The Research
Nov. 25th, 2006 07:27 pm![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
peterbirksGawd, this research is hard-bloody-work.
I decided that, if I was going to play these stupid freeroll points-spending tournaments, it would be a good excuse to do some analysis of my Full Tilt game at the same time. This is a slow process. I've put in three hours, and I've scratched the surface of one small area.
I noticed that, although my win rate in the non-blind areas wasn't exactly stellar, neither was it disappointing. It was my Small Blind and Big Blind performances that were worse than I would wish. Clearly, if you are nine-handed, a relatively bad performance in the blinds will have a 10% greater effect than if you are 10-handed. So, I had to work out how to improve my performance in the blinds, which was currently clocking along at minus 0.14 BB per hand in the SB and minus 0.25BB per hand in the BB.
In fact, this is only slightly worse than my performance on other sites, I think (say, minus 0.12 and minus 0.23, on average), but the slightly poorer performances in some other positions (three and four off the button) and the fact that it's nine-handed, conspire to turn me from a 1BB per hundred winner into a 1BB per hundred loser.
So, I put in a filter to look at hands where I was in the Big Blind and there was a steal attempt against me. My conclusion (after much research) was that I was folding the Big Blind too often to a steal attempt.
To give you some numbers, of 991 hands where the first person to bet raised from either CO or button or small blind, I am winning just 107 of them. In 165 cases I defend and fold either on flop, turn or river. In all the rest (719 cases), I fold pre-flop.
Now, I looked at the average raise percentages in steal positions, and ran that through PokerStove.
Now, here's a question. What percentage equity do I need in the big blind to defend against a possible steal? In other words, if I put a range of hands into PokerStove, and I put my actual hand into Pokerstove, what equity do I need to call and what equity do I need to raise? And how does this alter if the potential stealer is in the SB rather than on the button?
Note that we are not talking about the strength of hands on the button or in the small blind (in fact, I noted that the strength of hands in the small blind was significantly higher than it was on the button). All I want to know is what equity percentage I need to make a call correct.
And what percentage do I need for a reraise? Once again, how does this percentage change from OOP to when I have position (i.e., CO or button raise vs SB raise)?
I have my own numbers in my head but, if those numbers are correct, then I should be playing a FAR wider range of hands from the Big Blind than I currently am.
Of course, once I decide on this, I then have to develop a strategy for how to play those hands. But here's a typical example where I think I have been going wrong. I think that this is one of the worst examples of my misplay. Most of them are far more marginal than this!
What it looks like to me is that I am being far too weak-tight OOP.
I am in the BB and get Ad Kc.
Button raises. Now, I don't know much about the guy at the time, but his figures currently show a 40% raise rate in this position. The default raise rate appears to be 28% to 32% (this is FTP only), with button raises being the likeliest, followed by CO raises, followed (a long way behind) by SB steals. In fact, SB steals aren't steals at all; they are nearly all genuine hands. So, button raises may be very loose indeed.
However, I didn't know all this at the time, so I just call.
This is mistake number one. I should three-bet. But, let's not worry about that for the moment.
Flop comes Jc Th 4c.
Even if button has capped it from my three-bet pre-flop, I should bet here. But, suppose I am still in my old-style mode. I've flat-called pre-flop.
I should still bet out. I have more than 50% equity against his range of hands. If I check, it should be with the intention of check-raising. If I am three-bet by the guy, I should cap it, and bet the turn. Only if I am raised on the turn, should I go into "call it down" mode. But no matter what happens with this hand, at no point should I fold. because I always have the odds to call against his initial raising range.
Now, I hear you say, this all sounds a bit wrong.
And, yes, in certain cases, it is. If your opponent is normally very unagressive post-flop, and if he starts three-betting you in a $1-$2 game, the chances are you are beaten, but you are still getting the odds to call it down to the end.
This, I think, is a major leak for me. I have simply been folding in these, ever-more frequent, heads-up situations when I have been getting odds to call. Now, this is an extreme example. But I could have something like KQ off against the more aggressive raiser, and still, just, be getting odds to call to the end (I think). Others might prefer the three-bet pre-flop and the bet out. It all depends here on the style of your opponent. Is he likely to fold a better hand? Is he likely to bet a worse one?
