very good video, a point to highlight is that on average 70% to 75% of the rake is paid in pots without showdown, if you take a rake environment like pokerstars NL200 for example and put 6 GTO bots to play, they will lose 3.5bb/100 on the red line and 1.5bb/100 on the blue line just for the rake, from the moment you put a recreational player, then it will depend on the type of mistake that player makes to know which one line the bots will take winrate, there are redline up (very calling station/spewy) and red line down recreatives, and the most interesting thing is that if you put a solver that can adapt to the recreative's strategy, instead of one that is playing a balanced strategy, this solver will explore even more the characteristics of this recreational, causing it to lose more in one line or another, in short, the solver is not concerned with red line/blue line, but as you said with the $$$...
Very interesting video, thanks! I often talk about the red line on my channel as well. But there’s definitely a lot more to it than those three tips. I, for example, still have a slightly negative red line, even though I have been following these three tips for a long time. It would be interesting to go a bit deeper into the topic.
I just watched this video and then immediately played a session. Over a dozen times I found myself about to make a passive skewing action before remembering this video. I found multiple river bluff jams that I wouldn't normally find that got through, I found a river raise that got paid off by worse. Didn't work every time, ran a diff line than I normally would w/ and lost a much bigger size pot than normal running into top of range, but I think it was far superior line. I don't know if any video I've watched has made as immediate and big impact as this one. Fantastic video.
I used to do strategy vids on my YT channel and advise players on being better at Dark Souls 3 PVP, and one of my biggest observations was that players were generally too passive. It seems stepping up the aggression is a good general strategy in a lot of games.
Hey Patrick, Thank you very much for the great content! In my opinion, the best there is nowadays. Can you make a video/comment about anticipating exploits on turn? For example, If villain has a bit of overcall turn, most combos of bluff become -EV. However, If the same player has ton of overfold River in most runnouts, it becomes +EV to bet/bet Turn and River? The same with bluffcathing. If villain underbluff a bit on Turn, but a ton on River, call/call becomes +EV always with the previous 0 ev combos bluffcatchers? If so, it's something to do intuitively, or can be done in Pio, even if it involves deviations on two streets with multiple possible runouts
Both OOP and IP, if you beat all of the valuebets that are represented in that sizing excluding traps, it will be theoretically indifferent to raise, if you start beating or chopping with any trap, it's a pure raise
What about at lower stakes like 25 & 50nl where you see a larger percentage of 3way flops? I’d imagine this would skew your redline down, thus more normal to see a declining or flat redline?
Multiway pots are pretty rare even at lower stakes so that shouldn't have a big impact on your overall graph. I don't think there are too many environments online where you shouldn't have at least a breakeven redline, given the way people play currently
Great video! Do you have any opinion on how redline is distributed across the player pool? I would think that only the most aggressive regs and fish will have winning redlines, and then factoring in that the sum of all redlines in a 6max game will be negative (as some commenters have pointed out), would this imply that the average reg with a modest winrate should expect to have a negative redline? Asking as a break even redliner 😅
You would mine to share the hands of the DDBB? Im not saying i do not believe you ofc, but i think would be great if every time a coach share a graph, we can in fact get access to the DDBB, kind of scientific approach, in the other way is like news without the source.
tbh i think that "just be more aggressive" is bad advice to improve redline winnings. if you bluff in bad spots (e.g. villains range is strong or he just doesnt give a f***) and fold at some point in the hand, your redline suffers extremely hard. yes, you need to be more aggressive, but in the right spots and with a lot of certainty about villains play or based on mass data analysis in order to leverage population exploits.
Thanks for the vid. What do you think of the River Call Efficiency in PT4 for evaluating if you are defending rivers enough? It seems pretty appealing as making correct explo river folds will leave the stat in a normal range, however I guess it can also be high/low if your get to river range is out of whack or you are not finding river raises. Do you think there is an ideal range for the stat?
It’s hard to say. I rarely look at that stat because as you suggested so many things go into it and there’s also a lot of variance. I think the only thing you can say for sure is you don’t want it to be
@@mobiuspoker it’s actually a pretty low variance stat over any reasonable sample size, because it’s completely disconnected from any other part of your game (i.e. it can’t really be affected by other stats or be higher or lower because of a player’s “style”). I think anything under about 1.3 is suboptimal. 1.0 would be very bad because it would mean you would’ve had the same result if you had folded literally every river you’ve called, and presumably you’ve occasionally called rivers with the nuts against a shove. +EV calls exist at equilibrium but -EV calls don’t, so RCE should always be >1 by a good margin if you’re playing well.
