Theory: Gigabet's "bands" and "The Finch Formula" Grand Unification

[curtains]

Assuming those chips come from the stacks of your opponents, then it should be =EV, so it doesn't matter whether I would or not. Offer me 1050 from their stacks instead and I may say yes.

[AtticusFinch]

[ QUOTE ]

Assuming those chips come from the stacks of your opponents, then it should be =EV, so it doesn't matter whether I would or not. Offer me 1050 from their stacks instead and I may say yes.

[/ QUOTE ]

I'll offer you 2k from their stacks, how's that? You'd still be wrong to do it.

[Exitonly]

i'm not sure who i agree with, but saying 'still be wrong to do it' isn't gonna cut it...

gotta explain taht.

[curtains]

[ QUOTE ]
[ QUOTE ]

Assuming those chips come from the stacks of your opponents, then it should be =EV, so it doesn't matter whether I would or not. Offer me 1050 from their stacks instead and I may say yes.

[/ QUOTE ]

I'll offer you 2k from their stacks, how's that? You'd still be wrong to do it.

[/ QUOTE ]

Well of course this is the common theme in this thread. An opinion that goes completely against basic math, without any evidence or logical support to back it up.

To be honest I don't think I should have to explain my point of view that a chip's value is actually constant from the beginning to end of a tournament in which you must win all the chips to be paid. This is what common sense would dictate, or so I believe would be the case for most people. When you have an opinion that is against common sense, you should carry the burden of explaining it fully if you want anyone to take it seriously, of which I don't because I've seen no compelling reason offered.

[AtticusFinch]

[ QUOTE ]
i'm not sure who i agree with, but saying 'still be wrong to do it' isn't gonna cut it...

gotta explain taht.

[/ QUOTE ]

I'm 3-tabling. I'll explain more later [img]/images/graemlins/wink.gif[/img]

[PrayingMantis]

[ QUOTE ]
Furthermore, ICM is an excellent tool for single-table analysis, and I don't suggest we replace it in those scenarios. It's not very effective for MTTs, however, because of the enormous range of stack sizes involved, among other issues. That is the problem I am trying to solve.

[/ QUOTE ]

It doesn't make any sense to claim that ICM is an excellent tool for SNGs but is not effective for MTT. If it's excellent (i.e true) for SNGs it is true for MTT, and thus it is effective for MTT too, only doesn't have a significant impact most of the time for clear reasons - being relatively far from the money most of the time.

Your theory basically negates the core idea of ICM. So you can't say that ICM is excellent and at the same time come up with a model which is completely different from a theoretical point of view (even if the results are the same sometimes, because of the nature of the formula you've invented).

BTW, I can easily come up with all kinds of formulas that at some points (specifically when you win or out) converge to what the results the the core ICM model says (with p(winning 1st)=(your share of chips in play)). However, this won't make my formulas correct or meaningful in any way.

Also I agree with curtains about your analogies, which are not convincing to say the least. In general it seems to me that your perception of EV for winner-take-all tournaments seems somehow confused. There's nothing wrong with coming up with new solutions to old problems, but in this case your solution seems absolutely arbitrary, and I don't see in what sense it helps to shade new (or true) light on this already solved problem.

There are also other very big problems with the general line of thinking you present, which I won't get into here for different reasons. Basically you are trying to fix something that isn't broken, and in fact works very very well. From my own expirieince on these boards, very often when people come with new and revolutionary ideas about MTTs, they do not understand the current and old ideas very well. I'm not sure if that is the case with you too, but there are some signs here that it might be.

[AtticusFinch]

[ QUOTE ]


Well of course this is the common theme in this thread. An opinion that goes completely against basic math, without any evidence or logical support to back it up.


[/ QUOTE ]

Cut me a little slack. I'm 3-tabling right now -- thus the terse responses. Despite that, I've made a number of logical arguments and analogies. If you want something more formal, you'll just have to wait for me to bust out. [img]/images/graemlins/wink.gif[/img]

[pfkaok]

[ QUOTE ]
But tell me this: Let's say you're in a field of 1000 people, and you start with 25% of the chips. Do you really think you have a 25% chance of winning? I don't.

[/ QUOTE ]

if you're good, you should win more than that.

the reason the chips would be worth less at that point is b/c usually 1000 people tourneys aren't winner take all.

