considering.
Yes, this is definitely a concern and a great question to ask! I am careful to offer a little bit of everything (especially block bets and overbets) in the original sim when using subtrees later on. Indeed there could be a major difference in flop strategy, for example, if your turn parameters are too limiting. As long as your original sim does not ignore any high frequency or high EV lines on later streets, your early street strategy should come fairly close to accurate.
Loading 14 Comments...
1) Between 20-21 min mark I see PIO shows some green boxes and red boxes under certain hands. What do they mean? I tried to pause the video and figure it out, but not sure if it's just representing a huge EV loss for red boxes and EV gain for green boxes? Looks like a massive EV loss for block bet on some of these hands compared to other sizes.
2) Also I noticed at 21:14 time stamp it doesn't look like any AX is donk betting. Does donk bet then folding to a raise with some of the AX hands that would check-fold makes sense? Too weak to bluff catch a hand like A6, but you don't wanna get bluffed either, so a hand like A6 (Top pair, but blocks some bluffs), maybe use a block bet for oop? That way a hand like AJ only calls on the river because you'll have Kx in your range as well. Also mix in some donk bets with 65o where you get some QJ QT J10 hands to fold.
3) 36min - 99 on 752ss hero XR 500 to 1400 and get a turn 752ss3hh. Hero chooses to check. In some recent six max videos, they talk about players floating with 2 overs and bdfd + bdsd and when another low card comes on the turn, they (six max 500z players) have been using a lot of 25-33% turn block bets. Do you think this turn block bet would be a good bet size for 99 on this hand?
4) I am not sure if I missed it or not, but on Hand #1 I didn't see what Kx5d was supposed to do on the river when you used half pot and he went for 6x XR. Also last hand #2, was the A2c just a pure bet (50%) on the turn for IP? Looks like half of the A2s are betting, but not sure what suites. The turn strategy seemed SUPER STRANGE to me with middling pairs betting and higher pairs checking, but bottom pair betting and top pair checking. Everything seems backwards. Would love to see another one of these videos.
This view is pretty confusing I'll admit - I selected 'Strategy + EV' which is simply overlaying two things at once, the Strategy (top half of the box, green, meaning J3s is checking pure) and the EV (bottom half of the box, red, meaning J3s is negative EV). It's not actually saying anything about which options have positive or negative EV, the numbers show that independently (all of which are negative, except checking, which should make sense out of why it checks pure).
You can see this is happening a little bit with A9/AT, but I think it only makes sense with the best kickers so that we actually have a shot at getting called by worse. A7/A4 were counterfeit on this river, so leading a hand that wins vs those combos makes sense. That said, I think we're seeing mostly Kx leading because we need to protect our checking range as well, and Ax is better suited for that. We're already allowed to bluff lead 65 and a good percent of our J-Q high unpaired hands, so it's not too important to lead any more often than this.
Yep, I think if I wasn't so concerned with keeping the pot small, I'd have normally block bet or possibly even bet large for value. I was trying to avoid stacking off at this moment, though, as I mentioned in the comment below.
4a. This might've been confusing since the suits were changed, but the hand class he had (Kx5s in my sim) was check/raising river almost half the time.
4b. The A2s without FD are betting more than half the time, while A2s with FD are pure checking behind. I'm supposed to check/jam pretty often on the turn, and these hands are bet/folding mostly, so I guess with the FD they don't want to face that play.
Kevin Rabichow So Green = +EV and Red = -EV so Orange = 0EV? Yellow = 0EV?
Looks like it's a sliding scale with red being below 0 and green being the highest, if you click the EV tab it's easiest to visualize. Here's the same node with only the EV tab selected.
Sorry one last question. The last hand with 99 you chose to flat preflop instead of 3 betting because you mentioned it was towards end of the session and you were up a couple of buy-ins. This seems like a bad mindset, but is this legend thing more of a tourney? I would think you would want to optimize your EV every hand whether it's your first hand or last hand of the session. You briefly mentioned ICM, but appears to be a cash game, so not sure how ICM plays a part.
Yes, this was an ICM effect I'm referencing. Each "win" in the early stages of the Legends Showdown was worth some amount of money, and I think in this case I was securing my place in the semi-finals by winning this match, so it was actually worth quite a lot to finish in the green.
nice video as per usual, Kevin!
how did you manage to get this box up in PIO to quickly find a particular flop texture from your saves? that could save me quite a bit of time haha.
I believe he have all saves in 1 folder and use name like board texture
for example AdJcTs.cfr
File -> Load Tree in same folder
thanks guys. will try that out during my next study session!
The keyboard shortcut for this is shift+ctrl+O, as others pointed out this navigates any sims you have saved in the same folder as the sim you're currently looking at
Hey Kevin,
good video as always! The way I understand the "Create Subtree" function to work, is that it basically uses the ranges that arrived at the selected node from the original tree setup, to then establish a new equilibrium with the new parameters. Wouldn't it be necessary to re-run the whole sim with the new parameters and compare it to the "subtree sim" in order to be certain, that the original bet sizing options didn't influence the ranges in a too drastic way?
An example: if the original sim only had a block probe size for OOP on the turn (after flop went check check) and now you create a subtree with an overbet size for OOP, it seems quite obvious that this would be a high EV option for OOP on most textures, since IP's turn range is not solved for that option and might lack enough hands that can defend properly against the overbet. This effect gets stronger the more limited betsizes on later streets become. If OOP can never bet big on turn and river but is allowed to XR flop, then IP is incentivized to bet less on the flop.
I am not sure if I have expressed my thought clearly. But I was wondering if you have looked closer at this and know, whether my thinking here is wrong or if the effect isn't really worth considering.
Thank you!
Yes, this is definitely a concern and a great question to ask! I am careful to offer a little bit of everything (especially block bets and overbets) in the original sim when using subtrees later on. Indeed there could be a major difference in flop strategy, for example, if your turn parameters are too limiting. As long as your original sim does not ignore any high frequency or high EV lines on later streets, your early street strategy should come fairly close to accurate.
well done vid mister Krab!
Be the first to add a comment
You must upgrade your account to leave a comment.