@@matta5749 Not sure about this. All river bluff catching stats usually take a while to converge. Maybe it depends on what our definition of what a "big" sample is, but I think you want at least 50k hands to start taking the stat seriously at small levels of statistical uncertainty, and double that amount if you want to compare it to someone else's stat or your own stat in a previous sample. It's also pretty easy to gain a false sense of security from a high river call efficiency. If you are making lots of very marginally +EV calls with your bluff catchers, you will drive the stat down even though your calls were correct. Conversely, if you miss thin value raises, you will drive the stat up.
im guessing thats a bovada graph ...do u think its Possible do have a positive red line in a tough pool like stars it seems like i cant get it to positive no mater what i do
So interesting video! thank you for sharing this! Let me ask a question related to tip #3. What do you think is a good size for thin value bet on the river? Some players bet small (like 30-50%) to get call from opponent's marginal hands. Is this idea correct? Or should we bet bigger (like 75% as you mentioned in the video) when we are in position?
Thanks! It depends on the scenario, but as long as the hand isn't took weak to go for a 2/3 pot bet size on the river, then in theory 2/3 pot, 1/2 pot, and 1/3 pot are usually very similar or identical in EV for value bet combos.
Any time you reach a point where 3 players have money in (including the 2 blinds) and someone folds before showdown, there will leakage of the dead money that is lost as red and won as blue. So btn opens, sb folds, bb calls and someone wins at showdown results in sb losing 1/2 bb to their redline, but the winner adds that amount to their blue line. Since the green line is zero sum in a rake free game and red+blue=green and in each hand the total blue across players can't be negative, but total red can, it means red will be negative. In HU, though, you can't win dead money at showdown, so there red and blue will each net zero. This is true for any table where all players have the same strategy, not just GTO. I guess in theory, a strategy that never folds in multiway pots would have a zeroed red line playing itself. GTO clearly doesn't play that way as SB folds to open raises quite often.
In midstakes pools like 200-500nl reg stakes or zone. Do you think roughly simplifying preflop solves to basically match the frequencies of 3bets/4bets is aggressive enough or does your data suggest it should be pushed even farther?
I think the most important consideration is your post flop edge. The higher your win rate, the more you can justify expanding pre flop. In zoom I would try to at least expand vs. fish and the nittiest regs.
Good points. Though, your point about checking when your value wins around 45-50% of the time and betting when it wins more is not correct, you need a lot higher equity to valuebet IP. The math is such that for valuebetting being higher EV than checking back, you need to beat 50% of the calling range (and a bit more to compensate for the EV loss from getting raised). Because your opponent will fold a portion of his range that you beat, it's more common to need something closer to 70, 80% equity to valuebet. If your opponent is a station or he doesn't bluff raise as much, you should valuebet slightly thinner, but never anywhere close to around 50% initial equity
I think you misunderstood the stat I was talking about. You should win the hand about 45% of the time *after you check back*, not after you bet. If you win significantly more than that after checking, you're likely checking back too strong, in other words, missing value.
You said a solver vs a solver would have a losing redline… at equilibrium wouldn’t both lines just be 0? Can you explain why you think a solver would have a losing redline
I believe what would happen in a rakeless game is all the GTO players would win x bb/100 in showdown and win -x bb/100 in non-showdown, and the net result would be all players breaking even in the long run. Losing at non showdown is sort of just how poker is supposed to work in theory.
If 3 or more people put money in, and remember every hand starts with money from the two blinds, and anyone folds before showdown, then there is red loss for the folders that "converts" to blue for the winner. In heads up it is zero sum every hand.
i think its a copywriting mistake to explain what the lines are and then in tip nr 1 you state that everyone knows that aggression on river improves red line
very good video, a point to highlight is that on average 70% to 75% of the rake is paid in pots without showdown, if you take a rake environment like pokerstars NL200 for example and put 6 GTO bots to play, they will lose 3.5bb/100 on the red line and 1.5bb/100 on the blue line just for the rake, from the moment you put a recreational player, then it will depend on the type of mistake that player makes to know which one line the bots will take winrate, there are redline up (very calling station/spewy) and red line down recreatives, and the most interesting thing is that if you put a solver that can adapt to the recreative's strategy, instead of one that is playing a balanced strategy, this solver will explore even more the characteristics of this recreational, causing it to lose more in one line or another, in short, the solver is not concerned with red line/blue line, but as you said with the $$$...
The solver is totally irrelevant here. Ur playing as a human not as a bot.
