Thursday, October 05, 2006
Fold Equity - Bringing It All Together
Kieran asked below for some examples, so I thought this would be a good reason to bring this all together in one place. Most of it is already scattered here and there throughout this blog, and I link the posts where relevant. The linked posts go into more detail.
Here's the classic situation I'm talking about. Blinds are 400-800. Hero has 10K in the cut-off, and it's folded round to him with AT. Most people, in this spot, would do something like make it 2400, either without thinking at all or with the rationale "I can get away from this if someone reraises me". That's fine if your opponents simply reraise on the strength of their own hand, but if I'm sitting in the blinds, or someone like me, or even a non-thinking maniac, you have a problem.
The problem is that I'm going to reraise you with a bunch of hands, like small pairs, suited connectors etc. If the conditions are right, I might reraise you with nothing. I can do this because I can estimate, from my hand and a guess at your raising and calling ranges, how likely I am to win the hand if you call. Typically it's 30-35% with something decent, and no lower than 25% even with filth. Adding this to the chance you will fold gives me a profitable semi-bluffing situation.
If I was in the cut-off with AT here I would just move in. I know that this is a profitable move because I have worked it out in advance (various posts here) . It looks clumsy and you'll have to field a lot of bitchy comments live, but there's no doubt in my mind that it's the best play. Basically, you should not put 20% or more of your stack in the pot if you don't want to face a reraise. I don't know how often I've seen this live and on TV, often from players with big reputations, where they raise just the right amount to invite a reraise, and then agonise over the call.
I've never met him, but one guy who keeps cropping up with these reraises in tournament reports seems to be Ali Mallu, and he's a frequent winner [1]. A useful metric in the blinds is, if your stack [2] is about 2.5-4 times the pot after the late position raise, you should consider moving in. At the higher end, it's better to have something half-decent like 76s or KJ maybe. Between 2.5-3, you can consider it with any two. And if you have a pair, however small, it's basically an automatic reraise with a stack of 5 x the pot or less.
A similar situation can crop up the other way round with smaller blinds. Suppose it's 100-200 now, and the cut-off makes it 600 out of 10K. You're in the blind with 88 or AJ or something. Putting in 2500 here is just giving the same situation back to the raiser. He can move in and give you the dilemma. Here I would either move in or (more usually) flat call and check-raise a favourable flop.
Now, Kieron also asked for an example where we don't have fold equity. At the risk of stating the obvious, we don't have fold equity when someone is already all-in. You would think that was obvious, but wait a moment. Let's say it's 100-200 and a short stack moves in for 800. Hero, with 4000, picks up 66 and decides to "get it heads up". That's pretty common in my experience. It's also pretty ... let's be generous and say risky (second half of that post). What I'm saying there is, suppose this master plan works and we "get it heads up" with our 66. Given the small chance that our opponent might have an undercard or underpair (and don't forget he can find an overpair too), we're now 55-60% to win 1900, which is an EV of about +300. Now, supposing no one had raised, would you smash in your 4000 with 66 to win 300 in blinds ? Say, with 5 people behind you ? If you would, you'd better not :-). And most people wouldn't. The "isolation reraise" doesn't add up to a good play here.
While I'm on a roll, another example is raising when the big blind is already committed. Sometimes, especially live, the best of us can do this inadvertently (and so can I). Make it part of your pre-raise routine to check the big blind's stack, and tighten right up if he's committed.
This stuff is huge in tournaments. Sklansky said "Tournament pros know that these loose move-ins [as in my first example] are correct; in fact, this knowledge is the main reason many of them win money at all playing tournaments". When, for example, Greenstein says in his book that if he has 10 big blinds or less, he always moves in or folds [except with AA], I'd love to know whether he's been through the analysis as above ; he instinctively knows this ; or that's what he found to work through experience [3]. I can't stress it highly enough. Don't let others steal-reraise you all in. Steal-reraise other players all in as much as you can.
[1] Of course I've no idea whether he's an overall money winner or not ...
[2] Or your opponent's stack including his raise (but not his call). Simply, how much you have to put in.
[3] I'm not implying that Greenstein is one of Sklansky's players who wouldn't win otherwise.
Here's the classic situation I'm talking about. Blinds are 400-800. Hero has 10K in the cut-off, and it's folded round to him with AT. Most people, in this spot, would do something like make it 2400, either without thinking at all or with the rationale "I can get away from this if someone reraises me". That's fine if your opponents simply reraise on the strength of their own hand, but if I'm sitting in the blinds, or someone like me, or even a non-thinking maniac, you have a problem.
The problem is that I'm going to reraise you with a bunch of hands, like small pairs, suited connectors etc. If the conditions are right, I might reraise you with nothing. I can do this because I can estimate, from my hand and a guess at your raising and calling ranges, how likely I am to win the hand if you call. Typically it's 30-35% with something decent, and no lower than 25% even with filth. Adding this to the chance you will fold gives me a profitable semi-bluffing situation.
If I was in the cut-off with AT here I would just move in. I know that this is a profitable move because I have worked it out in advance (various posts here) . It looks clumsy and you'll have to field a lot of bitchy comments live, but there's no doubt in my mind that it's the best play. Basically, you should not put 20% or more of your stack in the pot if you don't want to face a reraise. I don't know how often I've seen this live and on TV, often from players with big reputations, where they raise just the right amount to invite a reraise, and then agonise over the call.
