Edge needed to go all in first hand?

[Beck]

Preflop, your ICm value is 0,1 of course.
If you double up, and take another player out, your ICM is now 0.1844.

So is it correct to assume that to make a profitable all in fist hand, you only need an edge of 0.1844X=0.1 --> X=0.5423 or 54.23% edge??

Seems like you only need a rather small edge to make it worth risking the whole tourney, so please enlighten me, where did I go wrong?

[Freudian]

I think ICM calculations lose quite a bit of their value when you do them in hand 1. There are simply too many hands from there to when its threehanded, even with a healthy stack.

[david050173]

Tournament Poker for advanced players has a discussion of this. You don't want to take gambles when you have a small edge if they will prevent you from taking a gamble when you have a bigger edge later.

[Slim Pickens]

I looked at this a little, but never really got a good answer.

How big of an edge do you need to call? (http://forumserver.twoplustwo.com/showthreaded.php?Cat=&Number=1917881&page=&view=&s b=5&o=&vc=1)

Oh yeah, long story short, my guess is 62.5%.

Slim

[Bigwig]

[ QUOTE ]
Preflop, your ICm value is 0,1 of course.
If you double up, and take another player out, your ICM is now 0.1844.

So is it correct to assume that to make a profitable all in fist hand, you only need an edge of 0.1844X=0.1 --> X=0.5423 or 54.23% edge??

Seems like you only need a rather small edge to make it worth risking the whole tourney, so please enlighten me, where did I go wrong?

[/ QUOTE ]

One thing to remember about ICM is that it doesn't take into account how much better you are than the average player in your game, and how likely it is that you'll get a better situation later. That's probably impossible to quantify mathematically. Generally speaking, I need a pretty big edge to risk the whole tournament on the first hand. K's, or possibly A's preflop. On the flop is another story.

[AleoMagus]

Bigwig is correct here so I'm just seconding what he has already said.

ICM is just a mathematical model of equally skilled players. Ideally, if we are playing in the first place, it is often safe to assume that we are actually much better than the competition. If this is the case, we can pass up bigger edges than the ICM suggests because we can count on better opportunities coming our way later.

That said, I don't think even for expert players, it is all that much higher than the ICM suggests. I remember a discussion on here once where some suggested that a pro might fold AA if pushed against on the first hand of the WSOP. That is definitely taking this line of thought way too far.

In a SNG, I'd take even a small edge on the first hand, just because I know what it means to my hourly rate, and I know what a big stack means to me. How you play a big stack matters also. If a big stack increases your $EV more than the math indicates because that is your style of game, then you might be inclined to try to build one with some early gambles.

As I get closer to the bubble I will pass on bigger and bigger edges if it means all my chips.

I know that early on, we tend to play very tight in SNGs and don't emulate proper ring play, but this is NOT because of ICM considerations. ICM considerations matter more as we near the bubble. The reason why we play different early is because of skill inequities and just the kind of game we excel at.

Regards
Brad S

[viennagreen]

Aleo,

I find some of your statements to be contradictory.

"As I get closer to the bubble I will pass on bigger and bigger edges if it means all my chips."

"In a SNG, I'd take even a small edge on the first hand"

and

"ICM considerations matter more as we near the bubble."

It seems that you play opposite to the ICM?

Doesn't it make more sense to pass on small edges early, and capitalize on them late?

I think that part of the reason why you do/say this is because of the amount of time you have committed--- as a SNG progresses, you have more invested.... but isn't the whole point of surviving to the point of "near the bubble" is so that you can take advantage of small mathematical edges?

[Nottom]

If I have TT and my opponent went all-in while flashing AKs, I'd call. I don't know if its correct, but I think my TimeEV makes up for any EV lost in that tourney.

The problem with preflop coin flips is half the time you are thinking about gambling (maybe becasue you already have a raise in the pot) you aren't really in a coinflip.

[BradleyT]

If you're 4 handed with equal stacks and always take a coinflip how often will you make it into the money?

Mathematically when you're 4 handed, how often should you make it into the money?

[Freudian]

[ QUOTE ]
If I have TT and my opponent went all-in while flashing AKs, I'd call. I don't know if its correct, but I think my TimeEV makes up for any EV lost in that tourney.

The problem with preflop coin flips is half the time you are thinking about gambling (maybe becasue you already have a raise in the pot) you aren't really in a coinflip.

[/ QUOTE ]

If you had the choice to play a ninehanded 10+1 with 1600 chips but you had to pay $21 to enter, would you prefer to do so over paying $11 for 800 chips.

Payouts are 50-30-20

[XChamp]

[ QUOTE ]
Aleo,

I find some of your statements to be contradictory.

"As I get closer to the bubble I will pass on bigger and bigger edges if it means all my chips."

"In a SNG, I'd take even a small edge on the first hand"

and

"ICM considerations matter more as we near the bubble."

