Thoughts on Bubble pushing

[Sabrazack]

I just thought of something that i have not considered whilst playing before. When you are in the small blind on the bubble, how much does the BB:s stack size affect your decision whether to push or fold. I'm assuming here you and the BB have atleast 5bb:s.

Imagine this situation

Blinds 100/200

BB 4000
SB (You) 1200
BTN 1400
CO 1400

I just realized (tell me if im wrong) that pushing into a big stack on the bubble is ALOT worse than pushing into a equally or slightly smaller/bigger sized one. Not just because the big stacks calling range might be alot wider since he feels he can afford it(this i already counted with) , but also for the following reasons.

1. If he folds, you get the blinds. But this does not really reduce the big stacks chance of making it ITM by alot. If you steal from a smaller stack it will reduce his shot of making it ITM more, and therefore further increase your chance.

2. If he calls and wins, you are out either way. Big or small.

3. If he calls and loses, the bubble is still on. You are now in great shape to make it ITM, but there is still no guarantee. If an equal stack calls you and you win you are now ITM with a good chance of getting atleast second.

Would like some of the SNG experts tell me if my thinking is correct or flawed. ICM takes this into account right?

[bigt439]

This is actually a very good post. While going through hh's the other day I realized this and played around with similair situations for about 20 minutes. It actually prompted me to start a word file to store rare hh's that teach me big concepts I did not know before. Your thinking is actually a very nice summary as to why this is true.

[DMACM]

This is a good question. ICM does say not to push into big stacks but I thought that this was because if you win the hand your not ITM, but if you lose you are not. Its been suggested that there are strategic advantages to having a big stack, and this is the only plausible reason Ive found for that.

[Freudian]

You are correct in a way.

But when you steal blinds it is better to push into one player than two or three, no matter what stacksizes are.

You can't really choose who sits to the left of you, so you have to work with what you have.

I think that the concept that you're stealing ITM% from a small stack more so than from a big stack is accurate. Though, I think it would probably distribute to yourself equally for both scenarios. If you understand my meaning.

OCW

It is very insightful, but Gigabet touched on this topic in his "theory of stack sizes" post, I can post a link in like 20 mins (don't ask [img]/images/graemlins/smirk.gif[/img]) unless someone wants to do it for me.

[Apathy]

[ QUOTE ]
Would like some of the SNG experts tell me if my thinking is correct or flawed. ICM takes this into account right?

[/ QUOTE ]

I'll assume you mean SNGPT when you say ICM, if you don't or are confused let me know.

Yes for the most part SNGPT takes this into account, but it doesnt really take into account the change in playing style you may see from doubling off of the bigstack here. Some people will become very tight once they get to a medium stack putting you in a great position to cripple everyone on the bubble. Others get angry and call everything figuring out those ranges and how often they will be pushing in hands you are not in and how THAT will affect future bubbel hands is your job.

DISCLAIMER: I don't own SNGPT or use it on a regular basis but I used to do these calcs by hand and I am familiar with the program.

[Sabrazack]

I really didn't understand much of Gigabet's post. All i got from it was that chips were blocks n stuff. But i think i know what you mean now that you mention it. And i think i might understand some of what he was saying if it has anything to do with this.

[Sabrazack]

SNGPT does ICM calculations. Correct?

[Apathy]

[ QUOTE ]
SNGPT does ICM calculations. Correct?

[/ QUOTE ]

Yes. But asking if there is something ICM doesn't consider is different from asking if there is something SNGPT doesn't consider.