Very interesting video, thanks! I often talk about the red line on my channel as well. But there’s definitely a lot more to it than those three tips. I, for example, still have a slightly negative red line, even though I have been following these three tips for a long time. It would be interesting to go a bit deeper into the topic.
I just watched this video and then immediately played a session. Over a dozen times I found myself about to make a passive skewing action before remembering this video. I found multiple river bluff jams that I wouldn't normally find that got through, I found a river raise that got paid off by worse. Didn't work every time, ran a diff line than I normally would w/ and lost a much bigger size pot than normal running into top of range, but I think it was far superior line. I don't know if any video I've watched has made as immediate and big impact as this one. Fantastic video.
Wow, that's awesome. Thanks for the comment!
I used to do strategy vids on my YT channel and advise players on being better at Dark Souls 3 PVP, and one of my biggest observations was that players were generally too passive. It seems stepping up the aggression is a good general strategy in a lot of games.
Hey Patrick, Thank you very much for the great content! In my opinion, the best there is nowadays.
Can you make a video/comment about anticipating exploits on turn?
For example, If villain has a bit of overcall turn, most combos of bluff become -EV. However, If the same player has ton of overfold River in most runnouts, it becomes +EV to bet/bet Turn and River?
The same with bluffcathing. If villain underbluff a bit on Turn, but a ton on River, call/call becomes +EV always with the previous 0 ev combos bluffcatchers?
If so, it's something to do intuitively, or can be done in Pio, even if it involves deviations on two streets with multiple possible runouts
Best poker channel ❤
🤩
Great contents as always. Thank you.
P.S. I think someone stole your couch.
Great video. Really gets me thinking. Thanks so much!
Glad you found it helpful! Thanks for watching :)
Love it! 💙What's the rule of thumb for raising river IP (both as bluff and value?)
Could make several whole videos on this topic :)
@@mobiuspoker looking forward to it ;)
Both OOP and IP, if you beat all of the valuebets that are represented in that sizing excluding traps, it will be theoretically indifferent to raise, if you start beating or chopping with any trap, it's a pure raise
What about at lower stakes like 25 & 50nl where you see a larger percentage of 3way flops? I’d imagine this would skew your redline down, thus more normal to see a declining or flat redline?
Multiway pots are pretty rare even at lower stakes so that shouldn't have a big impact on your overall graph. I don't think there are too many environments online where you shouldn't have at least a breakeven redline, given the way people play currently
Great video! Do you have any opinion on how redline is distributed across the player pool? I would think that only the most aggressive regs and fish will have winning redlines, and then factoring in that the sum of all redlines in a 6max game will be negative (as some commenters have pointed out), would this imply that the average reg with a modest winrate should expect to have a negative redline? Asking as a break even redliner 😅
Yes, I think the average player definitely has a losing redline
You would mine to share the hands of the DDBB? Im not saying i do not believe you ofc, but i think would be great if every time a coach share a graph, we can in fact get access to the DDBB, kind of scientific approach, in the other way is like news without the source.
tbh i think that "just be more aggressive" is bad advice to improve redline winnings. if you bluff in bad spots (e.g. villains range is strong or he just doesnt give a f***) and fold at some point in the hand, your redline suffers extremely hard. yes, you need to be more aggressive, but in the right spots and with a lot of certainty about villains play or based on mass data analysis in order to leverage population exploits.
Thanks for the vid.
What do you think of the River Call Efficiency in PT4 for evaluating if you are defending rivers enough? It seems pretty appealing as making correct explo river folds will leave the stat in a normal range, however I guess it can also be high/low if your get to river range is out of whack or you are not finding river raises. Do you think there is an ideal range for the stat?
It’s hard to say. I rarely look at that stat because as you suggested so many things go into it and there’s also a lot of variance. I think the only thing you can say for sure is you don’t want it to be
@@mobiuspoker it’s actually a pretty low variance stat over any reasonable sample size, because it’s completely disconnected from any other part of your game (i.e. it can’t really be affected by other stats or be higher or lower because of a player’s “style”).
I think anything under about 1.3 is suboptimal. 1.0 would be very bad because it would mean you would’ve had the same result if you had folded literally every river you’ve called, and presumably you’ve occasionally called rivers with the nuts against a shove. +EV calls exist at equilibrium but -EV calls don’t, so RCE should always be >1 by a good margin if you’re playing well.
@@matta5749 Not sure about this. All river bluff catching stats usually take a while to converge. Maybe it depends on what our definition of what a "big" sample is, but I think you want at least 50k hands to start taking the stat seriously at small levels of statistical uncertainty, and double that amount if you want to compare it to someone else's stat or your own stat in a previous sample.