It seems that you play opposite to the ICM?

Doesn't it make more sense to pass on small edges early, and capitalize on them late?

I think that part of the reason why you do/say this is because of the amount of time you have committed--- as a SNG progresses, you have more invested.... but isn't the whole point of surviving to the point of "near the bubble" is so that you can take advantage of small mathematical edges?

[/ QUOTE ]

I don't see how those are contradictory. I think they are pretty coherent.

[multifast1]

The only way I see to determine the edge required is to first put an expectation value on your goal (ITM, 1st place, etc..) if you go all-in and win versus folding.

An example: Goal is to be ITM
- If doubling up first hand, expectation is 80% ITM
- If folding, expectation is 40% ITM

Therefore: Edge Required = 0.8/0.4... or 2/1

Obviously you have to put in your own numbers based on your skill level relative to your opponents and how you play with a big stack versus short stacked... but I think that's the only mathmatical way to approach it.

[Misfire]

[ QUOTE ]
The reason why we play different early is because of skill inequities and just the kind of game we excel at.

[/ QUOTE ]

I'm not sure if this is clear thinking or not, but here's just some random thoughts that came to me...

It's not really possible to fold into the money, but it is (in my cheap 5+.50 games) possible to fold into, say, 5th. You could then be pretty safe assuming that before then you'll catch better or safer opportunities to increase your chipstack than taking a risk flipping a coin for all your chips on the first hand.

[Newt_Buggs]

[ QUOTE ]
Tournament Poker for advanced players has a discussion of this. You don't want to take gambles when you have a small edge if they will prevent you from taking a gamble when you have a bigger edge later.

[/ QUOTE ]
[ QUOTE ]
If I have TT and my opponent went all-in while flashing AKs, I'd call. I don't know if its correct, but I think my TimeEV makes up for any EV lost in that tourney.

The problem with preflop coin flips is half the time you are thinking about gambling (maybe becasue you already have a raise in the pot) you aren't really in a coinflip.

[/ QUOTE ]


Let me see if I can show the contrast between these two good points. I think that in the first quote Sklansky is assuming that your ONLY goal is to make as much money as possible in THIS tournament.

Online though you can just fire up another table if you get busted 1st hand. Technically wouldn't this mean that any play that is marginally +EV is the right play because you lose no invested time and are making money long term. Assuming the previous ICM calculations are correct, calling an all in with 10s in lvl 1 against AKo will certainly decrease your ROI (you are more likely to win the tournament by folding and outplaying your opponents later) but increases your hourly rate and hence your overall profit. Yes, by folding you are passing up a chance to make even more money in the tournament, but you are also giving up the chance to play another tournament in its place

[Blarg]

In which you will again pay a fee to play, but then pass up the chance to make more money if you can so you can again play another tournament instead of maximizing your value on this one?

Where does this circle end? And how many fees do you have to pay before that happens?

[KenProspero]

I'm not sure your analogy is appropriate. Even with AA, you're going to lose ..... ummmmm 25% of the time +/- so, isn't it more like playing 9 handed with twice the chips for $26 or so, vs playing 10 handed with the same number of chips for $11.

[david050173]

Lets assume on the first hand of every tournament you can take this coinflip. Should you? Winning a coinflip on the bubble means that you make the money most of the time. Winning a coinflip now gives you a better chance. You will need to make the money 75% of the time when you win to get a 37.5% winning percentage. I am guessing hitting the money 75% of the time after an early double up is a bit optimistic . I would also guess that you have a better chance of winning (not just hitting the money) if you win this gamble.

In RL you never know if you are dominated, dominating, or coinflipping. If people are only pushing with pocket pairs, you should be folding AK (that 5% difference will add up) but if you through in AQ-AT calling might be a decent move. Similiar if they are only pushing/calling with AA-JJ and AK, pushing with TT is going to be a long term losing proposition (you win 30 chips most of the time, but are a big dog the 15 or so % of the time that you are called for all your chips).

If I don't have notes (ie he way over plays weak aces/small pairs), I don't want to get into this coinflips early with AK,AQ, and pairs smaller than Jacks.

[viennagreen]

are we talking about a true coinflip or a small edge? say a small edge--- 52%.

how often do you make the money by taking a 52% edge 10-handed with equal stacks?

[Nottom]

[ QUOTE ]


If you had the choice to play a ninehanded 10+1 with 1600 chips but you had to pay $21 to enter, would you prefer to do so over paying $11 for 800 chips.

Payouts are 50-30-20

[/ QUOTE ]

Well thats not really what we're doing is it? Note I chose my hands carefully, TT is actually a pretty decent favorite over AKs. I would certainly not gamble the other way without enoug money in the pot to make it worthwhile.

So now we are looking at a spot where 55% of the time I have 1600 chips for $11 or I lose (which happens 60% of the time anyway) and I get another 800 chips for $11.