I've never met him, but one guy who keeps cropping up with these reraises in tournament reports seems to be Ali Mallu, and he's a frequent winner [1]. A useful metric in the blinds is, if your stack [2] is about 2.5-4 times the pot after the late position raise, you should consider moving in. At the higher end, it's better to have something half-decent like 76s or KJ maybe. Between 2.5-3, you can consider it with any two. And if you have a pair, however small, it's basically an automatic reraise with a stack of 5 x the pot or less.
A similar situation can crop up the other way round with smaller blinds. Suppose it's 100-200 now, and the cut-off makes it 600 out of 10K. You're in the blind with 88 or AJ or something. Putting in 2500 here is just giving the same situation back to the raiser. He can move in and give you the dilemma. Here I would either move in or (more usually) flat call and check-raise a favourable flop.
Now, Kieron also asked for an example where we don't have fold equity. At the risk of stating the obvious, we don't have fold equity when someone is already all-in. You would think that was obvious, but wait a moment. Let's say it's 100-200 and a short stack moves in for 800. Hero, with 4000, picks up 66 and decides to "get it heads up". That's pretty common in my experience. It's also pretty ... let's be generous and say risky (second half of that post). What I'm saying there is, suppose this master plan works and we "get it heads up" with our 66. Given the small chance that our opponent might have an undercard or underpair (and don't forget he can find an overpair too), we're now 55-60% to win 1900, which is an EV of about +300. Now, supposing no one had raised, would you smash in your 4000 with 66 to win 300 in blinds ? Say, with 5 people behind you ? If you would, you'd better not :-). And most people wouldn't. The "isolation reraise" doesn't add up to a good play here.
While I'm on a roll, another example is raising when the big blind is already committed. Sometimes, especially live, the best of us can do this inadvertently (and so can I). Make it part of your pre-raise routine to check the big blind's stack, and tighten right up if he's committed.
This stuff is huge in tournaments. Sklansky said "Tournament pros know that these loose move-ins [as in my first example] are correct; in fact, this knowledge is the main reason many of them win money at all playing tournaments". When, for example, Greenstein says in his book that if he has 10 big blinds or less, he always moves in or folds [except with AA], I'd love to know whether he's been through the analysis as above ; he instinctively knows this ; or that's what he found to work through experience [3]. I can't stress it highly enough. Don't let others steal-reraise you all in. Steal-reraise other players all in as much as you can.
[1] Of course I've no idea whether he's an overall money winner or not ...
[2] Or your opponent's stack including his raise (but not his call). Simply, how much you have to put in.
[3] I'm not implying that Greenstein is one of Sklansky's players who wouldn't win otherwise.
Comments:
<< Home
Andy,
Shhhhh. You’re giving too much away.
Incidentally, referring back to your AT on the button example. There was a piece written a few years ago by the much derided Paul Samuels in which he advocated moving in with 22 in similar circumstances. He was widely ridiculed for this move. I don’t know if you were one of his detractors and I’m not even sure if the situation he described was identical to the one you offer. I seem to remember his figures stacked up, but a number of well known players concluded he was barking. He may well be, but I think in this situation he might have been right.
Shhhhh. You’re giving too much away.
Incidentally, referring back to your AT on the button example. There was a piece written a few years ago by the much derided Paul Samuels in which he advocated moving in with 22 in similar circumstances. He was widely ridiculed for this move. I don’t know if you were one of his detractors and I’m not even sure if the situation he described was identical to the one you offer. I seem to remember his figures stacked up, but a number of well known players concluded he was barking. He may well be, but I think in this situation he might have been right.
I probably am giving too much away but it was all in this blog here and there anyway. What's maybe more surprising is that out of all the poker websites, books and magazines I've read, I've so rarely seen this concept mentioned, and basically never set out in full as here. Sklansky touches on it in his latest No-Limit book ; reading between the lines of various blogs, a few players are clearly aware of it (Matros, Chan, Grundy). But it just goes to show how pointless most of these strategy articles are. IMO this is fundamental to playing NL tournaments.
Speaking of pointless strategy articles :-), I don't remember that one from PS. For me, with a small pair it would depend on the stack sizes, but there's no doubt again that moving in is much better than putting 1/4 of it in.
Andy.
Speaking of pointless strategy articles :-), I don't remember that one from PS. For me, with a small pair it would depend on the stack sizes, but there's no doubt again that moving in is much better than putting 1/4 of it in.
Andy.
Actually, I've found it (the PS article).
http://www.pokerpages.com/articles/archives/samuel16.htm
He was making that move with 25x bb so not quite the same at all. My mind was playing tricks with me.
http://www.pokerpages.com/articles/archives/samuel16.htm
He was making that move with 25x bb so not quite the same at all. My mind was playing tricks with me.
AT on cut-off with 400-800 blinds and 10k effective stack.
What about making it 3k and calling any push? Or is it too problematic play if BB chooses call-and-push-any-flop approach?
aksu
What about making it 3k and calling any push? Or is it too problematic play if BB chooses call-and-push-any-flop approach?
aksu
A call from the blinds is the problem yes. If your opponent doesn't have an Ace and correctly puts you on the Ace then he can give you a real problem, much more so than if he auto-pushes.
Andy.
Post a Comment
Andy.
<< Home