Anyway, if anyone has any suggestions onthe equity percentages I should be looking at pre-flop when I am in the big blind, I'd be eternally grateful. So much so, that I'll actually publish here a table of averages and list of hands. See, fair trade (and somewhat cheered) Birks.
++++++++++++++++
I read a fascinating piece this morning on brain damage. The brain fascinates me. Because of ancient concepts such as "the soul" and Cartesian dualism, research into the physical aspects of how our brain makes us "think" are way behind research into the rest of the body. And ever more weirdieties crop up. One woman turned up with "Cotard Delusion", a belief that she was dead. After some research, it was theorized that this was, chemically/physically, akin to "Capgras delusion", where a person suddenly thinks that someone very close to them although physically identical to that person, is actually an imposter.
The brain damage is the same, but in the first case, the person is someone who "internally attributes" (sees themselves as the cause of events) whereas the the other person "externally attributes" (sees events as being caused by other people).
In this instance, the person suffering from the illness (it's the same illness, it just manifests itself as two different delusions) recognizes the person as a loved one, but the emotional reaction, the "love", isn't there. The Capgras sufferer immediately assumes that this must be because an imposter has appeared in the place of the person that he or she loves, because if it wasn't an imposter, the love would still be there. But the Cotard sufferer says "goodness, why am I feeling no love for this person whom I love deeply? I must be dead. Is this heaven?"
To suffer this kind of brain damage must be horrific, but, in my own self-analystical way, it would also be fascinating. If you are a reader of Phil Dick, and aware that the brain plays the weirdest tricks when it gets physically damaged, it must be almost exciting to be in this position where one part of your brain knows something rationally, but the other part of your brain feels that the opposite is true, emotionally.
I think that I'm fairly strongly "internally attributional" -- perhaps to an extreme degree. I frequently have to tell myself that I am not the cause of things and that I am not the centre of things. People won't notice if I don't turn up to a party, and that couple laughing behind me on the escalator are not laughing at me. This isn't the same as paranoia, although there are links to it, obviously. perhaps it is part of the whole package. If you don't think like this, at least a bit, then you can't believe that you can make a difference.
Are any serious poker players externally attributional? I dunno, but I rather doubt it. Team players of the worst kind might be. But team leaders, surely not. And individualists like poker players? No way. Of course I'm the centre of attention! Look at me!.
I decided that, if I was going to play these stupid freeroll points-spending tournaments, it would be a good excuse to do some analysis of my Full Tilt game at the same time. This is a slow process. I've put in three hours, and I've scratched the surface of one small area.
I noticed that, although my win rate in the non-blind areas wasn't exactly stellar, neither was it disappointing. It was my Small Blind and Big Blind performances that were worse than I would wish. Clearly, if you are nine-handed, a relatively bad performance in the blinds will have a 10% greater effect than if you are 10-handed. So, I had to work out how to improve my performance in the blinds, which was currently clocking along at minus 0.14 BB per hand in the SB and minus 0.25BB per hand in the BB.
In fact, this is only slightly worse than my performance on other sites, I think (say, minus 0.12 and minus 0.23, on average), but the slightly poorer performances in some other positions (three and four off the button) and the fact that it's nine-handed, conspire to turn me from a 1BB per hundred winner into a 1BB per hundred loser.
So, I put in a filter to look at hands where I was in the Big Blind and there was a steal attempt against me. My conclusion (after much research) was that I was folding the Big Blind too often to a steal attempt.
To give you some numbers, of 991 hands where the first person to bet raised from either CO or button or small blind, I am winning just 107 of them. In 165 cases I defend and fold either on flop, turn or river. In all the rest (719 cases), I fold pre-flop.
Now, I looked at the average raise percentages in steal positions, and ran that through PokerStove.
Now, here's a question. What percentage equity do I need in the big blind to defend against a possible steal? In other words, if I put a range of hands into PokerStove, and I put my actual hand into Pokerstove, what equity do I need to call and what equity do I need to raise? And how does this alter if the potential stealer is in the SB rather than on the button?
Note that we are not talking about the strength of hands on the button or in the small blind (in fact, I noted that the strength of hands in the small blind was significantly higher than it was on the button). All I want to know is what equity percentage I need to make a call correct.
And what percentage do I need for a reraise? Once again, how does this percentage change from OOP to when I have position (i.e., CO or button raise vs SB raise)?
I have my own numbers in my head but, if those numbers are correct, then I should be playing a FAR wider range of hands from the Big Blind than I currently am.
