I would like some help on this hand. He is pushing whenever it is folded him on the BTN (and I am SB) or he is SB (and I am BB). He did it 6 times in a row, so he must be pushing with any two. What should I do here?

I'd call here.

You don't know he's really any two but he may well be very wide.

I think the computer will say it's close including ICM but the chance to be the big-stack on the bubble is worth extra. Also being a mid-stack getting abused by this guy is worth less than the computer says, so the decision is not that close.

I was in a similar situation yesterday on the bubble with a9 I called and got sucked out on by 62o . I think in this situation your second in chips with plenty of bb I would wait it out until I had a pair or a stronger ace to call with.

Assuming this is 9man turbo? top 3 paid?

If you are confident vil is always jamming any two here then A9o is a close-ish call but you have to be pretty certain for this to be profitable because we have 15bb and 2nd in chips. I would fold because there is no need to risk our stack at this stage, we are way too deep and we dont have enough stats to be confident of his range IMO.

the "computer" will say its close because it thinks villain is shoving his whole range...in reality the upper part of the deck is missing when he rips like this

its an easy call vs a lot of people

Don't underestimate lektar advice here. Being proactive in getting into profitable game positions is a big skill most non stt players struggle with. I would def take the risk here, cos you can boss the table from there if u win. And as pointed out, avoiding a spot where you will later have to fold A9 to his shove for icm reasons...had villain folded the last 6 times I would say fold!

Def a call. A9o here is in the top portion of his range. You are going to be dominating him a lot here

Am not calling here, by no means am saying this is the right decision.

At 18bb is he really pushing that wide, Id be expecting to see 99+ A 10 +, 55 - 45 favourite at best, more likely 65 - 35 dog.

Sorry missed first post, if he's consistently pushing would be a profitable call, but given how deep you are I'd like to wait it out.

LektorAJ and Colinb7 I've just found out about the Gigabite dilemma/block theory.

Would you say this is a good example of that? Or perhaps if we make it a bit more marginal, say A8o, A7o (don't know if those hands are low enough ... basically a hand that is -EV but not by a lot) or the like. Basically making a slightly -EV play in order to put yourself in a position that will allow you to be +EV later on?

Here if you call and win you escape from the current block of stacks and can (as Colinb7 pointed out) boss the table around, right?

Ya it's a call if your confident he's pushing atc. You'l question whether you should be waiting for a less marginal spot when you get it in and lose but this is the nature of sngs. If you don't make these calls you'l be run over in the end phase of a sng. Don't be hard on yourself if you make the correct decision and lose.

Gigabet is said to have gone bust around 2006 or 2008 as soon as other people could play even a bit. Grouping stacks into blocks is nonsense.

It may lead to some correct moves accidentally. The basic idea is that there is ICM (based on the formula) and RealICM (based on the real dollar values of the stacks), which are a bit different.

RealICM is not calculable - but if you are in a situation where the immediate ICM is making you play "bad" (i.e. heavily negative chip EV) to preserve your stack, then ICM is overvaluing your stack, so that means you should not play so bad to preserve it, which means it is worth more, which means you should preserve it and so on in an endless loop - the point at which those two forces are in balance is the realICM value of your stack.

So although we don't know how to calculate RealICM (we would have to be able to simulate the entire tournament to completion) with reasoning like the above we can get ideas of how to make ICM more correct. It all depends on a lot of factors though, even the relative seating of the stacks (if you are the 2nd stack you really don't want the 1st stack on your left, so you might be more willing to gamble to overtake it) and who has to pay blinds next.

Basically in this hand some people are saying that realICM would increase the value of the big stack and decrease the value of the mid stack so the apparent unfavourable odds caused by ICM are actually not that bad under RealICM.

Erm.... what he said!!