Hi Tyler
7:55 you talked about the protection bet preferring to be quarter pot. It seems the concept of betting for protection has changed a lot over the years and to me protection betting is simply value betting. Our objective is to have our opponent make a mistake and call with worse hands that don't have the eq to call and if they fold well that's ok too as they lose their pot share. It is basically just not letting them realize their eq for free. Is this an accurate view of this concept?
If we have an opponent that will call too wide then wouldn't our hand want to bet a bit bigger?
12:00 I appreciated the talk on taking a line that had lower deviation and how that applied to making it as a poker player. Maybe that would be a an idea for a future video. I would love to hear your thoughts on making it as a pro with proper bankroll management and life roll management.
With a value hand, we want to bet the sizing that gets the most money from the worst hands. Usually against bad players that means bigger especially earlier in hands. Protection betting/value betting in GTO is substantially more complicated.
Protection betting is a bet whose primary purpose is to deny equity realization to our opponents. The equity share of the losing hand in no limit is quite small so it takes a small amount of money to "protect" the pot.
The solver seems to like sizing schemes that size around protection ideas, because it lets us bet the most hands on earlier. If we bet too big, then we bet much less frequently.
If we split sizings, then we can have a protection region and bigger sizing which is more polarized to maximize value extraction.
@Bankroll Management I really like that idea. I will see what I can put together.
3min for the turn probe looking at some old data for micro/low stakes six max on Iggy for any 2 bet pot these are the numbers I have for the pool. MDA would just get to probe entire range once flop goes XX or IP plays B-X and hero probes the river. Do you have any H2N data on this breaking down the bet size, HU, MW, etc?
I tend to the think MDA is overrated on ignition especially against the pros, because some many people try the obvious exploits that the professionals tend to adjust quickly. It's possible recreational players though have these leaks.
4min 762r hero OB the flop. Villain has a 0% FCB. Hindsight if someone has a low FCB should we just never cbet bluff our air? Just stick to equity driven range and size up with our value?
30min on Q52dd-7d-Qx any thoughts on playing XC-XC-D
1) IP barrels more often on the turn than in the past since the turn is over folded.
2) If the turn is over folded, then the river is not barreled often enough, so we should donk bets some rivers to get value from AA/KK/Qx or sticky JdJx-8d8x region.
3) I agree with you that KQ should probably check back as its hard to get called by worse on this river, but how likely is he to fold Qx to a donk bet?
I don't put much weight on the fold to cbet number after 49 hands. It's basically worthless.
It's possible donk is the best play here. The tricky spot is what to do against raise. Bet/Folding vs a bluff is very expensive as is losing $1000 against the nuts rather than $350. I tend to shy away from spots where I expose myself to big mistakes and I think donk betting could be one of those situations.
Turn we’re fist pumping kind of love X calling or X raising turn
River we’re like Oh no! Haha but sad when we call he has AQ KQ QJ Q10, is raising this river a big mistake?
If he has a boat, can we assume he would be betting bigger on this river?
Or do we need to have some flats with our flushes
It's more that Qx is a marginal value bet so I expect it to bet/fold to the jam. If he calls flushes+ here, then he actually has more Ace-high flushes than lower flushes due to my card-removal, so not only do I lose to fullhouses, but I also lose to the flush range.
For this hand, I think you can bluff turn pretty often despite having 9d. I would have missed this too given that the diamond is a negative blocker, but the solver is betting 60% frequency overall so hands like this do make it in sometimes.
The solver is mixing equivalent trashy hands like this, including no diamond hands, 1 diamond hands, and even 2 diamond hands. So as a human you might want to just simplify and bet all the no diamond hands and check the rest, but then you are still missing some bluffs with this hand class as you would still have an extra bluff allowance beyond non-diamond hands on the turn.
Another MDA based consideration is that people overfold the turn after calling a block-sized cbet. I know you advocated against MDA exploits in the video, but if you bump your frequency slightly with this type of hand here I don't think it would be noticeable deviation that your opponent could instantly recognise and counter-exploit you for.
For the river of this hand, the solver is also bluffing it at a fairly high frequency. The solver also suggests OOP will bet some middling AX on the river, which I don't think regs will do that often, so I node-locked that out, and even then the solver will mix a decent number of bluffs with this hand.
As a human I think the turn bet makes more sense than the river bet. I tend to think players are going to bluff some K-highs that solver tends to showdown in these spots, so the range is a little narrower than the solver anticipates.
I tend to think players are going to bluff some K-highs that solver tends to showdown in these spots
Yeh I think thats a fair assumption. I node-locked a few K high bluffs in, Q9 still bluffing river sometimes. Important to note that the solver has some KX in the XC range for BB here, so BB bluffing K high hands occasionally removes calling hands too!
With regards to the MDA debate in the video where you said that a turn DCB exploit could net someone 1BB a few times then cost them 21BB as soon as their opponent realises the attempted exploit - do you think this is true in all games, or specifically those where hole cards become available to all players after the hand?
For reference, when do you find out hole cards in these games? Is it immediately after each hand, or only at a later time?
There's a 24-hour delay but the pools are small enough that unique sizing schemes are readily identifiable, so lots of "good" exploits can go -EV very quickly, just from identifying the pattern and player who uses them.
Loading 19 Comments...