I'm REALLY not agreeing with this log thing here, as i believe if you're just looking at your chances to win, and not prize pool, they should AT LEAST double when you double up early.

Just think of all the times you're forced to make -EV plays when you're shortstacked b/c you might not be able to wait for a +EV spot. you're sometiems forced to take the LEAST -EV. when you're big stacked you can take of advantage of certain +EV spots that smaller stacks can't... and you NEVER have to take unprofitable -EV decisions b/c there's no pressure.

[AtticusFinch]

[ QUOTE ]

It doesn't make any sense to claim that ICM is an excellent tool for SNGs but is not effective for MTT. If it's excellent (i.e true) for SNGs it is true for MTT

[/ QUOTE ]

This simply does not follow. First of all, I don't claim it's "true." I claim it provides good results when the number of players is small. To say that this necessarily means the same formula can be extrapolated to cover any number of players is absurd.

[ QUOTE ]

Your theory basically negates the core idea of ICM.


[/ QUOTE ]

This is also not true. In fact, the base case of my formula comes directly from ICM, namely that an average chip stack has an average chance of winning.

[ QUOTE ]

BTW, I can easily come up with all kinds of formulas that at some points (specifically when you win or out) converge to what the results the the core ICM model says (with p(winning 1st)=(your share of chips in play)). However, this won't make my formulas correct or meaningful in any way.


[/ QUOTE ]

I agree 100%. I also think that cEV = $EV is a perfect example. It works well for cases below double the average value.

[ QUOTE ]

I don't see in what sense it helps to shade new (or true) light on this already solved problem.


[/ QUOTE ]

To say that this problem has been "solved" is preposterous. ICM is useful, but flawed in a number of ways I won't bother to recount here, even in 1-table situations.

[PrayingMantis]

[ QUOTE ]
[ QUOTE ]
It doesn't make any sense to claim that ICM is an excellent tool for SNGs but is not effective for MTT. If it's excellent (i.e true) for SNGs it is true for MTT


[/ QUOTE ] This simply does not follow. First of all, I don't claim it's "true." I claim it provides good results when the number of players is small. To say that this necessarily means the same formula can be extrapolated to cover any number of players is absurd.


[/ QUOTE ]

If you really think that ICM provides good results when the number of players is small, but you can't extrapolate from it to cover any number of players, you simply do not understand ICM at all.

[ QUOTE ]
[ QUOTE ]
your theory basically negates the core idea of ICM.



[/ QUOTE ] This is also not true. In fact, the base case of my formula comes directly from ICM, namely that an average chip stack has an average chance of winning.

[/ QUOTE ]

This is very confused and not clear. First, what is "average chance"? And if the "base case" of your formula comes from the core idea behind ICM, namely that p(winning)=(your share of chips in play), why is your formula different from this? And if it is indeed different, why does it need to "come" from the same idea behind ICM? Not clear at all.

[ QUOTE ]
[ QUOTE ]
BTW, I can easily come up with all kinds of formulas that at some points (specifically when you win or out) converge to what the results the the core ICM model says (with p(winning 1st)=(your share of chips in play)). However, this won't make my formulas correct or meaningful in any way.

[/ QUOTE ]

I agree 100%. I also think that cEV = $EV is a perfect example. It works well for cases below double the average value.

[/ QUOTE ]

This sentence is not clear. I really can't understand what you mean by it. cEV=$EV is not a formula for deciding the p of coming first. What do you mean by "it works well for cases below double the avarage value"? What do you mean by "value", for instance?

[ QUOTE ]
[ QUOTE ]
I don't see in what sense it helps to shade new (or true) light on this already solved problem.


[/ QUOTE ]To say that this problem has been "solved" is preposterous.

[/ QUOTE ]

I don't see why it is preposterous. It is solved in the sense that there is a very useful solution, that is very logical, simple and even intuitive in many ways. Many simulations in the past had shown that it works very well, so you have some very good evidence too.

[ QUOTE ]
ICM is useful, but flawed in a number of ways I won't bother to recount here, even in 1-table situations.

[/ QUOTE ]

It is flawed like any model that deals with a game played by human beings. Your solution seems very far from being a better one.