Liveblogging a HU match.
 

Mark and I are setting up to play, and I'm going to be giving play-by-play in my lj here. Come on by if you're interested.

Re: Liveblogging a HU match.
 

Nice. Where you playing anyways?

Re: Liveblogging a HU match.
 

I hope you're making Mark drink beer. Him throwing all that math at you has to be -EV.

Re: Liveblogging a HU match.
 

[ QUOTE ]
I hope you're making Mark drink beer. Him throwing all that math at you has to be -EV.

[/ QUOTE ]

T is a much better player than I am, probably more so HU. I'm just a geek donk with too much education.

Re: Liveblogging a HU match.
 

I'll post a more complete hand history later, but some quick stats that might be of interest:

T: raised 40 times (67%) from the button, folded 20 times (33%).
G: raised 31 times (52%) from the button, folded 29 times (48%).

(No limping!)

In the big blind vs. a raise:
T: folded 14 times (45%), called 12 times (39%), reraised 5 times (16%).
G: folded 20 times (50%), called 16 times (40%), reraised 4 times (10%).

T's HU aggression shows here, I think.

I think my defending percentage is about where it should be against an aggressive raiser. Consistently defending less often than 50% is bad in a 1-chip/2-chip game like we were playing... do you see why?

Exercise 1: Work out how often you should defend in a 1-chip/3-chip game.

Exercise 2: Given an opponent (T) who defends 55% of the time, what percentage of the time should Hero (G) be raising the first round?

Re: Liveblogging a HU match.
 

Here's a hand from last night, number 104 in our match. G has the button.

G gets 4478A and raises, T calls. T draws 3, G discards A4 and gets K7. T checks, G checks.

I felt that the odds were good that T had caught up, and that any 2-card draw he has is better than mine. I would bet with an (unpaired) 9 or lower. My two paired cards help marginally here.

T raps pat, G draws 64 (87644). T bets, G folds.

T's perspective:
<font color="white">
I make a loose defense with 54 and draw 239. Mark draws two but I whiff my checkraise attempt. I'm pat, he draws two again and folds the third round, showing 87644.
</font>

Good? Bad? Indifferent?

Re: Liveblogging a HU match.
 

[ QUOTE ]
Here's a hand from last night, number 104 in our match. G has the button.

G gets 4478A and raises, T calls. T draws 3, G discards A4 and gets K7. T checks, G checks.

I felt that the odds were good that T had caught up, and that any 2-card draw he has is better than mine. I would bet with an (unpaired) 9 or lower. My two paired cards help marginally here.

T raps pat, G draws 64 (87644). T bets, G folds.

T's perspective:
<font color="white">
I make a loose defense with 54 and draw 239. Mark draws two but I whiff my checkraise attempt. I'm pat, he draws two again and folds the third round, showing 87644.
</font>

Good? Bad? Indifferent?

[/ QUOTE ]

I think you are 100% on point with your check. He will have improved by 1 card at least 50% if not greater of the time.

TT [img]/images/graemlins/club.gif[/img]

Re: Liveblogging a HU match.
 

[ QUOTE ]
[ QUOTE ]
Here's a hand from last night, number 104 in our match. G has the button.

G gets 4478A and raises, T calls. T draws 3, G discards A4 and gets K7. T checks, G checks.

I felt that the odds were good that T had caught up, and that any 2-card draw he has is better than mine. I would bet with an (unpaired) 9 or lower. My two paired cards help marginally here.

T raps pat, G draws 64 (87644). T bets, G folds.

T's perspective:
<font color="white">
I make a loose defense with 54 and draw 239. Mark draws two but I whiff my checkraise attempt. I'm pat, he draws two again and folds the third round, showing 87644.
</font>

Good? Bad? Indifferent?

[/ QUOTE ]

I think you are 100% on point with your check. He will have improved by 1 card at least 50% if not greater of the time.

TT [img]/images/graemlins/club.gif[/img]

[/ QUOTE ]

Well, let's see here...

I know T has two low cards, and I've see 44778KA. That leaves 43 cards in the deck. T's best case for improving is holding 47, so any 23568 and maybe 9 is good (23 cards), worst case is having 23, so a 45678 or possibly 9 is good (19 cards).

Thus, given what I know, T will brick on all three cards somewhere between (24/43)*(23/42)*(22/41) = 16% and (20/43)*(19/42)*(18/41) = 9% of the time. Since there are only 4 SB in the pot, trying to take it down here is not profitable. (No better than 5.25:1 odds of doing so.)

If he has improved to a 2-card draw, what can he have that I'm obviously ahead of? Not much, by my estimation:

985 (very rough defense, might fold)
984
983
982
97w
9ww (some of these matchups are interesting, T could be ahead.)
876 (unlikely)
875

Compare with
87w
86w
8ww
76w
7ww
6ww
www
+ all one-card draws
+ all pat hands

The three card nine draws are nonobvious, I might have to think a bit more about them.

ETA: of course, the combination of folding equity and whatever equity I do have against T may put me over 50%. But I'm comfortable with a check here.

Re: Liveblogging a HU match.
 

I would almost certainly fold 98x to a bet, FWIW.

I think most of the time when you have a two-card draw and the opponent draws three, it's not good to autobet, especially if you have position. In fact I think Mark bets here substantially too often, and it's part of why I felt comfortable defending with some pretty crappy three-card draws.

The vast majority of situations I've identified where betting a card ahead is bad are when you've drawn 2, villain has drawn 3, and you either improve by one card or don't improve at all.