I'm all confused all of a sudden [Archive] - Two Plus Two Older Archives

ilya

05-17-2005, 01:46 AM

So I did some calculations to see what win% would be necessary for a push from CO to be profitable given

1, equal stacks of 5xBB
2, varying probalities of getting called in one of the three spots behind you(30%, 40%, 50%).

The win% required drops dramatically as the number of players increases. What confuses me, however, is how high the necesary win% is if there are only 4 players left. For example, if you figure to get called a total of only 30% of the time (each opponent calls with 66+, AT+, KQ), your win% against your opponents' range must be at least 38%! So, pushing a hand like QTo is drastically wrong from an ICM perspective...yet it feels necessary to push it. Is the ICM *that* warped near the bubble?! Or has my intuition drifted *that* far off course?

curtains

05-17-2005, 01:55 AM

[ QUOTE ]
So I did some calculations to see what win% would be necessary for a push from CO to be profitable given

1, equal stacks of 5xBB
2, varying probalities of getting called in one of the three spots behind you(30%, 40%, 50%).

The win% required drops dramatically as the number of players increases. What confuses me, however, is how high the necesary win% is if there are only 4 players left. For example, if you figure to get called a total of only 30% of the time (each opponent calls with 66+, AT+, KQ), your win% against your opponents' range must be at least 38%! So, pushing a hand like QTo is drastically wrong from an ICM perspective...yet it feels necessary to push it. Is the ICM *that* warped near the bubble?! Or has my intuition drifted *that* far off course?

[/ QUOTE ]

That calling range is really as high as 30% with 3 players to act??

dfscott

05-17-2005, 01:58 AM

If I understand you correctly, I've noticed something similar as well when checking my tourneys in eastbay's calculator. The CO position on the bubble is extremely unfavorable, and hands that are easy button pushes are often very negative EV in the CO.

I've always attributed this to the fact that FE is such a large part of your win% at this point, and every additional potential caller takes a big chunk out of that. I even did some ballpark calculations on when pushing any two is correct the majority of the time (note: these are very rough , are based on a PP structure, assume an "average" calling range regardless of stack size, and do not take relative stack size of the potential callers into account).

Caveated per above, I discovered that with 9BB, a SB push with any two is normally +EV. With 6BB, you can do it on the button. But you need to have less than 3BB to do it from the CO, and of course by then, your FE is fairly weak, so the model breaks down. So, it is rare that pushing any two from the CO is ever correct, which may be why you run that drop-off.

However, this is all idle speculation and I have nothing to back it up, so I'd be interested to hear the statisticians opinions...

ilya

05-17-2005, 01:58 AM

[ QUOTE ]
[ QUOTE ]
So I did some calculations to see what win% would be necessary for a push from CO to be profitable given

1, equal stacks of 5xBB
2, varying probalities of getting called in one of the three spots behind you(30%, 40%, 50%).

The win% required drops dramatically as the number of players increases. What confuses me, however, is how high the necesary win% is if there are only 4 players left. For example, if you figure to get called a total of only 30% of the time (each opponent calls with 66+, AT+, KQ), your win% against your opponents' range must be at least 38%! So, pushing a hand like QTo is drastically wrong from an ICM perspective...yet it feels necessary to push it. Is the ICM *that* warped near the bubble?! Or has my intuition drifted *that* far off course?

[/ QUOTE ]

That calling range is really as high as 30% with 3 players to act??

[/ QUOTE ]

Well, no. It's 28.5% for that range.

ilya

05-17-2005, 02:03 AM

[ QUOTE ]

I've always attributed this to the fact that FE is such a large part of your win% at this point

[/ QUOTE ]

Actually it seems to be more of a function of the value of having any chips at all so close to the money. When it's 10-handed with everyone at 5xBB, and the same calling ranges for your 3 remaining opponents, your win% needs to be only 20% (!!) for pushing to be ICM-correct. Even if there are only 6 players, the necessary win% is still only a modest 26%.

Scuba Chuck

05-17-2005, 03:38 AM

Two factors that make your numbers ...well off.

One is the fact that with equal stacks, there is not likely to be callers of top 28.5%. Second, pushing from UTG is not going to be as attractive as button, and then SB is even more attractive.

ilya

05-17-2005, 03:44 AM

[ QUOTE ]
Two factors that make your numbers ...well off.

One is the fact that with equal stacks, there is not likely to be callers of top 28.5%. Second, pushing from UTG is not going to be as attractive as button, and then SB is even more attractive.