It's also pretty easy to gain a false sense of security from a high river call efficiency. If you are making lots of very marginally +EV calls with your bluff catchers, you will drive the stat down even though your calls were correct. Conversely, if you miss thin value raises, you will drive the stat up.
@@mobiuspoker Yes it's correct that a high RCE isn't necessarily good (and is probably always suboptimal), but a low RCE of 1 and
im guessing thats a bovada graph ...do u think its Possible do have a positive red line in a tough pool like stars it seems like i cant get it to positive no mater what i do
Not Bovada, his main sites were Stars, WPN, and Chico
So interesting video! thank you for sharing this!
Let me ask a question related to tip #3. What do you think is a good size for thin value bet on the river? Some players bet small (like 30-50%) to get call from opponent's marginal hands. Is this idea correct? Or should we bet bigger (like 75% as you mentioned in the video) when we are in position?
Thanks! It depends on the scenario, but as long as the hand isn't took weak to go for a 2/3 pot bet size on the river, then in theory 2/3 pot, 1/2 pot, and 1/3 pot are usually very similar or identical in EV for value bet combos.
@@mobiuspoker Thank you! That is a great insight for me!
BTW I love this channel because I can ask you a question directly like this :)
@@jeramae123 Correct, it's much more common to use the blocker size when out of position.
Interesting solver v solver has losing redline. How were you able to figure this out?
Experiments with solvers vs. solvers seem to show this. It's expected, actually, I'd be surprised if it turned out any other way.
Any time you reach a point where 3 players have money in (including the 2 blinds) and someone folds before showdown, there will leakage of the dead money that is lost as red and won as blue. So btn opens, sb folds, bb calls and someone wins at showdown results in sb losing 1/2 bb to their redline, but the winner adds that amount to their blue line. Since the green line is zero sum in a rake free game and red+blue=green and in each hand the total blue across players can't be negative, but total red can, it means red will be negative. In HU, though, you can't win dead money at showdown, so there red and blue will each net zero. This is true for any table where all players have the same strategy, not just GTO. I guess in theory, a strategy that never folds in multiway pots would have a zeroed red line playing itself. GTO clearly doesn't play that way as SB folds to open raises quite often.
In midstakes pools like 200-500nl reg stakes or zone. Do you think roughly simplifying preflop solves to basically match the frequencies of 3bets/4bets is aggressive enough or does your data suggest it should be pushed even farther?
I think the most important consideration is your post flop edge. The higher your win rate, the more you can justify expanding pre flop. In zoom I would try to at least expand vs. fish and the nittiest regs.
Awesome content my man. I took a few good things away from this one.
If it's true that people tend to be too nitty, why would we fold less on the river?
Good points. Though, your point about checking when your value wins around 45-50% of the time and betting when it wins more is not correct, you need a lot higher equity to valuebet IP.
The math is such that for valuebetting being higher EV than checking back, you need to beat 50% of the calling range (and a bit more to compensate for the EV loss from getting raised). Because your opponent will fold a portion of his range that you beat, it's more common to need something closer to 70, 80% equity to valuebet.
If your opponent is a station or he doesn't bluff raise as much, you should valuebet slightly thinner, but never anywhere close to around 50% initial equity
I think you misunderstood the stat I was talking about.
You should win the hand about 45% of the time *after you check back*, not after you bet. If you win significantly more than that after checking, you're likely checking back too strong, in other words, missing value.
@@mobiuspoker I see what you mean, you meant overall with every hand. I thought you meant with a specific value hand
You said a solver vs a solver would have a losing redline… at equilibrium wouldn’t both lines just be 0? Can you explain why you think a solver would have a losing redline
I believe what would happen in a rakeless game is all the GTO players would win x bb/100 in showdown and win -x bb/100 in non-showdown, and the net result would be all players breaking even in the long run. Losing at non showdown is sort of just how poker is supposed to work in theory.
If 3 or more people put money in, and remember every hand starts with money from the two blinds, and anyone folds before showdown, then there is red loss for the folders that "converts" to blue for the winner. In heads up it is zero sum every hand.
Banko?
smash that mf like button
Can't relate mines in the sky
😁
Lmao I may be one of the few idiots who needs help with his blue line instead of his red line
i think its a copywriting mistake to explain what the lines are and then in tip nr 1 you state that everyone knows that aggression on river improves red line
How 1:01 is that a losing redline? its its breakeven over like 400k hands 🤣 Just a cringy 'hidden brag' graph.
Patrick Howard does not play online poker. Ask him for his screen names.