Of course, once I decide on this, I then have to develop a strategy for how to play those hands. But here's a typical example where I think I have been going wrong. I think that this is one of the worst examples of my misplay. Most of them are far more marginal than this!
What it looks like to me is that I am being far too weak-tight OOP.
I am in the BB and get Ad Kc.
Button raises. Now, I don't know much about the guy at the time, but his figures currently show a 40% raise rate in this position. The default raise rate appears to be 28% to 32% (this is FTP only), with button raises being the likeliest, followed by CO raises, followed (a long way behind) by SB steals. In fact, SB steals aren't steals at all; they are nearly all genuine hands. So, button raises may be very loose indeed.
However, I didn't know all this at the time, so I just call.
This is mistake number one. I should three-bet. But, let's not worry about that for the moment.
Flop comes Jc Th 4c.
Even if button has capped it from my three-bet pre-flop, I should bet here. But, suppose I am still in my old-style mode. I've flat-called pre-flop.
I should still bet out. I have more than 50% equity against his range of hands. If I check, it should be with the intention of check-raising. If I am three-bet by the guy, I should cap it, and bet the turn. Only if I am raised on the turn, should I go into "call it down" mode. But no matter what happens with this hand, at no point should I fold. because I always have the odds to call against his initial raising range.
Now, I hear you say, this all sounds a bit wrong.
And, yes, in certain cases, it is. If your opponent is normally very unagressive post-flop, and if he starts three-betting you in a $1-$2 game, the chances are you are beaten, but you are still getting the odds to call it down to the end.
This, I think, is a major leak for me. I have simply been folding in these, ever-more frequent, heads-up situations when I have been getting odds to call. Now, this is an extreme example. But I could have something like KQ off against the more aggressive raiser, and still, just, be getting odds to call to the end (I think). Others might prefer the three-bet pre-flop and the bet out. It all depends here on the style of your opponent. Is he likely to fold a better hand? Is he likely to bet a worse one?
Anyway, if anyone has any suggestions onthe equity percentages I should be looking at pre-flop when I am in the big blind, I'd be eternally grateful. So much so, that I'll actually publish here a table of averages and list of hands. See, fair trade (and somewhat cheered) Birks.
++++++++++++++++
I read a fascinating piece this morning on brain damage. The brain fascinates me. Because of ancient concepts such as "the soul" and Cartesian dualism, research into the physical aspects of how our brain makes us "think" are way behind research into the rest of the body. And ever more weirdieties crop up. One woman turned up with "Cotard Delusion", a belief that she was dead. After some research, it was theorized that this was, chemically/physically, akin to "Capgras delusion", where a person suddenly thinks that someone very close to them although physically identical to that person, is actually an imposter.
The brain damage is the same, but in the first case, the person is someone who "internally attributes" (sees themselves as the cause of events) whereas the the other person "externally attributes" (sees events as being caused by other people).
In this instance, the person suffering from the illness (it's the same illness, it just manifests itself as two different delusions) recognizes the person as a loved one, but the emotional reaction, the "love", isn't there. The Capgras sufferer immediately assumes that this must be because an imposter has appeared in the place of the person that he or she loves, because if it wasn't an imposter, the love would still be there. But the Cotard sufferer says "goodness, why am I feeling no love for this person whom I love deeply? I must be dead. Is this heaven?"
To suffer this kind of brain damage must be horrific, but, in my own self-analystical way, it would also be fascinating. If you are a reader of Phil Dick, and aware that the brain plays the weirdest tricks when it gets physically damaged, it must be almost exciting to be in this position where one part of your brain knows something rationally, but the other part of your brain feels that the opposite is true, emotionally.
I think that I'm fairly strongly "internally attributional" -- perhaps to an extreme degree. I frequently have to tell myself that I am not the cause of things and that I am not the centre of things. People won't notice if I don't turn up to a party, and that couple laughing behind me on the escalator are not laughing at me. This isn't the same as paranoia, although there are links to it, obviously. perhaps it is part of the whole package. If you don't think like this, at least a bit, then you can't believe that you can make a difference.
Are any serious poker players externally attributional? I dunno, but I rather doubt it. Team players of the worst kind might be. But team leaders, surely not. And individualists like poker players? No way. Of course I'm the centre of attention! Look at me!.
