All in question

[LordVTR]

Hi,

I have been playing low buy in ($6 and $11) NL HE SnGs for a few months now and have started to see an improvement in my results since reading the helpful posts on this forum.

Last night I was playing an $11 10 player SnG and was in the last 3 with the short stack with around 3500 chips. Blinds were at 300/600 I pushed from the button with 9 5o to steal the blinds, a tactic used successfully for the previous few orbits by all 3 players. I was called by the Big stack and when the cards were flipped he held 8 4o!!! He paired his 4 on the board and i was out, c'est la vie.

My question is, was this just a dumb call by him, or is there a more advanced theory that made that call for him a +EV? If so, would somebody explain it to me cos I was totally gobsmacked to be called by that hand!!! My thinking is he probably just clicked the call button by mistake and got lucky!!!

Regards

[givememyleg]

ALL YOU CANN EATYY CBABY

[taipan168]

My question is, was this just a dumb call by him, or is there a more advanced theory that made that call for him a +EV? If so, would somebody explain it to me cos I was totally gobsmacked to be called by that hand!!! My thinking is he probably just clicked the call button by mistake and got lucky!!!

Hi and welcome to FTR, it's good to have you here. Feel free to post in the SNG forum if you have questions about SNGs, I promise we don't bite!

It was a terrible call by opp to your standard push. The thing about poker is that sometimes you can make a terrible decision, get lucky and suck out. It might have been a misclick, or he might have gotten sick of you pushing and thought, f*ck it, let's gambolero!

[LordVTR]

Taipan,

Cheers for the reassurance!!! I wasnt overly bothered, as you say, sucky things happen in poker and at least I had made it to the money! Was bizarre cos the guy had played pretty solid the whole game which made me wonder if there was more to it!!

I have read a lot of the posts on the SnG forum and have found it of great benefit to my game, so really appreciate the help!

Regards

[losttrem]

How much bigger was his stack? Dan Harrington wrote about the 10-1 rule, which says that you can call an all in with any two cards when your chips are ten times as many as the raiser's chips.

[losttrem]

Oops.... just realised, that you wrote it was a 10 player SnG. So theres no way he has 35000 chips. Sorry.

[Xanadu]

Well, how bad was his call? Definitely a bad call, but not nearly as bad as you think. And not as bad as your push. His heart was in the right place, but he was really only a little too loose on the call if he assumed you were pushing any 2. If you are pushing any 2 (I'm assuming he's in the BB, otherwise it's a horrible call from the SB), he only needs 95o or better to break even ICM wise. With a more reasonable assumption that you were pushing top 70%, he needs T8o, 97s or better.

Having run through the #s, if you are pushing 70% (95 is right at the bottom of 70%), and he calls with 84o, he will lose .0125 of the prize fund on average. This is bad, but I bet not as bad as you thought. On the other hand, even if we assume SB always folds to simplify the calculations, If he calls with 70% (which is pretty close to correct if you are pushing 70%), you lose .02 of the prize fund on average. Almost twice as bad of a mistake! You will in fact lose more than this because SB will have a much tighter range to call you and you will be much worse off when he does. Once everyone has been stealing on the button without resistance for a while, it becomes pretty obvious that everyone is just about pushing any 2. The longer this goes on, the more someone will just decide to make a stand, and you don't want to be holding 95o when that happens. Although they will likely have crap, it will usually be better crap than yours.

[LordVTR]

Xanadu wrote:

Well, how bad was his call? Definitely a bad call, but not nearly as bad as you think. And not as bad as your push. His heart was in the right place, but he was really only a little too loose on the call if he assumed you were pushing any 2.

So I guess what you are saying is that I was overplaying the push on any 2 option? With the blinds this big tho, and an impending rise in blinds, did I have any alternative but to face the prospect of getting even more short stacked and therefore having even less leverage when pushing? any advice or comments would be appreciated as I'm still struggling to get my head round the theory here!!

