PDA

View Full Version : So I've had a few beers, but I want to start a thread.


citanul
04-15-2005, 12:26 AM
So I've been thinking about this problem lately, and before I start letting it slip out of my head, I wanted to get something "on paper" as it were. This is mostly inspired by the raptor hand with AJs, which drove me to think more and more about "proper" bubble play. For a brief lesson in proper bubble play for anyone who is lacking in high level bubble play, I quote donbuttons "just push every time it's folded to you." I've joked before about not being sure if he's serious about this, because well, it's not entirely far from what you should be doing, if you look at an equity tool, such as eastbay's. For another general synopsis: push, don't call.

So, here we are at my question, or whatever.

In proper bubble play, it is correct to push very often, but to call almost never, even if you know your opponent's hand, and you know you are ahead. I believe this to be most true at the highest levels, where know not to call who are not necessarilly 2+2ers, perhaps not out of reasoning directly tied to $ev as such, but out of quite justified fear of mutuallly assured destruction. ie, even if people haven't necessarilly looked at icm numbers for hours on end, it's possible for them to have come to the conclusion that just calling with the best hand ain't where it's at.

However, even very good opponents (or at least very name recognizable opponents) don't play this well. There are many "top" sng players who call in spots where they "shouldn't." (I don't particularly know why I put that in quotes, since it doesn't necessarilly make sense.) So the dynamic clearly has to change when you're playing against these players, those who call too much, even if they just call a bit too much, or perhaps it changes in different ways when they call too much, depending on what those ranges that they call extra with are.

My gut feeling is that if people call too much, you have to tighten up when you push at/through them. Clearly if they somehow manage to call too little for what is the proper amount (hah, like someone could do that) you push at them even more). But here's the real question: say everyone at the table loosens up their calling standards by some amount. I think that that dictates that you as well open up your calling standards. I could be full of [censored], I don't know. But I think that if they're going to start to call your pushes which are coming fast and furious often with any two cards with say, any ace and K9+, any pair, you have to also open up your calling standards (perhaps not as much as theirs).

I don't really know where the question comes in here really, but I thought this was interesting, that I think that you need to clearly open up your calling as well, so I thought I'd put it out there for discussion. I like this stuff, and I know there are others out there too. I didn't pay much attention to the recent strassa post called "playing the bubble perfectly" or whatever, and it's possible that it's a very similar post, but much better. (it probably is, knowing my luck.) So I apologize if it is so and this has been recently beaten to death.

Anyways, have at thee, what do you all think?

citanul

Maulik
04-15-2005, 12:36 AM
This post is right on the money. You'r pretty much suggesting, play tight when people are loose and loose when they're tight. This is especially true during bubble play. Just analyze the number of pushes and if they are pushing entirely too much start calling with some of your weaker calling hands. Perhaps some that you may call when you're HU.

Bottom line: You don't call because you want to steal their blinds, that's better than having to let www.partypoker.com (http://www.partypoker.com) deal a coinflip out for the worst hand pre-flop.

raptor517
04-15-2005, 12:50 AM
good post cit. i think you made quite a few very good points here. (possibly too many /images/graemlins/wink.gif). hmm, i think, in certain situations, you do have to loosen up quite a bit with yer calling requirements. i really dont like to, but it could be necessary against some very agressive opponents that do in fact call a lot. you have to learn how to adjust to situations like this when you cant get FE out of your crap pushes. who knows, against super loose opponents you would have to call with that AJ hand, assuming they call every push every time you shove with anything. that would make any player's edge dissappear quick. holla

citanul
04-15-2005, 01:17 AM
note that in case this thread gets interesting, i deceptively named it something stupid so that no one will know where to go looking for it.

citanul

Blarg
04-15-2005, 01:30 AM
I agree that the bubble is especially not a place to do a lot of calling. That's just a way to bleed off chips when your real goal is to accumulate some quickly if you can, hopefully with a great hand of course, but probably primarily through stealing. You may only get one or two chances to steal, so you have to be very focused on it. You have to use your folding equity while you can still have some chips to intimidate players with; doing much calling is just taking the point off your spear right when you need to be poking people. Money lost calling is money you can't push people out of pots with.