[/ QUOTE ]

Of course I don't think any one opponent is likely to call with as wide a range as top 28.5%. The figure is cumulative. That is, if each of your three opponents will call with 66+, AT+, and KQ (about 10% of hands), then if you push in the CO, you will be called about 28.5% of the time.

I don't quite understand your 2nd point. Do you mean people will be more leery of calling an UTG push since it represents more strength?

Scuba Chuck

05-17-2005, 03:51 AM

[ QUOTE ]
I don't quite understand your 2nd point. Do you mean people will be more leery of calling an UTG push since it represents more strength?

[/ QUOTE ]

You understand my second point. It's more attractive (in fact correct) to push a VERY wide range of hands from SB if folded to you on the bubble. The range of hands narrows as you move from the SB to UTG.

Blarg

05-17-2005, 03:59 AM

Jeez this seems so foreign from the world of the 11's, where people will call with any face card, 98 suited, and the like.

ilya

05-17-2005, 04:10 AM

[ QUOTE ]
[ QUOTE ]
I don't quite understand your 2nd point. Do you mean people will be more leery of calling an UTG push since it represents more strength?

[/ QUOTE ]

You understand my second point. It's more attractive (in fact correct) to push a VERY wide range of hands from SB if folded to you on the bubble. The range of hands narrows as you move from the SB to UTG.

[/ QUOTE ]

Oh yeah, of course. I was just saying that I was surprised to see just how tight you have to be in the CO on the bubble, especially considering how loosely you should push in the same position when it's 6-handed+. I think I'd started being too LAGgy on the bubble-CO.

curtains

05-17-2005, 04:12 AM

btw I ran that scenario in eastbay's and according to that its a reasonably clear push with QTo in cutoff there against those calling standards.

ilya

05-17-2005, 04:20 AM

[ QUOTE ]

btw I ran that scenario in eastbay's and according to that its a reasonably clear push with QTo in cutoff there against those calling standards.

[/ QUOTE ]

?? I ran it in dethgrind's ICM calculator....hmmm

ilya

05-17-2005, 04:34 AM

[ QUOTE ]

btw I ran that scenario in eastbay's and according to that its a reasonably clear push with QTo in cutoff there against those calling standards.

[/ QUOTE ]

EDIT: NEVER MIND, I HAD FAILED TO ACCOUNT FOR THE FACT THAT HOLDING QT MAKES IT LESS LIKELY YOUR OPPONENTS WILL HAVE AN AQ, AT, KQ, QQ, OR TT.

I just don't understand how you arrived at that result. I just ran the numbers again using dethgrind's ICM calculator.

If you fold, and it's folded to the BB (pretty much your worst-case folding scenario), your equity is

0.2504

If you push and win blinds, your 2600 stack will have an equity of

0.2941

If you push and beat BB, your 4200 stack "" ""

0.3902

If you push and beat SB, your 4400 stack "" ""

0.3971

If you push and beat button, your 4600 stack "" ""

0.4036

If you assume that the button, SB, and BB will call you with equal frequency, the three equity values average out to

0.397

So.

You fold, your equity is at least 0.2504

You push, your equity is

0.749*0.2941 +

0.251*0.33 [QTo win% against 66+, AT+, KQ] *0.397 =

= 0.2532

Yes, that's more than your equity if you fold...but that 0.2504 number was the worst-case scenario. If you fold, there may be a confrontation behind you, sharply increasing your equity.

Blarg

05-17-2005, 04:39 AM

I can't imagine a game where at least one out of three people wouldn't have looser calling standards than that anyway.

theredpill5

05-17-2005, 05:46 AM

Where are these calculators that you speak of ? What is eastbay's calculator and where do I find it ?

ilya

05-17-2005, 05:53 AM

[ QUOTE ]
Where are these calculators that you speak of ? What is eastbay's calculator and where do I find it ?

[/ QUOTE ]

eastbay's tool (http://sitngo-analyzer.com/)

shejk

05-17-2005, 07:45 AM

Sure, one might have looser calling standards, but say 99+, AQ+ is not at all unreasonable for quite a few opponents in this scenario.

Barrett's Last Privateer

05-17-2005, 07:53 AM

Perhaps a useful question is, what are your (Blarg at the 11's, Scuba, Ilya, shejk etc. at your repsective levels) calling standards when on the bubble with evenish stacks.

KQ?
KJ?
44?

shejk

05-17-2005, 08:09 AM

Gotta take many more variables into account, I was just making a point that these standards, in some situations against specific opponents will be loose.

