![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
So, here's the Matt Matros hand that I mentioned yesterday. It's on page 253, for those who have a copy of "The Making Of A Poker Player".
Blinds are 400-800 with a 100 ante, which makes it a 2,100 pot at a nine-handed table. It's early on day 2, so I think it's fair to assume that we are not anywhere near the bubble.
Matros opens for 2,600 with Jh 8d. Female pro to Matt's left makes it 5,000, the blinds fold and Matros calls.
12,100 in the pot.
Flop comes Q J 7 with two diamonds.
Pro bets 2,000, Matros calls.
16,100 in the pot.
Turn was Ace of diamonds. Matros bets 6,000 and now an odd thing happens. Female pro makes a genuine mistake. She says "raise", but she thought Matros had bet only 1,200 rather than 6,000. So when she put in an extra 4,000, she was compelled to raise the minimum, another 6,000. It's possible that this threw her off her stroke.
Matros thinks that he is losing here, but he hopes to push opponent off KK or AA. He reckons that, if he fails, he has outs and half-outs.
So, he puts himself all-in, calling the 6,000 raise and putting in an extra 25,000, representing the flush.
Opponent goes into tank, and eventually folds.
It later transpires that opponent had QQ here, second-top set.
Now, let's look at the numbers. There was 16,100 in the pot after the flop. Add the 12,000 from female pro and the 12,000 from Matros, and you have 40,000 in there. Female pro is being asked to put in 25,000 to win 40,000.
How likely does Matros have to be to have the flush to make a call correct?
Well, if it's 100% that Matros has the flush, the call is wrong, because she is getting 7-to-4 about a 7-to-2 shot.
But, suppose Matros makes this kind of move half the time, and opponent knows that he makes this kind of move half the time.
If female pro calls, then 39% of the time Matros will be and front and will win = -25,350
11% of the time Matros will be in front but female pro will boat up = +7150
41% of the time female will be in front and will win = +26,650
9% of female will be in front and Matros will hit flush to win = -5,840
That gives a positive EV from calling of 2,610 chips.
But, clearly, female pro does not think Matros would make that move 50% of the time. She thinks he would only make it something like 10% of the time, which makes her correct action a clear fold.
Matros spends some time early in the book talking about playing the WSOP computer game, and winning the championship a majority of the time. How? By being aggressive and making opponents fold.
Until very recently in major tournaments there was clearly a disconnect between the number of times people made moves like this and the number of times that they were called. For this reason the players who made the aggressive moves could win far more of their fair share of tournaments (see the WSOP computer game example that Matros cites) than if their opponents knew how often such moves owuld be made.
As Andy observed in his blog, "knowledgeable" players are now calling him on thinner values. The "commonsense" view is that you fight back against the stealers by reraising them. But if they are in Kill Phil mode, going all-in, you have to fight back by calling. A larger number of MTT players online these days are doing just that, bubble or not.
Fortunately (for the raisers) there are a sufficient number who are folding.
Most of the stuff that has been written on MTTs (see, for example, "Winning Online") has been from the point of view of the stealer (who is likely to fold to your raises, who is likely to call). But let's look at it from the point of view of the stealee.
There's no hard and fast rule, unfortunately. You have to know each player putting in the raise and their likely ranges (or, in this case, the percentage of times that a play is a semi-bluff and the percentage of times it's a genuinely made hand) and react accordingly. "Always call" is as wrong as "always fold". But the default should perhaps be one of assuming a steal rather than assuming a good hand. Once again, Raymer has it right. Just try to make every play positive EV. Whether that's 20% about a 15% shot or 95% about a 90% shot makes no difference. In the long run, it's being on the right side of the bet that matters, not whether you are odds against or odds on.
Please note, absolutely none of this applies to super satellites, which are a breed apart. :-)
Blinds are 400-800 with a 100 ante, which makes it a 2,100 pot at a nine-handed table. It's early on day 2, so I think it's fair to assume that we are not anywhere near the bubble.
Matros opens for 2,600 with Jh 8d. Female pro to Matt's left makes it 5,000, the blinds fold and Matros calls.
12,100 in the pot.
Flop comes Q J 7 with two diamonds.
Pro bets 2,000, Matros calls.
16,100 in the pot.
Turn was Ace of diamonds. Matros bets 6,000 and now an odd thing happens. Female pro makes a genuine mistake. She says "raise", but she thought Matros had bet only 1,200 rather than 6,000. So when she put in an extra 4,000, she was compelled to raise the minimum, another 6,000. It's possible that this threw her off her stroke.
Matros thinks that he is losing here, but he hopes to push opponent off KK or AA. He reckons that, if he fails, he has outs and half-outs.
So, he puts himself all-in, calling the 6,000 raise and putting in an extra 25,000, representing the flush.
Opponent goes into tank, and eventually folds.
It later transpires that opponent had QQ here, second-top set.
Now, let's look at the numbers. There was 16,100 in the pot after the flop. Add the 12,000 from female pro and the 12,000 from Matros, and you have 40,000 in there. Female pro is being asked to put in 25,000 to win 40,000.
How likely does Matros have to be to have the flush to make a call correct?
Well, if it's 100% that Matros has the flush, the call is wrong, because she is getting 7-to-4 about a 7-to-2 shot.
But, suppose Matros makes this kind of move half the time, and opponent knows that he makes this kind of move half the time.
If female pro calls, then 39% of the time Matros will be and front and will win = -25,350
11% of the time Matros will be in front but female pro will boat up = +7150
41% of the time female will be in front and will win = +26,650
9% of female will be in front and Matros will hit flush to win = -5,840
That gives a positive EV from calling of 2,610 chips.
But, clearly, female pro does not think Matros would make that move 50% of the time. She thinks he would only make it something like 10% of the time, which makes her correct action a clear fold.
Matros spends some time early in the book talking about playing the WSOP computer game, and winning the championship a majority of the time. How? By being aggressive and making opponents fold.
Until very recently in major tournaments there was clearly a disconnect between the number of times people made moves like this and the number of times that they were called. For this reason the players who made the aggressive moves could win far more of their fair share of tournaments (see the WSOP computer game example that Matros cites) than if their opponents knew how often such moves owuld be made.
As Andy observed in his blog, "knowledgeable" players are now calling him on thinner values. The "commonsense" view is that you fight back against the stealers by reraising them. But if they are in Kill Phil mode, going all-in, you have to fight back by calling. A larger number of MTT players online these days are doing just that, bubble or not.
Fortunately (for the raisers) there are a sufficient number who are folding.
Most of the stuff that has been written on MTTs (see, for example, "Winning Online") has been from the point of view of the stealer (who is likely to fold to your raises, who is likely to call). But let's look at it from the point of view of the stealee.
There's no hard and fast rule, unfortunately. You have to know each player putting in the raise and their likely ranges (or, in this case, the percentage of times that a play is a semi-bluff and the percentage of times it's a genuinely made hand) and react accordingly. "Always call" is as wrong as "always fold". But the default should perhaps be one of assuming a steal rather than assuming a good hand. Once again, Raymer has it right. Just try to make every play positive EV. Whether that's 20% about a 15% shot or 95% about a 90% shot makes no difference. In the long run, it's being on the right side of the bet that matters, not whether you are odds against or odds on.
Please note, absolutely none of this applies to super satellites, which are a breed apart. :-)