Madd
04-15-2005, 07:18 AM
If the other players have loose calling standards and push often, then let them bust each other and slide into the money without a fight. Their calling standards don't change that a call with AJ is -$EV.

Voltron87
04-15-2005, 10:23 AM
I tried to add this to my favorites but instead alerted the mods about you. Ooops. Too early to read it though. I'll play Duke Nukem for 30 minutes and read it during my 11 o'clock class.

EdgePort
04-15-2005, 10:28 AM
This is something that I have really realized over the past few days and I think has/will improve my ITM. Too manys times I was calling an all-in when I had a good hand AJ, type of thing, when I should just let it go. I would know I actually had the best hand, but it is just not worth giving up half my stack on the bubble when I can use it in better ways.

Phil Van Sexton
04-15-2005, 10:39 AM
[ QUOTE ]
However, even very good opponents (or at least very name recognizable opponents) don't play this well. There are many "top" sng players who call in spots where they "shouldn't." (I don't particularly know why I put that in quotes, since it doesn't necessarilly make sense.) So the dynamic clearly has to change when you're playing against these players, those who call too much, even if they just call a bit too much, or perhaps it changes in different ways when they call too much, depending on what those ranges that they call extra with are.

[/ QUOTE ]

My guess is that these top players don't make these bad calls as often as you think. They are counting on the fact that you aren't going to forget the one time he spite called with with J9. You'll never know about the 10 times he folded AT.



The best image for a player to have on the bubble:
- Will call a steal with any reasonable hand.
- Only steals with good hands.

The ideal strategy on the bubble:
- Don't call without a great hand.
- Steal with any 2 cards.

As in any form of poker, you want your table image to be the opposite of how you actually want to play.

How do you achieve this? If you play in 200s where everyone knows you, you can throw in some crazy calls against a tough opponent who you face often and won't forget.

If you know your opponent is stealing with any 2, then calling with J9o is -$EV, but it isn't horrible. It's probably a coin flip vs a random hand, so maybe you give up 2% according to ICM, so $20 in a 200/15 sng. It may be worth $20 to get tagged as a lunatic.

You probably heard second-hand about these crazy plays from name players in games you weren't even in. That's effective advertising.

If you are Gus Hansen, you can call an allin with T8o on a televised event. He did that in an event with a $25k first prize. I think he had like 30% of the chips in play at the time, so his $EV if he folded was $7500. If he called, it was maybe $5000. That's a huge percentage drop in $EV, but it really only cost him $2500.

Since Gus plays in 4000/8000 mixed games, I doubt he'll miss the $2500. It was worth it to him to cement his image as a lunatic that will call you with T8 for a ton of chips. (If you missed the episode, a stunned Antonio flipped over 77 and Gus flopped an 8).

In the lower limits, it's pointless to make -$EV calls because no one is going to remember, and you don't face the same players every day. However, I think it is worth it to make borderline +$EV calls for the same reasons.

I think Scuba had a case where he had 2100 and needed to call a 330 allin with T9 on a J8x flop. It was close, but it was +$EV to call. There were arguments made for passing here, but none of them were convincing to me.

You have a chance to call with Ten-high, it's +$EV, and it won't cost you that much.....I am all over it.

To my opponents, this call might look just as crazy as the Gus Hansen call. In fact, I've improved my table image AND made a small increase in $EV.

If you say, "I'll just fold, and then steal the next 2 hands." That's fine, but now you are killing your table image. Your opponents see you fold for only 330 more (though they don't know you cards). They will then see you stealing more than your share.

Eventually, your strategy becomes transparent. This is bad because a "don't call, steal with any 2" strategy is exploitable. If your opponents call a lot, you can just wait from them kill each other, but if they only call a lot when YOU bet....that's really, really bad.

1C5
04-15-2005, 10:41 AM
2 questions for you and anyone else.

First. How does this change when you are on the bubble and you have 3-4X BB or less. Say someone is aggressive and pushing lots to steal the blinds. Blinds are 150/300 and you have 800 chips.
As I said, this big stack is aggressive and bullying the table (as he should).
He is probably pushing with almost any 2.