Hi Tyler
7:55 you talked about the protection bet preferring to be quarter pot. It seems the concept of betting for protection has changed a lot over the years and to me protection betting is simply value betting. Our objective is to have our opponent make a mistake and call with worse hands that don't have the eq to call and if they fold well that's ok too as they lose their pot share. It is basically just not letting them realize their eq for free. Is this an accurate view of this concept?
If we have an opponent that will call too wide then wouldn't our hand want to bet a bit bigger?
12:00 I appreciated the talk on taking a line that had lower deviation and how that applied to making it as a poker player. Maybe that would be a an idea for a future video. I would love to hear your thoughts on making it as a pro with proper bankroll management and life roll management.
Thanks!
Soundspeed really great questions!
With a value hand, we want to bet the sizing that gets the most money from the worst hands. Usually against bad players that means bigger especially earlier in hands. Protection betting/value betting in GTO is substantially more complicated.
Protection betting is a bet whose primary purpose is to deny equity realization to our opponents. The equity share of the losing hand in no limit is quite small so it takes a small amount of money to "protect" the pot.
The solver seems to like sizing schemes that size around protection ideas, because it lets us bet the most hands on earlier. If we bet too big, then we bet much less frequently.
If we split sizings, then we can have a protection region and bigger sizing which is more polarized to maximize value extraction.
@Bankroll Management I really like that idea. I will see what I can put together.
3min for the turn probe looking at some old data for micro/low stakes six max on Iggy for any 2 bet pot these are the numbers I have for the pool. MDA would just get to probe entire range once flop goes XX or IP plays B-X and hero probes the river. Do you have any H2N data on this breaking down the bet size, HU, MW, etc?
I tend to the think MDA is overrated on ignition especially against the pros, because some many people try the obvious exploits that the professionals tend to adjust quickly. It's possible recreational players though have these leaks.
4min 762r hero OB the flop. Villain has a 0% FCB. Hindsight if someone has a low FCB should we just never cbet bluff our air? Just stick to equity driven range and size up with our value?
30min on Q52dd-7d-Qx any thoughts on playing XC-XC-D
1) IP barrels more often on the turn than in the past since the turn is over folded.
2) If the turn is over folded, then the river is not barreled often enough, so we should donk bets some rivers to get value from AA/KK/Qx or sticky JdJx-8d8x region.
3) I agree with you that KQ should probably check back as its hard to get called by worse on this river, but how likely is he to fold Qx to a donk bet?
I don't put much weight on the fold to cbet number after 49 hands. It's basically worthless.
It's possible donk is the best play here. The tricky spot is what to do against raise. Bet/Folding vs a bluff is very expensive as is losing $1000 against the nuts rather than $350. I tend to shy away from spots where I expose myself to big mistakes and I think donk betting could be one of those situations.
Tyler,
Always enjoy your commentary on the hands in the videos you make. Nice job.
Thanks Tyler.
Thanks 777!
Hey Tyler another great video man thank you
Hey Truepower, I appreciate the love!
Really fascinating runout here
Turn we’re fist pumping kind of love X calling or X raising turn
River we’re like Oh no! Haha but sad when we call he has AQ KQ QJ Q10, is raising this river a big mistake?
If he has a boat, can we assume he would be betting bigger on this river?
Or do we need to have some flats with our flushes
It's more that Qx is a marginal value bet so I expect it to bet/fold to the jam. If he calls flushes+ here, then he actually has more Ace-high flushes than lower flushes due to my card-removal, so not only do I lose to fullhouses, but I also lose to the flush range.
For this hand, I think you can bluff turn pretty often despite having 9d. I would have missed this too given that the diamond is a negative blocker, but the solver is betting 60% frequency overall so hands like this do make it in sometimes.
The solver is mixing equivalent trashy hands like this, including no diamond hands, 1 diamond hands, and even 2 diamond hands. So as a human you might want to just simplify and bet all the no diamond hands and check the rest, but then you are still missing some bluffs with this hand class as you would still have an extra bluff allowance beyond non-diamond hands on the turn.
Another MDA based consideration is that people overfold the turn after calling a block-sized cbet. I know you advocated against MDA exploits in the video, but if you bump your frequency slightly with this type of hand here I don't think it would be noticeable deviation that your opponent could instantly recognise and counter-exploit you for.
For the river of this hand, the solver is also bluffing it at a fairly high frequency. The solver also suggests OOP will bet some middling AX on the river, which I don't think regs will do that often, so I node-locked that out, and even then the solver will mix a decent number of bluffs with this hand.
matlittle
Another insightful post, thank you!
As a human I think the turn bet makes more sense than the river bet. I tend to think players are going to bluff some K-highs that solver tends to showdown in these spots, so the range is a little narrower than the solver anticipates.
Yeh I think thats a fair assumption. I node-locked a few K high bluffs in, Q9 still bluffing river sometimes. Important to note that the solver has some KX in the XC range for BB here, so BB bluffing K high hands occasionally removes calling hands too!
matlittle
Never discount the GTO K-high calldowns.
With regards to the MDA debate in the video where you said that a turn DCB exploit could net someone 1BB a few times then cost them 21BB as soon as their opponent realises the attempted exploit - do you think this is true in all games, or specifically those where hole cards become available to all players after the hand?
For reference, when do you find out hole cards in these games? Is it immediately after each hand, or only at a later time?
There's a 24-hour delay but the pools are small enough that unique sizing schemes are readily identifiable, so lots of "good" exploits can go -EV very quickly, just from identifying the pattern and player who uses them.
Be the first to add a comment
You must upgrade your account to leave a comment.