Apart from the pusher's standards, I also consider game texture.

Are people to my right stealing a lot?
Is there someone at the table calling too much?
What's my estimated fold equity when I push?

Disclaimer: This is not a complete list.

microbet

05-17-2005, 10:38 AM

I think if you fabricated a hand and posted it you would see that you want decent cards to push here. QT is probably a borderline case. 5 x BB makes you really anxious to push, but the fact that everyone has 5 x BB on the bubble makes you a little more cautious. Usually when you have 5 x BB on the bubble you are a short stack.

Moonsugar

05-17-2005, 10:58 AM

You're "intuition" is off.

Or, you have caught the "push anything from anywhere" disease which seems to be sweeping this forum.

sabre170

05-17-2005, 11:09 AM

I second your remarks.
As long as I can tell an opponent is pushing any two, which I very often can, I can profitably call with any A, K, 2 broadway cards, or pair.

Last night I caught a 43o with A3o. He wasn't desperate, but he thought I was. /images/graemlins/ooo.gif

Sabre170

microbet

05-17-2005, 11:39 AM

You get to loosen up so much that you fold 70% of the time and you win 60% (your range against random) of the 30% of the times you call.

Depending on the situation, maybe you were both right.

KJ o

05-17-2005, 12:23 PM

[ QUOTE ]
Jeez this seems so foreign from the world of the 11's, where people will call with any face card, 98 suited, and the like.

[/ QUOTE ]

Not my experience. Is this because you tend to play at peak hours and I tend to play at non-peak hours, since I'm in Sweden?

I'd love to see figures on stuff like average VPIP/tourney level, buy-in level and hour of the day over a large sample of tourneys. Anyone has anything like that?

Does it differ as much as people often suggest here?

sabre170

05-17-2005, 12:25 PM

Not sure what point you are making.
Me winning once makes up for 4 or 5 folds.

Sabre170

microbet

05-17-2005, 12:48 PM

If the blinds were high enough compared to your stack this wouldn't be true. There is another range of blinds where what you said is true, but it is still a good push considering the chances that the villian's any two hand will win.

You seemed to be vaguely indicting pushing with any two, so I was vaguely supporting pushing with any two.

But, I guess I was just being a troublemaker.

tech

05-17-2005, 01:31 PM

[ QUOTE ]
As long as I can tell an opponent is pushing any two, which I very often can, I can profitably call with any A, K, 2 broadway cards, or pair.

[/ QUOTE ]

Just because you are a favorite over a random hand doesn't mean it is profitable (in terms of $EV) to call.

sabre170

05-17-2005, 02:59 PM

Which is better on the bubble with even stacks?

A. push from button with any 2, when everyone knows you will push with any 2

B. call from BB with top 35%, when button pushed with any 2

I agree it is a matter of style. You might even make a case for both, which takes playing the player out of the game entirely.

Sabre170

sabre170

05-17-2005, 03:01 PM

If I have only 5BB or less, I should call with any 2.
Thanks for the discussion, I thought I was the one being the troublemaker.

I love to play against the "push with any 2" crowd.

Sabre170

ilya

05-17-2005, 03:59 PM

[ QUOTE ]
I second your remarks.
As long as I can tell an opponent is pushing any two, which you almost never really can,

[/ QUOTE ]

ain't wishful thinking sweet

anyway, calling with something like A3o in the situation I described would be pretty bad. You would have to be sure that your opponent was pushing literally _any_ two cards before you could so much as break even $EV-wise.

not saying that you would, just pointing out that merely knowing your opponent is pushing very liberally often still doesn't mean you can call.

ilya

05-17-2005, 04:07 PM

[ QUOTE ]
If I have only 5BB or less, I should call with any 2.

[/ QUOTE ]

This is utter nonsense.

EDIT: I just realized you are probably kidding. Sorry.

ilya

05-17-2005, 04:17 PM

[ QUOTE ]
You're "intuition" is off.

Or, you have caught the "push anything from anywhere" disease which seems to be sweeping this forum.

[/ QUOTE ]

Actually it was my calculations that were off.
*Given the calling ranges I specified for my opponents*, pushing QTo in that spot would be correct, those close enough that pushing would also be reasonable.
Of course, if even one of my opponents will call with a much wider range than 66+, AT+, KQ - whether because he is making a legitimate strategy adjustment or because he just likes to play sheriff - pushing that QTo becomes unprofitable.

ilya

05-17-2005, 04:21 PM

[ QUOTE ]
you win 60% (your range against random) of the 30% of the times you call.

