nittyoldman10 years, 6 months agoyes more HU PLO action plz plz plz MR. Galfond!! ok now that I watched the video, I see you are heads up in the tourney for the last part of the series...anyways, I do hope you put out more HU cash PLO after this.
48:30 you call AJ35 sb vs bb and x/c AJ4r, turn T goes chk/chk and on 6o river you bet 3/4ths pot.
I've seen you do this a bunch and I never really understood why when you feel it's very difficult for you to get called and that you have the best hand a really high percentage of the time why you rarely seem to choose to size your bet down to like 1/4th-1/3rd pot in order to rep weaker value bets? To me it seems that theoretically, this would be the better play, why don't you agree?
In practice, the best bet-size here is going to be pretty dependent on your opponent.
In theory, if my opponent is truly capped and my range is mostly value, it doesn't matter a whole lot what I choose as my bet-size.
Let's pretend I have 10 hands here - 8 VBs (none of which he beats) and 2 bluffs.
If I bet 110k into 110k with all of them, his best play is to always fold. My EV is the 110k pot with my full range, and his EV is 0.
If I bet 70k into 110k, this lays him ~2.6:1 (meaning he needs ~28% equity to call). He still should always fold and my EV is identical: the 110k pot (his EV is 0)
If I size my bet down to 1/3 pot (36.666k into 110k), he is now getting exactly 4:1. He's indifferent to calling or folding, but our EVs are the same. No matter what strategy he chooses, my EV is 110k and his is 0.
In this toy game, the only way I can capture less than the full pot in EV is by betting under 1/3 pot. Anything 1/3 pot or higher nets me the same EV.
Interestingly, betting 1/3 pot is the only viable strategy of these three that doesn't allow him to make a mistake.
In practice there's a lot more to consider, including occasionally running into the nuts or even a set. That will cause me to want to size my bet down, and a toy game that includes that possibility would require me to bet smaller.
The reason I chose the bet-size I did is that the couple of bluff hands I have are KK54 and QQTx type stuff, and I don't think I'd bet 1/3 pot with those hands. Not that I believe I shouldn't bet 1/3 pot with them... I just think that put in that situation in the moment, I would bet larger. I wanted to size my VB the same way.
If I were to play this way I would just about never get paid off. The times I see you do it in videos I also almost never see you get paid off. I generally choose to bet the large amount w/ my bluffs and the smaller amount w/ my value, though I don't think many of my opponents catch on.
Does this mean that I'm just not bluffing enough in these spots? Or is it the amount that you are winning with your bluffs is greater than the amount I'm making with my smaller value bets?
Phil, Isn't this an exploitable strategy? If you underbluff the river, he should always fold. Unless the EV of him never defending the pot is greater than the EV of your value bets (which never get paid off because he always folds), you have to go with the bet sizing that represents your value/bluff ratio. Or am I wrong here?
@Peter - I have always taken balance/believability too far in that I say to myself. "would I bluff this way?" and if the answer is no I don't VB that way. I think that I should probably size down here for exploitative reasons but it's hard to bring myself to do it.
@Chael - I don't consider myself the resident expert on Game Theory, but I believe I know the answer here.
In the toy game in which he is capped and my range is 80% value, there are several strategic options which allow me to capture the whole pot with my entire range (as shown above). They all give me an equal result and therefore are equally viable strategies.
If I size down to make him indifferent to calling/folding, sure he can call some bets if he wants to. My EV and his EV are still the same though. I get the full pot and he gets nothing.
Forcing your opponent to be indifferent isn't the end goal... it's usually a means to the end goal.
Your goal is to find the strategy with the highest EV (assuming villain responds in a maximally exploitative way). In many spots, this will mean that we'd like to make him indifferent (since if he's not indifferent, he will choose the best strategy), but in spots like this it doesn't matter- we get the whole pot due to the range imbalance as long as we don't bet too small.
(If a player knows your strategy, you should never be able to capture more than 100% of the pot.)
Yes, he should fold every time, but I wouldn't call that an exploitation of our strategy.
What's interesting to me, which I realized while typing the previous response- If we are forced to choose between any of the strategies that involve one bet-size, betting 1/3 pot is the worst option in theory-practice (still in the toy game, but against an imperfect opponent). It's the only strategy against which our EV is exactly 100% of the pot no matter how he responds.
If we bet larger, our minimum EV is 100% of the pot, but if he ever calls a bet we improve that EV to >100% pot.
At Phil, in your example, it appears that you're giving yourself a perfectly polarized range, do you really think that this is the case? I'd assume that while your range could likely have close to 80% equity, I would assume that it's more condensed. In other words, you'll have lots of value hands that beat the majority of your opponents hands, but you'll also have value hands which will lose to some of your opponents stronger hands. In addition, the bottom of your range is still likely ahead a small chunk of the Villain's bottom range.
If the latter is correct, my guess is that theoretically you'll still want to bet larger, but not for the same reason that we deserve the entire pot. Instead, when a very large proportion of your range is made up of very strong hands, then you realize more EV by betting larger and being called less than by betting smaller and being called more.