At what point do you have to make a stand? What pocket pair do you call with? AQ? AK? AJ?
This is something I am not sure about just yet?


Also keep in mind what level you play. People in the $11s will call with any A and any 2 face cards all the time. They don't know what the gap concept is.

Irieguy
04-15-2005, 11:35 AM
This is one of the best bubble strategy/theory posts I've ever seen.

It all seems simple and perhaps obvious... but I've never seen it so clearly in print before.

Thank you, Phil. Now shut up.

Irieguy

citanul
04-15-2005, 11:59 AM
Phil, Thanks for your post.

First off, I wanted to say that I'm not talking about like, Gus Hansen, I'm talking about bean_mo, gambaholic, those types of "name players." So instead of hearsay and from television, I'm talking about real players with good results that I play with more frequently than I really should.

Next wanted to share some math that I did. I haven't done an ICM in a while, so yeah, it's possible there's some screwups.

even stacks, 4 handed

everyone starts with equity of:

1/4x1000=250

heads up in the blinds, i push with a random hand:

you are the bb, and call a push with J9o against a random hand:

you have 53.251% equity in the pot:

so 53.251% of the time you have 5000 chips 3 handed, for an equity of:

38.33%, or 383.3

when you lose, 46.749% of the time, you have 0 equity.

so your total equity from calling is 204.11

so to call with J9o here is giving up 45 bucks of equity. very bad. not very bad to do it once or twice as an advertizer, but very bad nonetheless.

and let's see, i should be able to get function that is dollars given up in terms of percentage likelihood you win the hand, right:

so you start with 250, and when you win x% of the time you wind up with 383.3, losing 1-x of the time, you wind up with 0.

so dollars given up is 250 - (383.3x) basically is the dollars that you give up.

250-383.3x=y

when y=0, x=65.22%, so to call correctly, assuming your opponents play the bubble well, you should believe you have an edge over your opponents of the 65/35 variety. Interestingly, when I run AJs with even stacks against a truly random hand, it runs just better than this (a marginal winner, at 65.393, which is a couple bucks.

so the only really interesting part that's semi-unique here:

dy/dx=383.3

and making that in terms of "percentages" that means that basically, dy/dx=3.833
so for every 1% you are wrong by, at the 200s, you give up 3.833 dollars.

doing a quick check to make sure that's correct, in the J9o example, this figure gives that the caller is wrong by $45.87, which is right! hot damn i like math sometimes.

however, it does mean that you can advertize "relatively cheaply" sometimes.

Pooh-bah biatches?

citanul

hansarnic
04-15-2005, 12:00 PM
Well I composed this before seeing that some of my thunder has already been stolen by other replies. But what the hell, I'll post anyway....


Yes, obviously you have to adjust to different game conditions and ICM only tells you what to do if everyone follows the strategy it suggests. But how should you adjust?

Let’s create some scenarios to see how ‘non-optimum’ play should affect your call / push standards.

Let’s make the game 4-handed with equal chips and standard SnG prize money. Let’s also say the blinds are tiny, but each players actions can only be push /or call all-in. Let’s also say my opponents all know my strategy in advance.

In the first scenario let’s say my strategy the moment we’re down to 4 is to fold every single hand. Literally.

What should my opponents do? Clearly pushing against my big blind is correct and clearly if one of them does so every round my EV is basically zero.

So how does this change the game? All my opponents share out my equity value. If we are all equal in chips, they all have a third of the prize pool each and I have nothing.

So when UTG pushes, what kind of hand does one of my opponents need to have to call? Well, a bigger one than normal obviously.

He’d actually need JJ-AA (using ICM but with the assumed equities modified to take account of the fact that I’m worth nothing).

But if I wasn’t playing this idiotic strategy and was ‘equally skilled’ to my three opponents then we are back to normal ICM and he could add AK, AQ&Js and pairs 77-TT to his calling range (he’d need to be a 65% favourite).

So when someone is in the game who is very tight (both calling & pushing) on the bubble, then it is correct to push more (if that’s possible) and call even less.

