Razz: Defending against steal - Page 3

Praxising · #21 11-05-2007, 10:17 PM

[ QUOTE ]
Right, Huck Seed. I'm just saying that, just because he'd fold A23 in some spots, that doesn't mean he (or hellmuth) would not raise here every time.

[/ QUOTE ]
I agree. I didn't invoke to inform this specific situation, it was about generally not having some "must play" about any hand because of theoretical mathematical advantage only. As for ol' Phil, his advice is don't play them unless you are stealing. Doesn't mean he follows his own advice, of course.

As long as we are on this, if you pull out all your starting 8s and all your starting 7s, what's the difference in win rate? Might as well have some hard data here, might show something more useful to us than opposing opinions no matter where they come from.

And if you get the chance to analyze the math in my two trny hands, I'd appreciate it. I know it's a lot of work.

RustyBrooks · #22 11-05-2007, 10:46 PM

I'm playing 2 tourneys right now but I'll read your hands later. I am definitely winning with my 8s... I'll get exact #s later.

iStackBooks · #23 11-06-2007, 04:51 AM

[ QUOTE ]
Some believe "situational" is at least as important to consider as mathematical

[/ QUOTE ]

It sounds like you are arguing for me? The situation is exactly what should be influencing your decision to play a bad 8...

Praxising · #24 11-06-2007, 01:16 PM

[ QUOTE ]
I'm playing 2 tourneys right now but I'll read your hands later. I am definitely winning with my 8s... I'll get exact #s later.

[/ QUOTE ]I wouldn't be surprised if you are winning with 8s, or if you weren't, I guess. I was interested in the comparison of win/loss between 8s and 7s. You know, like how much difference does it really make?

[ QUOTE ]
It sounds like you are arguing for me? The situation is exactly what should be influencing your decision to play a bad 8...

[/ QUOTE ] I'm not arguing at all. I'm not attempting a proof. However, can you give me a situation where you would play 876 and one where you wouldn't?

RustyBrooks · #25 11-06-2007, 01:41 PM

[ QUOTE ]
[ QUOTE ]
I'm playing 2 tourneys right now but I'll read your hands later. I am definitely winning with my 8s... I'll get exact #s later.

[/ QUOTE ]I wouldn't be surprised if you are winning with 8s, or if you weren't, I guess. I was interested in the comparison of win/loss between 8s and 7s. You know, like how much difference does it really make?

[/ QUOTE ]

OK, I dragged out my stats.
9s : .25bb/hand
8s: .49bb/hand
7s: .66bb/hand
6s: .75bb/hand
5s: .75bb/hand

This is lifetime stats, including .25/.5 through 1/2. If I limit it to .5/1 and 1/2 I get very similar numbers. 40k hands sample. Note that I am actually a minescule winner over this sample size. It's kind of embarassing actually. A better player would probably have better numbers for all of these.

If you use pokertracker I can describe to you how to get these stats.

Note that I do fold some 8s, depending on relative position, cards out, my up card, etc, etc. Situations where I'd fold an 8 include:
* I'm first to act with something like 875, 8 up - I instamuck here unless there are very few low cards showing, or there are like lots of 8s and 7s up.
* I have a quality 8 like A48 with the 8 up, and I'd have to call 2 bets cold, or I'd have to call 1 bet but there are too many low cards after me, and they're my outs. I'd actually probably fold to 2 bets even if I had the A up unless the rest of the up cards were pretty favorable to me.

According to pokertracker, I play 3 card 8s 92% of the time, I go to showdown with them 33% of the time, and I win at showdown 55% of the time. My showdown rates for better starting cards are a bit higher, but the win rates are not that much better. So I probably play 3 card 8s better post-3rd. That's kind of ironic.

Situationally, when someone raises in steal position and you have an awful 8, you have no information - his hand can range from any 2 down, to the best 2 down, and he'll play 3rd the same way. The ONLY time I'd fold 876 here, is if the player was so bad and predictable that I could be almost positive that he would not steal. And I have not met that player yet.

AlanBostick · #26 11-06-2007, 02:20 PM

If the alleged ante thief is raising through one or two low cards, you should give her credit for a hand ... or at least wait for a good situation to play back. (Note, though, that if you are the bring-in with an 8 or 9, that means that the deck is poor in good ones.)

But if you're in a situation where you have a king in the door, its folded to the player to your right, and she raises, and you have two good ones in the hole lower than her door card, you should consider defending.