[Renton]

depending on the pot odds he was getting (i didn't bother calculating them because im really tired), it can be the right call with 84o. If the money he's posting in the blind is a certain percentage of your push raise, then he's obliged to call with any hand.

[villagenut]

Well, how bad was his call? Definitely a bad call, but not nearly as bad as you think. And not as bad as your push. His heart was in the right place, but he was really only a little too loose on the call if he assumed you were pushing any 2. If you are pushing any 2 (I'm assuming he's in the BB, otherwise it's a horrible call from the SB), he only needs 95o or better to break even ICM wise. With a more reasonable assumption that you were pushing top 70%, he needs T8o, 97s or better.

Having run through the #s, if you are pushing 70% (95 is right at the bottom of 70%), and he calls with 84o, he will lose .0125 of the prize fund on average. This is bad, but I bet not as bad as you thought. On the other hand, even if we assume SB always folds to simplify the calculations, If he calls with 70% (which is pretty close to correct if you are pushing 70%), you lose .02 of the prize fund on average. Almost twice as bad of a mistake! You will in fact lose more than this because SB will have a much tighter range to call you and you will be much worse off when he does. Once everyone has been stealing on the button without resistance for a while, it becomes pretty obvious that everyone is just about pushing any 2. The longer this goes on, the more someone will just decide to make a stand, and you don't want to be holding 95o when that happens. Although they will likely have crap, it will usually be better crap than yours.

I would like to add a question here for Xanadu or any other math wiz. I am still trying to work my way through all the math of Poker. You said that someone "needs 95o to break even ICM wise" can you explain this a little further for a relative newbie. Also, you refer to top 70%. Are you talking about the top 70% of ranked hands?

[Xanadu]

Here's an article which shows optimal short stack play when HU:

http://www.cardplayer.com/magazine/article/15250

You can't use it 3-handed, but if you look at the chart, your 95o wasn't quite good enough to push HU, so it is certainly not good enough 3handed.

ICM is Independent Chip Model. It is a fairly accurate model of the expected value of prize money for each stack size in a tournament assuming equal skill among players. What it does is assume that each player's chance of placing first is the same as their proportion of the total chips in the tournament. The rest of the places are determined in the same way and the chance of each place is multiplied by the prize for that place and they are all added up giving the EV in prize money.

Here's an example of how it works if you are interested:

Suppose 3 players are left in a SNG with 50-30-20 payout and their stacks are 5000, 3000, and 2000. Here's how to get the expected value of the big stack:

He finishes 1st 50% of the time because he has 50% of the chips.
When the middle stack finishes first (30% of the time), he will finish 2nd 5/7 of the time (his proportion of the chips remaining between him and the short stack) and will finish 3rd 2/7 of the time.
When the short stack wins (20%), he will finish 2nd 5/8 of the time and 3rd 3/8.

The chances of each place are:
1st: 50%
2nd: (30% * 5/7) + (20% * 5/8) = 34%
3rd: since he must finish third when not 1st or 2nd, we don't have to do as much math ... it's all that's left: 16%

Now each of these are multiplied by the prize and added:

(50% * .5) + (34% * .3) + (16% * .2) = .384

So the big stack expects to win 38.4% of the prize fund

But you don't have to do any math, that was just so you know what's going on if you use ICM. Fortunately, you can just go this webpage and plug it all in:

http://sharnett.bol.ucla.edu/ICM/ICM.html

Now the big question is 'why is this useful?'

Well, it's useful because each chip you gain in a tournament actually has a lower real money value than the one before. Straight pot odds calculations are correct for money games, but are not correct for tournaments. This is because you have a theoretical endless supply of chips in a money game (because if you bust, you can rebuy), but if you lose all your chips, you are out of a tournament.