However, I think it's probably best when oop with this range to have multiple bet sizes. You'll want smaller bet sizes for thinner value which are protected with your nut range. And then you'll want larger bet sizes with the strong part of your range. So maybe Peter wasn't completely off in wanting to size your bet down, but I do think it would be a mistake if that was your only bet size.
I often see in the forums and frankly in the videos the logic that DeSalle makes which is that it's "hard to have a bluff" therefore we should size our bet down. While we don't have many traditional bluffs which don't have any showdown component, this doesn't mean that we shouldn't turn made hands into bluffs. An equilibrium will form with the indifference being at the TOP of our bluffing range where we'll be indifferent between turning a made hand into a bluff and checking it. Therefore, the Villain responds by defending less than 1-A.
I could probably solve this situation on the river in a couple hours, if you wanted to lay out a more detailed version of both players ranges. It's not too hard for me to nearly solve it for multiple bet sizes if you devise ranges where blockers aren't relevant.
still very much enjoying this mamoth series.....congrats on shipping the $109+R last night...any chance you were recording that MTT session? rare to see you grinding NL MTTs exclusively.
I started playing the SCOOP earlier this year in preparation for the WSOP. I ended up doing pretty well and enjoying it too. I've played a few days of MTTs since coming back (just the bigger days with special events), and I plan on playing some WCOOP once it rolls around.
I find MTT days relaxing and fun, and I think it helps me to fit some in between stressful nosebleed games.
I wasn't recording, and likely won't record a ton of WCOOP days, but if I win a big one I'll be happy to do a HH review.
(PayPal requires that you authorize us to open a time period of one year to bill you monthly. In your Settings, you can downgrade or change your payment type at any time during your subscription with Run It Once)yeah i can change it but why i have to do anything to not pay for more than a month. after i opened paypal it had nothing but a spot to insert my creditcard information
Amazing series!! I would really enjoy it if you record a couple live hours of a wcoop sunday! In my opinion, you don't need to get HU to make the days recording worth it. I would enjoy a couple hours of live commentary during a WCOOP sunday regardless of your results.
Loading 19 Comments...
epic series. can't wait for HU
ok now that I watched the video, I see you are heads up in the tourney for the last part of the series...anyways, I do hope you put out more HU cash PLO after this.
48:30 you call AJ35 sb vs bb and x/c AJ4r, turn T goes chk/chk and on 6o river you bet 3/4ths pot.
I've seen you do this a bunch and I never really understood why when you feel it's very difficult for you to get called and that you have the best hand a really high percentage of the time why you rarely seem to choose to size your bet down to like 1/4th-1/3rd pot in order to rep weaker value bets? To me it seems that theoretically, this would be the better play, why don't you agree?
Was thinking the exact same thing. If it's incredibly hard to have a bluff then I don't like the 70-80k bet. 35-40k bet must be better.
In practice, the best bet-size here is going to be pretty dependent on your opponent.
In theory, if my opponent is truly capped and my range is mostly value, it doesn't matter a whole lot what I choose as my bet-size.
Let's pretend I have 10 hands here - 8 VBs (none of which he beats) and 2 bluffs.
If I bet 110k into 110k with all of them, his best play is to always fold. My EV is the 110k pot with my full range, and his EV is 0.
If I bet 70k into 110k, this lays him ~2.6:1 (meaning he needs ~28% equity to call). He still should always fold and my EV is identical: the 110k pot (his EV is 0)
If I size my bet down to 1/3 pot (36.666k into 110k), he is now getting exactly 4:1. He's indifferent to calling or folding, but our EVs are the same. No matter what strategy he chooses, my EV is 110k and his is 0.
In this toy game, the only way I can capture less than the full pot in EV is by betting under 1/3 pot. Anything 1/3 pot or higher nets me the same EV.
Interestingly, betting 1/3 pot is the only viable strategy of these three that doesn't allow him to make a mistake.
In practice there's a lot more to consider, including occasionally running into the nuts or even a set. That will cause me to want to size my bet down, and a toy game that includes that possibility would require me to bet smaller.
The reason I chose the bet-size I did is that the couple of bluff hands I have are KK54 and QQTx type stuff, and I don't think I'd bet 1/3 pot with those hands. Not that I believe I shouldn't bet 1/3 pot with them... I just think that put in that situation in the moment, I would bet larger. I wanted to size my VB the same way.
If I were to play this way I would just about never get paid off. The times I see you do it in videos I also almost never see you get paid off. I generally choose to bet the large amount w/ my bluffs and the smaller amount w/ my value, though I don't think many of my opponents catch on.
Does this mean that I'm just not bluffing enough in these spots? Or is it the amount that you are winning with your bluffs is greater than the amount I'm making with my smaller value bets?
Phil, Isn't this an exploitable strategy? If you underbluff the river, he should always fold.
