Daliman vs Sklansky

[Aleo]

Yeah, I'm not sure how much stock people are putting in ICM for these kinds of situations, but I think in the next couple years, this kind of analysis is really going to spread beyond the SNG forum.

ICM is a powerful tool, and for the most part, I am a true believer. At very least, I think the kind of analysis you have just provided is more meaningful than any of the usual discussions I see about these kinds of situations.

Just the other day, I won a copy of Poker Academy Pro 2.0 and toyed around with it for a bit because it is supposed to be the most advanced poker AI on the market today. After about 1 hr of playing SNGs, it was brutally obvious to me that the program did not make any kind of $EV decisions and simply made perfect CEV decisions. I talked to their support people and they knew that the tournament AI was flawed in this respect, but as far as I could tell had never even heard of ICM type calculations.

Talking with other players I respect, and reading on other forums, I have seem a few other models for approximating $ equity but none come close. Most have not even heard of ICM by that or any other name.

I saw this formula just the other day, and it actually seems to have a fair amount of respect

The Landrum-Burns formula:

With N players left and T chips in play, first place pays F, and the
total prizes for the players left total to P.

Then a player with C chips has an equity of:

((T-C)/T * (P-F)/(N-1)) + C*F/T

I'm not going to get into a big discussion about all the things that are wrong with this formula, but suffice it to say, I've seen some fairly indepth conversations trying to deal with tournamnet situations with formulas like this which are not good at all.

I think that as people start to realize how good ICM is, we are going to see a lot of "spitting into the wind" going on and you will not be the only one dismantling widely held beliefs.

Nice post. I'll look through the numbers a little more carefully in the next couple days, but if you aren't missing anything, I think it will be hard to dispute your results.

Regards
Brad S

[raptor517]

rawr. one day ill make a really smart sounding post. give it a few years. holla

[skipperbob]

[img]/images/graemlins/smile.gif[/img]"GFY" [img]/images/graemlins/smile.gif[/img]

[Aleo]

Luckily, smart and smart-sounding are two different things (though not mutually exclusive, of course)

Besides which, you can settle for the envy of most players on this forum. Oh to be a teenager again, multitabling the 215s.

Um... Holla?
Brad S

[bluewilde]

[ QUOTE ]
Um... Holla?

[/ QUOTE ]
Well done, sir.

[Maulik]

great post. Let's PM Sklansky to see his response.

[MagnoliasFM]

Sklansky = overpaid donkey. he not number one. Daliman number one.

No sarcasm intended.

[skipperbob]

I'm not one of David's pals, but he did give me the best puzzle I've ever heard....If anybody wants it = PM me

[durron597]

Great post. I agree with most of your analysis except this point:

[ QUOTE ]
I am going to give him a relatively strong initial raise range of 66 , ATs , AJ , KQ , and QJs or better in all cases. This is 9.06% of the total hands dealt, and 3.35%(the raise-callable hands) equals 36.97% of the initial raising hands, which I will round to 37%.

[/ QUOTE ]

I think Sklansky is assuming that the initial raiser's range is much tighter than that. Of course, I'm not sure how it could be much tighter without making it like 99+, AQ/AJs, which is simply based on the size of the raise (in Sklansky's mind). I think that you would have had to play with this opponent for the entire tournament to put him on a range that tight, and I agree with your conclusion against an unknown.

However since I have not played at a WPT final table I really can't come up with a good initial raising range in the dark.

[hoyasnaxa]

Great post. I don't have the link but somebody went through Harrington's book and showed that many of the decisions in the book could be dismantled in a similar light. I was also wondering if the HU all-in experiment was still going on and how you were doing if it was. Thanks.