There is a 'how to use' link on the ICM calculator. This shows you how to use it to make decisions. This is not a tool for use while playing, but an analysis tool for checking your play and studying up on how to play correctly in tournament situations. I used the ICM calculator and pokerstove to get the results I posted earlier in this thread. Basically you want to estimate your winning chances (I use pokerstove by putting in the range of hands I think my opponent may have against my hand), and then do the ICM calculations for 3 cases: when you fold, when you call and win, and when you call and lose. If it is you pushing, you need to estimate the chances your opponent will fold, and then the 2 cases where he calls and you either win or lose. Multiply the EV from the ICM calculator for each case by the chance of it happening and compare it to what happens if you just fold. If folding has a lower EV, pushing/calling is the better play.

And yes, top 70% is top 70% of hands ranked by HU value.

[b-rabbit]

Here's an article which shows optimal short stack play when HU:

http://www.cardplayer.com/magazine/article/15250

You can't use it 3-handed, but if you look at the chart, your 95o wasn't quite good enough to push HU, so it is certainly not good enough 3handed.

ICM is Independent Chip Model. It is a fairly accurate model of the expected value of prize money for each stack size in a tournament assuming equal skill among players. What it does is assume that each player's chance of placing first is the same as their proportion of the total chips in the tournament. The rest of the places are determined in the same way and the chance of each place is multiplied by the prize for that place and they are all added up giving the EV in prize money.

Here's an example of how it works if you are interested:

Suppose 3 players are left in a SNG with 50-30-20 payout and their stacks are 5000, 3000, and 2000. Here's how to get the expected value of the big stack:

He finishes 1st 50% of the time because he has 50% of the chips.
When the middle stack finishes first (30% of the time), he will finish 2nd 5/7 of the time (his proportion of the chips remaining between him and the short stack) and will finish 3rd 2/7 of the time.
When the short stack wins (20%), he will finish 2nd 5/8 of the time and 3rd 3/8.

The chances of each place are:
1st: 50%
2nd: (30% * 5/7) + (20% * 5/8) = 34%
3rd: since he must finish third when not 1st or 2nd, we don't have to do as much math ... it's all that's left: 16%

Now each of these are multiplied by the prize and added:

(50% * .5) + (34% * .3) + (16% * .2) = .384

So the big stack expects to win 38.4% of the prize fund

But you don't have to do any math, that was just so you know what's going on if you use ICM. Fortunately, you can just go this webpage and plug it all in:

http://sharnett.bol.ucla.edu/ICM/ICM.html

Now the big question is 'why is this useful?'

Well, it's useful because each chip you gain in a tournament actually has a lower real money value than the one before. Straight pot odds calculations are correct for money games, but are not correct for tournaments. This is because you have a theoretical endless supply of chips in a money game (because if you bust, you can rebuy), but if you lose all your chips, you are out of a tournament.

There is a 'how to use' link on the ICM calculator. This shows you how to use it to make decisions. This is not a tool for use while playing, but an analysis tool for checking your play and studying up on how to play correctly in tournament situations. I used the ICM calculator and pokerstove to get the results I posted earlier in this thread. Basically you want to estimate your winning chances (I use pokerstove by putting in the range of hands I think my opponent may have against my hand), and then do the ICM calculations for 3 cases: when you fold, when you call and win, and when you call and lose. If it is you pushing, you need to estimate the chances your opponent will fold, and then the 2 cases where he calls and you either win or lose. Multiply the EV from the ICM calculator for each case by the chance of it happening and compare it to what happens if you just fold. If folding has a lower EV, pushing/calling is the better play.

And yes, top 70% is top 70% of hands ranked by HU value.

LONGEST POST EVER!

[Xanadu]

LONGEST POST EVER!

If it was true before you posted, it wasn't after.

And I'm pretty sure I've done better

[LordVTR]

Wow!!! This theory goes so deep!!! Xanadu, many thanks for taking the time to explain this. I won't pretend I fully understand it, but it does show that in a tourney/SnG environment, stack size does affect the decisions I should or shouldn't make, very insightful post!

So bottom line is, I need to be a little more selective about the frequency that I push any 2 when up against a player who probably understands the math in these situation!