With a good card in the door, the presumed thief should be raising with any two in the hole, which means your hand is actually ahead of her range. If you catch good on fourth and fifth, especially if the villain bricks, you are in a good spot. If you brick on fourth, give it up and wait for a better spot.

At Sabyl Cohen and Steve Landrum's wedding last August, I had a long talk about defending bring-ins in stud games with Bill Chen and Jerrod Ankemann. We talked about razz, and I took the position that the bring-in with a brick in the door was a mandatory fold. Bill argued strenuously what I said above: that if you have two good ones in the hole, you are ahead of your attackers possible range of hands and should play.

Praxising · #27 11-06-2007, 02:23 PM

[ QUOTE ]

OK, I dragged out my stats.
9s : .25bb/hand
8s: .49bb/hand
7s: .66bb/hand
6s: .75bb/hand
5s: .75bb/hand

This is lifetime stats, including .25/.5 through 1/2. If I limit it to .5/1 and 1/2 I get very similar numbers. 40k hands sample. Note that I am actually a minescule winner over this sample size. It's kind of embarassing actually.

[/ QUOTE ]
Thanks for doing this, I'd like to veer off-topic a second. This looks pretty good to me, but I'm not a math person, as we all know. Still, you are averaging .58 bb per hand. Not including your 4s and 3s which would bring it up, let's call it .6. So, per hour (assuming you play few Ts or higher) you are going to be playing maybe ten hands? So wouldn't this make your win rate 6bb per hour? I thought that was good? Looks good to me, anyway.

So, back to topic, I think after I had a big slump, I got very conservative in terms of limiting variance. I see here we have a 24% difference between the 9 and 8 and a 17% between the 8 and 7. If you are very careful with how you play the 8s, you must be twice as careful in how you play the 9s. I'm wondering, of these hands, how many of the winning 9s and 8s were successful steals?

I think the differences in the win rates flattening out after 7 is pretty interesting.

RustyBrooks · #28 11-06-2007, 02:36 PM

[ QUOTE ]

Thanks for doing this, I'd like to veer off-topic a second. This looks pretty good to me, but I'm not a math person, as we all know. Still, you are averaging .58 bb per hand. Not including your 4s and 3s which would bring it up, let's call it .6. So, per hour (assuming you play few Ts or higher) you are going to be playing maybe ten hands? So wouldn't this make your win rate 6bb per hour? I thought that was good? Looks good to me, anyway.

[/ QUOTE ]

I'm paying .1bb/hand (10bb/100) in antes, and .03bb/hand (3bb/100) in bringins. But actually I play more like 25 hands per hundred. If I can average .6bb/hand that I play, that should be
+15bb/100 from winning
-13bb/100 from antes/bringins
for a total of 2bb/100
But also, I lose money stealing with worse hands, defending with worse hands. Or if my average is more like .5bb/hand that I play, then it's
+12.5bb/100 from winning
-13bb/100 from antes/bringins
whoops, -.5bb/100 after antes/bringins.

Which goes to show why averaging .5bb/hand and .6bb/hand that you play is a huge difference. And really, this is why missing a correct 1bb on a regular basis, or making incorrect 1bb calls on a regular basis, can easily swing you from a winning to a losing player.

RustyBrooks · #29 11-06-2007, 02:40 PM

Hmm, actually that's not quite right, because the antes/bringins are already factored into the hands that I won (that is I won .6bb/100 AFTER the antes/bringins). Uh, maybe I'll work this out later.

Hint: I'm barely above breaking even. About +.5bb/100. I feel bad about that, but there are hardly any long-term winners at 1/2 who are doing better (on FTP, I have almost no stats on people at PS), so I don't feel toooo bad. I've been really studying the ones who are maintaining a better win rate than me, and talking to some of them. I'm pretty sure one of them is a 2+2er.

SGspecial · #30 11-06-2007, 05:42 PM

[ QUOTE ]
Hmm, actually that's not quite right, because the antes/bringins are already factored into the hands that I won (that is I won .6bb/100 AFTER the antes/bringins). Uh, maybe I'll work this out later.

Hint: I'm barely above breaking even. About +.5bb/100. I feel bad about that, but there are hardly any long-term winners at 1/2 who are doing better (on FTP, I have almost no stats on people at PS), so I don't feel toooo bad. I've been really studying the ones who are maintaining a better win rate than me, and talking to some of them. I'm pretty sure one of them is a 2+2er.

[/ QUOTE ]
You may be neglecting the fact that at the microlevels you're usually playing against a hugely profitable player called "the rake" Making any profit at all I think puts you in the category of "good" player, and it probably takes a top 10% player to beat the game for >2BB/100.