What happens if my strategy is to call all-in / push every single hand when we get to 4 instead? Now what should my opponents strategy be? Well, in order to decide their strategy they first have to decide how effective they think my strategy is. This lets them estimate our respective EVs and make push / call decisions accordingly. Well, my equity is certainly not zero, but it surely must be lower than a quarter as I have to push and call with any two, right?

So let’s say they estimate my share of the pool at 19% playing my ‘any two’ strategy, rather than 25%. So my opponents have decided they are actually worth 27% of the prize pool. So if I’m in the BB the SB needs a pair of 9s or better to push with and no unpaired cards will do (He must be 70% favourite). So clearly my opponents need to be uber-tight when pushing.

Hmmm. That sounds like it plays to my advantage. They need 9s or better to push??? Looks like there won’t be a lot of pushing into me. What do they need to call? Well, obviously it’s the same edge as they need to push, so 9s or better.

So basically they can’t push and they can’t call. I’m liking my chances more and more. Doesn’t sound like I’m really only worth 19% of the prize pool if they play this way! Sounds like I’m actually the favourite!!

So by assuming my strategy is bad, they actually make it good.

So let’s have them assume my strategy is actually quite good and make me worth 31% of the prize pool & see if that looks more realistic. That means the other 3 are only worth 23% each. Now what do my opponents need to push / call with? A lot less. Pair of 6s or greater and A9+ KJ+ (bit more if suited).

So if they pick up, say, a top 15% hand they should push / call otherwise they have to get out of my way, in which case I get the blinds. Except that UTG has to be even a little tighter than that as the other two ‘normal’ opponents are still to act. So I probably pick up the blinds about 2 hands in 3, although when someone pushes / calls I’m likely to be about 65% dog for all my chips.

So by giving my strategy a little more ‘respect’, they can reduce its effectiveness somewhat but they still can only push about 1 hand in 8. Not sure whether this is the optimum push / call standard, but I can’t think they can get too much looser than this.

So if the blinds are high I like my any two strategy very much (assuming my opponents know what my strategy is). By playing ‘badly’ I am forcing them to back off. I am essentially making them tighter than they would like to be. And we all know folding too much with high blinds is not good. But if I only play that way & they know it, I really don’t see what else they can do. [note, obviously given high blinds the above push / call standards need to be modified accordingly, but the comparative adjustments that should be made vs. push / call standards where everyone plays ‘correctly’ should be about right].

But it gets even better for me. The reality is that my opponents won’t all play the same way either. If they know the other ‘normal’ players will gamble with a top 15% hand, they know a confrontation is coming pretty quickly. So why would they want to be the one to gamble with me? They’ll let someone else take the risk and make it to the money themselves before opening up. So the net effect may actually be to tighten everyone up as they all wait for someone else to make a stand.

So in summary:

If they are playing too tight you should be tighter when calling, but looser when pushing. If they are calling too much you need to tighten up your pushing standards a lot, and loosen up your calling standards a bit. But broadly speaking you are losing value on the bubble if someone plays this way.

Sound like you should play the ‘any two’ style? Only if your opponents know that’s your style and you won’t change it. But I think that’s why certain high-profile tournament players go out of their way to ensure that everyone knows that they'll gamble.

Phil Van Sexton
04-15-2005, 12:32 PM
[ QUOTE ]
First off, I wanted to say that I'm not talking about like, Gus Hansen, I'm talking about bean_mo, gambaholic, those types of "name players." So instead of hearsay and from television, I'm talking about real players with good results that I play with more frequently than I really should.

[/ QUOTE ]

Good point. I was just making examples that people could relate to. I haven't played with these guys, so I'm just speculating. I doubt anyone has 1000 examples of spite calls from any of these guys. Small sample + selective memory = table image that might not be accurate.

I think you did the math for a 100/9, not a 200/15. You were using prize pool=1000. Therefore, you are really saying it costs $90 at the 200s.

Anyway....I agree that calling with J9o gives an $EV of $205 at a 100/9, but you should also compute what your $EV would be if you folded.

You are BB. Let's say BB=400. If you fold, your ICM is 0.223, so $EV is $223. Calling with J9o costs you $18, not $45 in this case.