[/ QUOTE ]

He's not calling with a range, he's calling with K2o.

ilya

05-17-2005, 04:26 PM

[ QUOTE ]
Which is better on the bubble with even stacks?

A. push from button with any 2, when everyone knows you will push with any 2

B. call from BB with top 35%, when button pushed with any 2

I agree it is a matter of style. You might even make a case for both, which takes playing the player out of the game entirely.

Sabre170

[/ QUOTE ]

It is not a matter of style. It is a matter of math. If the button knows SB and BB will call very liberally, he should know better than to push any two cards. If the BB knows that button is pushing with a certain range, he should know better than to call just because his hand is an odds favorite over that range.

microbet

05-17-2005, 04:36 PM

Actually, it was A3o. I don't think your reply was entirely serious. If yours wasn't; mine wasn't.

Nottom

05-17-2005, 04:43 PM

[ QUOTE ]
Which is better on the bubble with even stacks?

A. push from button with any 2, when everyone knows you will push with any 2

B. call from BB with top 35%, when button pushed with any 2

I agree it is a matter of style. You might even make a case for both, which takes playing the player out of the game entirely.

Sabre170

[/ QUOTE ]

If you know he pushing any two and he knows you will call with your top 35% than he is making the bigger mistake.

But the fact is you are both making mistakes and giving that lost $EV to the other 2 opponents. Just becasue your mistake is less of a mistake, doesn't make it not a mistake.

Moonsugar

05-17-2005, 05:11 PM

Pushing is probably not correct and you gave the main reason why in your correction post.

sabre170

05-17-2005, 05:43 PM

[ QUOTE ]
If the BB knows that button is pushing with a certain range, he should know better than to call just because his hand is an odds favorite over that range.

[/ QUOTE ]

I'm trying to focus the discussion on the question of how much of a favorite you need to be to call a push of any two.

First you said I don't really know it is any two. Fair enough, but plenty of contributors here do advocate pushing any 2 in the kinds of bubble situations we are discussing.

Now you say it is a matter of math, not style. I think we can reconcile somewhat. My style is to take a 60-70% edge in calling a bubble push. You can beat me by being more selective about what you push, or by catching lucky.

Although I play incredibly tight, I have virtually never actually been blinded out. Have you?

Sabre170

Blarg

05-17-2005, 06:06 PM

[ QUOTE ]
Sure, one might have looser calling standards, but say 99+, AQ+ is not at all unreasonable for quite a few opponents in this scenario.

[/ QUOTE ]

I think a lot has got to depend on the level you're playing. Seriously, I see tons of calls by almost anything, as long as someone is in the blinds. Heck, I'm even seeing OVERcalls with things like 87s after two people have gone all in.

Anyway, calling with any face card and any kicker is extremely common at the 10's. Deciding what kicker constitutes a reasonable range for someone who calls with Aces is kind of absurd in those games, since most people call with any ace unhesitantly. It's just a whole different world.

On a side note, I was looking through my pokertracker imports the other day, sorting by who I played the most games against. Out of about 1800 opponents(not all of my stuff has been imported yet), I've only seen a handful of people as many as three times.

In the 10's, people just come and go and don't act in determinable, schooled, logical patterns all that much. I still take lots of notes, though, every chance I get, just in case.

But I've had to make many adjustments in my game because talk of aggressive bubble play like you find here on 2+2 doesn't work well when such a big percentage of people are literally willing to call with 75o. You may still win, but the way you win chips on the bubble by stealing is not by showing down your inferior cards, but by having everyone fold before you have to show them often enough that it's worth stealing with inferior cards in the first place.

Blarg

05-17-2005, 06:20 PM

[ QUOTE ]
Perhaps a useful question is, what are your (Blarg at the 11's, Scuba, Ilya, shejk etc. at your repsective levels) calling standards when on the bubble with evenish stacks.

KQ?
KJ?
44?

[/ QUOTE ]

It's hard to answer without considering reads. I'd have to add in what shjek said:

[ QUOTE ]
Are people to my right stealing a lot?
Is there someone at the table calling too much?
What's my estimated fold equity when I push?

[/ QUOTE ]

I'd also add, will my calling close the action?

I'm not doing a lot of calling with kings, but will call with low pairs occasionally and medium pairs always, if it's folded down to being heads-up with the raiser. I'll willingly surrender possible, "iffy" equity a lot of times in a calling situation if I feel I can get people to fold their own blinds once the blinds pass me.

