Thursday, 2 August 2012

An Ill advised WTF ICM hand

Hand played on Pokerstars
Seat 3 is the button

Seat 3:Villain ( $3302.00 USD )

Seat 4: Hero ( $5283.00 USD )

Seat 5: Player 3 ( $415.00 USD ) (Sitting out)

Hero posts small blind [$75.00 USD].

Player 3 posts big blind [$150.00 USD].

** Dealing down cards **

Dealt to Hero [  As 8h ]

Villain raises [$300.00 USD]

Hero ACTION????


Alright, it wasn’t long since my last post but I hope to post a few strategic hands each month to keep things interesting so here we go. 


So the BB/player 3 has been sitting out the tournament for a solid while now.  He is about to bust and the hero and villain are about to coast into the money once the BB goes through him another round – 5 more hands precisely until it’s guaranteed money for both of us, not bad eh?!


Well, whilst that’s what should be the case, it looks like the villain has other plans. Maybe he had a meeting and was running late?


So, once the villain opens to 400, against anyone not only with good knowledge on ICM but also observing the fact player 3 is sitting out, hero should be able to profitably push back 100% of cards since villains calling range should be non-existent or close to it. The presence of a cripple stack means it would be a disaster for villain to be in an all-in confrontation with hero because of ICM considerations – that’s to say that villain cannot make money as winning an all-in doesn’t burst the bubble and allow him to reach HU play, he can only lose money. Not only this, he is guaranteed a minimum of 2nd place in a few hands time, which in this $100 turbo is $195.82. Sure winning would see his equity increase in the tournament but does the equity gain justify being all-in with a blinding out cripple and therefore guaranteed equity in a few hands time? Fuck no…


Hero is dealt A-8o, and elects to shove back because of ICM considerations and the fact he is ahead of villain’s button opening range. Like a true champion, villain finds it in himself to “trap” the hero and raise-call his 22bb stack off with J-9s, WTF?!?


Remaining part of hand history:

Hero raises [$5208.00 USD]

Player 3folds

Villain calls [$3002.00 USD]

Hero wins $1981.00 USD

** Dealing Flop ** [ 2d, 5d, 8d ]

** Dealing Turn ** [ 8c ]

** Dealing River ** [ 2s ]

Hero shows [As, 8h ]

Villain shows [Js, 9s ]

Hero wins $6754.00 USD from main pot


Okay, so first off, doing this with AK would just be so terrible, so it’s going to be very costly with J-9s. The questions it raises, is how do we adapt in future, and what can we shove back against various opponents in future? Against a good player who knew ICM very well, you could shove back almost any two if they were raising wide although it’s a spot most good players will fold or ship anticipating being exploited unless you’ve no history, so this spot would be fairly rare. However, it wouldn’t necessarily be rare against a fish who can’t help himself from playing hands even when he shouldn’t be. Although some people will be sceptical about shoving back wide in this case if he’s not only raising wide but calling wide, this situation is an exception. The fact is, even if we lose we are going to have guaranteed equity in a few hands time because we are the big stack so the fact he is calling very wide is irrelevant – as long as we know he is opening wide enough - that's all that matters, whilst A-8o isn’t exactly wide and stands to be well ahead of villain’s range. It however would be a terrible proposition if we were the mid-stack since we’re risking our tournament life with soon guaranteed equity and we aren’t a big enough favourite to justify taking this risk, so folding to a raise in that case if the positions were the same would be the most +EV approach. Where you might shove 2-7o against a good player in these similar situations that arise on the bubble so long as you know it isn’t an induce, you certainly wouldn’t want to shove it back against a fish calling as wide as J-9s in this instance. When you observe such a hideous play, take a note on the opponent, colour code him, and adjust in the future and don’t abuse him on the bubble when you have the big stack and tighten up your shoving ranges against him.


Incidentally, assigning a hand range to a fish is a difficult thing to do since their ranges may be fluid, based on ego or what kind of mood they are in that particular day. It’s possible I won the previous couple of hands and this guy’s ego couldn’t cope with having to fold again and respect the situation even though that is the most +EV approach, and so he’s made his stand. He certainly isn’t thinking about his equity in the tournament anyway, that’s for sure.


In a couple of years of playing SNGs, this is one of the worst errors I’ve ever witnessed. His call is definitely a –EV proposition and a significant amount of equity gain goes to the cripple stack here since he has the chance at making money, but he isn’t even here. He literally could be having a nap yet the mid-stacks call is giving him a chance to make money which is exactly what happened – nice life. The question remains: how bad is it?


So, I thought I would plug the hand into an ICM calculator – SNG wizard, to really find out how costly this error was.


The wizard is unable to interpret situations from the original raiser’s perspective where there have been two actions pre-flop such as this one, so I’ve switched the positions of villain and player 3 around in this hand and given player 3 a fold on the button, so we can take a look at the hand from villains POV with hero shoving all-in blind versus blind against him. And as it is….Even if hero is shoving 100% of cards, calling with J-9s costs him 10.75% of the prize pool! That’s the equivalent to just over $60. Shoving 50% costs him 17.40% of the pool, whilst shoving 25% costs him 19.23% - need I go on? Mistakes that cost you anywhere from 0.5% to 1% of the pool are regarded pretty significant so you can grasp just how costly these numbers are. Of course this is from a blind on blind shoving perspective whereas in the initial hand he opened for a raise, so the fact is these figures might not quite be as costly with extra chips/equity in there, although given it was a min-raise and given what he has behind, the overall impact on these figures should be very negligible.  


There is an element of flaw however with these figures. The wizard isn’t able to account for the fact that the cripple stack in this situation is actually “sitting out”. This means as we’ve said already, he is guaranteed equity in a few hands time just by folding. In a standard 6max pay-out structure, 2nd place pays 35% of the total prize pool. Essentially, this means his mistake wasn’t actually any of those figures we arrived at earlier even though hero wouldn’t be shoving that wide anyway because villain opened the hand and villain is a fish, but more or less equivalent to him pissing away 35% of the pool which in this $100 buy-in is $195.83, and that’s not counting potential future money in the heads up. Even ignoring the fact the player is sitting out and the chance of free equity, calling with J-9s in the example assuming the player is sitting in against an 100% range will cost him 10.75% of the pool as we found out earlier, and only a calling range of Js through to As is recommended. 


It’s amazing. If you were walking down the street and seen an envelope with 195 bucks lying in it, would you pick it up? It doesn't seem this guy would…As PezRez once said, "In SNGs, ICM is king"


Live long and prosper