Converted to a 200, it's $36. This is what I was thinking when I guessed $20.

If the push came from the button and the SB folded, now your $EV is $207. So the advertisement costs $32 at the 200s.

I'm sure we can change the scenario more and get it below $20, but I think we get the idea. I don't think Irie wants me to compute the optimal time to make a spite call.

citanul
04-15-2005, 12:45 PM
Good point, I had forgotten to factor in that the other option is -ev too, so that you only lose the extra lost ev when you make your spite call. Also good point about the fact that i was calculating it from the 100s, not 200s. Der, no excuse there, what can I say, I heart Beer, or something.

Bah, not even a comment about the derivative? boo. that was the best part.

And yeah, I think no one wants you to be much more helpful than you already have been.

citanul

barry111
04-15-2005, 01:54 PM
Great post and great comments, and congrats on your new Pooh-Bah status have a couple more beers and git in thar and gambool /images/graemlins/grin.gif

Apathy
04-15-2005, 02:21 PM
I am going to respond to this thread while it is at the top of the board because I really don't want to see it around much longer.

I agree with the general comments in your OP, most of the replies have dealt with the more meta game issues of calling even if you lose a slight amount of EV so that people won't push on you as much. Of course this only matters at the highest levels and is quite important to do from time to time as it will make you money in the long run, mostly in the form of BB walkovers.


I would like to talk about something in you OP that I think you are a little bit off with. You suggest that if the 3 others on the bubble are all calling looser, you have to push tighter (of course) and call looser. Leaving the meta-game issue aside for a moment, this is not correct, assuming the table is typically aggressive. You will no doubt be correct in practically folding your way into the money, only making pushes with a very tight range of hands.

The time your example REALLY comes into play is a situation I have experienced before, which is the following:

Hero is CL with 7 /images/graemlins/heart.gif5 /images/graemlins/club.gif on the button and pushes into two medium stacks. The BB calls with KK and cripples the hero.

The hero gets lucky and doubles, as does the other shorty, the table is pretty equal now.

This table is also full of morons who dont understand icm and have no clue why i pushed garbage, they just think ill push anything. They would never consider folding the best hand PF against anyone.

All 3 opponents (or at least two of them as the case may be) are no calling me down with anything, but remaining tight against each other.

This creates a kind of collusion, even though its not collusion, between the other three players. In this circumstance, which is more common then I'm making it sound, it is CORRECT to call looser, not a lot looser, but looser then ICM suggests.

It is one of the MANY MANY applictations of the concept (in TPFAP) of taking a bad bet to avoid a worse one later that applies in SNGs.

No noone respond to this thread and hopefully it will die away....

zaphod
04-15-2005, 06:05 PM
[ QUOTE ]
No noone respond to this thread and hopefully it will die away....

[/ QUOTE ]

bump

(sorry could not resist)

Degen
04-15-2005, 08:16 PM
"a call with AJ is -$EV"

Why is that? When its 4 handed and the chips are flying I think AJ is a hand that will often dominate a lot of hands that are pushed in. Is that wrong?

What would you guys reccomend as the calling hands when blinds are say 200-400 and there are 3-5 players left.


AA, KK, QQ, AKs I think go without saying.

What about AKo, AQs, JJ, TT, AQo? I am calling with these hands most of the times if i'm in the blinds. Is this wrong?


Degen

Degen
04-15-2005, 08:17 PM
how do you add to favorites? I've always wanted to mark good threads....

Degen

citanul
04-15-2005, 08:45 PM
for an in depth discussion of AJ for instance, go look at the raptor post from a couple days ago. in general, icm or eastbay's tools help look at these situations to come up with answers like "AJ is a fold here." do not post in this thread to ask either how to get eastbay's tool or what icm is and how to use it. i swear i'll track you down if you do. AA-JJ, AK is basically all that's an acceptable calling range, i think.

to add a post to your favorites click the icon that is in the middle below a post's text.

citanul

Degen
04-15-2005, 08:54 PM
i know about ICM and eastbay's tool...thx for the referral to raptors thread

great post/topic btw

Degen