But that answer addresses what people at my level push with, and in a very very generalized way, and my response to it, not what people call with, which seems to be more the point of this thread.

Blarg

05-17-2005, 06:29 PM

[ QUOTE ]
[ QUOTE ]
Jeez this seems so foreign from the world of the 11's, where people will call with any face card, 98 suited, and the like.

[/ QUOTE ]

Not my experience. Is this because you tend to play at peak hours and I tend to play at non-peak hours, since I'm in Sweden?

I'd love to see figures on stuff like average VPIP/tourney level, buy-in level and hour of the day over a large sample of tourneys. Anyone has anything like that?

Does it differ as much as people often suggest here?

[/ QUOTE ]

I tend to play evening hours, till midnight or so, PST. I regularly see the calls I'm talking about. I was just pissing myself the other day at getting sucked out on and booted from the tourney 3 times out of 5 or 6 by guys calling with 75 offsuit (two in a row!) and 54 offsuit.

Seriously, these guys have "blinds disease." Kind of like beginning ring players, who feel that their blind is still "their money," and that once they've "invested" it by putting up that blind, they can't throw away their money by folding. They just don't know any better.

Which is great when you don't get sucked out on of course, because you make a lot of easy money. But, when it comes to stealing, it means you can expect to get an incredible percentage of calls on your pushes, and they will often be with astoundingly random hands. Which, though potentially profitable, changes your situation from "trying to steal the blinds" to "showing down the cards I'm trying to steal with very frequently so I have to win almost every push" and "showing down the cards I'm trying to steal with very frequently so people quickly notice if my cards are less than stellar, and then are even more encouraged to call almost every push."

I say this as someone who doesn't push like crazy, either, and am almost always playing people I've never played before, so they have no particular reason to be calling with garbage except for a love of their blind money or a desire to make sure nobody bullies them, or a simple misunderstanding that calling an all-in for your entire stack with 7-5 offsuit when your stack size doesn't force you to do it isn't really a good idea.

I think the times of day may very well matter a lot here. I notice the games get much looser when I play in the evenings. That's why I play those times, and why most people do.

Blarg

05-17-2005, 06:35 PM

[ QUOTE ]
[ QUOTE ]
If the BB knows that button is pushing with a certain range, he should know better than to call just because his hand is an odds favorite over that range.

[/ QUOTE ]

I'm trying to focus the discussion on the question of how much of a favorite you need to be to call a push of any two.

First you said I don't really know it is any two. Fair enough, but plenty of contributors here do advocate pushing any 2 in the kinds of bubble situations we are discussing.

Now you say it is a matter of math, not style. I think we can reconcile somewhat. My style is to take a 60-70% edge in calling a bubble push. You can beat me by being more selective about what you push, or by catching lucky.

Although I play incredibly tight, I have virtually never actually been blinded out. Have you?

Sabre170

[/ QUOTE ]

If you've virtually never been blinded out, you haven't played many games.

And I'd happily say that even if you were the best SNG player in the history of the world.

sabre170

05-17-2005, 06:51 PM

By the standards of many here, I don't play much.
I think I have been genuinely blinded out exactly once in about 1000 SnGs, on 3 sites.

Since keeping better records, I haven't been blinded out in about 200 SnGs on Empire. That's about 10 weeks of play for me.

By "blinded out," I mean forced to go all in to post a blind, and losing.

There have been lots of times I made a desperation push in early position in order to avoid being blinded out in the next few hands.

Sabre170

ilya

05-17-2005, 08:15 PM

[ QUOTE ]
[ QUOTE ]
If the BB knows that button is pushing with a certain range, he should know better than to call just because his hand is an odds favorite over that range.

[/ QUOTE ]

I'm trying to focus the discussion on the question of how much of a favorite you need to be to call a push of any two.

First you said I don't really know it is any two. Fair enough, but plenty of contributors here do advocate pushing any 2 in the kinds of bubble situations we are discussing.

Now you say it is a matter of math, not style. I think we can reconcile somewhat. My style is to take a 60-70% edge in calling a bubble push. You can beat me by being more selective about what you push, or by catching lucky.

Although I play incredibly tight, I have virtually never actually been blinded out. Have you?

Sabre170

[/ QUOTE ]

Well, the edge you need depends on the situation. In the example I gave, a 56% edge would be sufficient to call. In some situations, a 60% edge might not be enough. I definitely see where you are coming from...adjusting to opponents is essential, and people who advocate pushing any 2 regardless of context are often not giving very useful advice.

I can't resist pointing out that A3o is only 55% against a random hand.