Unless the EV of him never defending the pot is greater than the EV of your value bets (which never get paid off because he always folds), you have to go with the bet sizing that represents your value/bluff ratio.
Or am I wrong here?
@Peter - I have always taken balance/believability too far in that I say to myself. "would I bluff this way?" and if the answer is no I don't VB that way. I think that I should probably size down here for exploitative reasons but it's hard to bring myself to do it.
@Chael - I don't consider myself the resident expert on Game Theory, but I believe I know the answer here.
In the toy game in which he is capped and my range is 80% value, there are several strategic options which allow me to capture the whole pot with my entire range (as shown above). They all give me an equal result and therefore are equally viable strategies.
If I size down to make him indifferent to calling/folding, sure he can call some bets if he wants to. My EV and his EV are still the same though. I get the full pot and he gets nothing.
Forcing your opponent to be indifferent isn't the end goal... it's usually a means to the end goal.
Your goal is to find the strategy with the highest EV (assuming villain responds in a maximally exploitative way). In many spots, this will mean that we'd like to make him indifferent (since if he's not indifferent, he will choose the best strategy), but in spots like this it doesn't matter- we get the whole pot due to the range imbalance as long as we don't bet too small.
(If a player knows your strategy, you should never be able to capture more than 100% of the pot.)
Yes, he should fold every time, but I wouldn't call that an exploitation of our strategy.
What's interesting to me, which I realized while typing the previous response- If we are forced to choose between any of the strategies that involve one bet-size, betting 1/3 pot is the worst option in theory-practice (still in the toy game, but against an imperfect opponent). It's the only strategy against which our EV is exactly 100% of the pot no matter how he responds.
If we bet larger, our minimum EV is 100% of the pot, but if he ever calls a bet we improve that EV to >100% pot.
At Phil, in your example, it appears that you're giving yourself a perfectly polarized range, do you really think that this is the case? I'd assume that while your range could likely have close to 80% equity, I would assume that it's more condensed. In other words, you'll have lots of value hands that beat the majority of your opponents hands, but you'll also have value hands which will lose to some of your opponents stronger hands. In addition, the bottom of your range is still likely ahead a small chunk of the Villain's bottom range.
If the latter is correct, my guess is that theoretically you'll still want to bet larger, but not for the same reason that we deserve the entire pot. Instead, when a very large proportion of your range is made up of very strong hands, then you realize more EV by betting larger and being called less than by betting smaller and being called more.
However, I think it's probably best when oop with this range to have multiple bet sizes. You'll want smaller bet sizes for thinner value which are protected with your nut range. And then you'll want larger bet sizes with the strong part of your range. So maybe Peter wasn't completely off in wanting to size your bet down, but I do think it would be a mistake if that was your only bet size.
I often see in the forums and frankly in the videos the logic that DeSalle makes which is that it's "hard to have a bluff" therefore we should size our bet down. While we don't have many traditional bluffs which don't have any showdown component, this doesn't mean that we shouldn't turn made hands into bluffs. An equilibrium will form with the indifference being at the TOP of our bluffing range where we'll be indifferent between turning a made hand into a bluff and checking it. Therefore, the Villain responds by defending less than 1-A.
I could probably solve this situation on the river in a couple hours, if you wanted to lay out a more detailed version of both players ranges. It's not too hard for me to nearly solve it for multiple bet sizes if you devise ranges where blockers aren't relevant.
still very much enjoying this mamoth series.....congrats on shipping the $109+R last night...any chance you were recording that MTT session? rare to see you grinding NL MTTs exclusively.
Thanks, man :)
I started playing the SCOOP earlier this year in preparation for the WSOP. I ended up doing pretty well and enjoying it too. I've played a few days of MTTs since coming back (just the bigger days with special events), and I plan on playing some WCOOP once it rolls around.
I find MTT days relaxing and fun, and I think it helps me to fit some in between stressful nosebleed games.
I wasn't recording, and likely won't record a ton of WCOOP days, but if I win a big one I'll be happy to do a HH review.
haha Phil Gandalf. crushes PLO MTTs, crushes NLHE MTTs, its white magic!
擦
hey phil why the fuck do i have to pay for a entire year? and its impossible to pay without a creditcard?
1) Paypal is your friend
2) No you don't have to pay for an entire year
3) Profanity is no good
Just use Skrill or MB.
And lmao at Midori saying swearing isn't good.
(PayPal requires that you authorize us to open a time period of one year to bill you monthly. In your Settings, you can downgrade or change your payment type at any time during your subscription with Run It Once)yeah i can change it but why i have to do anything to not pay for more than a month. after i opened paypal it had nothing but a spot to insert my creditcard information
Chael, I might swear like a sailor IRL, but not on the forum!
Amazing series!! I would really enjoy it if you record a couple live hours of a wcoop sunday! In my opinion, you don't need to get HU to make the days recording worth it. I would enjoy a couple hours of live commentary during a WCOOP sunday regardless of your results.
Be the first to add a comment
You must upgrade your account to leave a comment.