snap fold.
if he's half decent thinking player you can min-raise and fold to a 4bet.
this isn't a easy shove because we have the bubble factor to deal with. we're shoving into the biggest stack at the table, for very little gain in terms of
equity. also, this guy is only raising 16% of hands. i know that figure is obviously skewed as this is later in the tournament, without reads in game this is impossible to guess. he might be raising everytime on the sb, which is what a lot of good players do with a big stack. since we can't define a range without a lot more info from the op, i'll leave it alone. Before you say it, his range does affect things, it affects the icm i'm about to go through. i'd say for this to be a shove he would have to be raising wider than 80% and calling less than 10% of hands, even then it would be iffy as we are gaining nothing really in terms of equity. our hand is strong, and also likely ahead of most villians range here, but our hand value shrinks as the bubble factor gets higher, so we need a much stonger hand when we're taking a risk.
in these situations on the bubble, the medium stack is always the one who needs to be the most risk averse. you can group the chip stacks up, its called 'stack grouping' and it's used to get a better idea when it comes bubble time. in this situation, we have one chip leader who has many more chips than anyone else. he is group 1, he has no stack rival, and can be very aggressive because of that. we are the middle stack here, we have more than the other two players, but not as much as the cl, we have to be very risk averse vs the cl as he is the only one who can knock us out, but we also have to be wary of the others stacks, as one lost pot will send us to the bottom of the chip count, so that makes us group 2. group 3 as you might work out is the two shortest stacks, they have equal stacks, they can be looser than group 2, as they need the chips to survive but they will have to be risk averse vs the bigger stack where possible.
just to do one more example of stack grouping. lets say stacks are 5000, 3000, 2000,500. the 5k stack doesnt care about anyone. the 3k stack is wary of everyone. the 2k stack is also quite wary of everyone, the 500 stack doesnt give a shit. So you see how the middle stacks are kind of forced sometimes to wait it out. this example here is kind of clear cut as there is a very short stack in play, but hopefully you get the idea of how this works.
one last thing i'd like to add, it has to do with table dynamics. there are times when we need to 3bet here or make shoves that could even possibly be -$ev. we make that -$ev shove because it will lead to better situations for the future of the tournament. Lets look at this example, lets say we know the SB chip leader, he is a good player who is very aggressive late in the game from all postions and likes to use his stack to bully shorter stacks, he especially likes to bully us, because he knows we are thinking about the game and he knows we need to be risk averse at the table. the other two players are really bad, and make loose calls because they do not know better.
lets sit back and look at the table dynamic here, this is really hard for me to explain without a diagram, so i'm going to post a very artistic ms paint diagram for you. the table dynamic is a bad one for us in the future context of this game. we will have two loose cannons who we cannot shove really wide on later when the blinds increase, they will call too much, forcing us to tighten our shoving ranges. basically we're not going to be gaining chips from them as much because they are loose. we also have a good player, who will not give up shoving and raising us from the sb, and also likely the button. so we're never getting a walk from the blinds unless we change something.
this is obviously an easy example, but table dynamics play a huge role in stt. in that above situation, we can then go against our teachings of icm, happily take a -ev situation, and shove on him, to improove our chip stack situation, but also to show that we are not going to take his shit all tournament long. we can also 3bet/shove him when he adjusts to our bvb aggression by picking off his button raises.
just to give some cold hard math here. have to do this maually since i dont have any of my old graphs or my version of sng wiz anymore. but basically i'm using an online icm calculator to see how much we're actually gaining here, not in terms of chips, but in terms of equity. equity is what we're concerned with in sngs until we reach heads up, once we reach heads up we dont care anymore.
numbers here are what their stack is worth, compared to the prize pool, which adds up to only 50% because thats all you can win....
co - 18.933%
button- 19%
sb-34.190%
Mr tangerine-27.877%
ok so thats before the hand played above.
if we shove here, and get away with it barring anything else happening (all of which sng wizards algorithm deals with)
cutoff and button stay the same
sb-32.548%
mr tangerine-29.712%
so a 2% gain to our stack, but more importantly, a 2% loss to villians stack. this actually worked out to be more than i would have thought. it's still not enough to risk shoving our whole stack over though. unless we know he will raise wide and fold lots, then just fold.
hope this rambling post helps. if you want me to run it through SNG wizard i can, but it will take a few days to get the developer to send me through a code.