
Dr Zen
New User
freddyvessant@gmail.com
Sep 14, 2006, 10:22 PM
Post #8 of 14(490 views)
Shortcut
|
On Thu, 14 Sep 2006 20:21:49 -0500, "eleaticus" <eleaticus@bellsouth.net> wrote: >"Dr Zen" <freddyvessant@gmail.com> wrote in message >news:b7sjg2996vrdh6pvcvmci7hhng8l2p5eda@4ax.com... >> On Thu, 14 Sep 2006 18:26:30 -0500, "eleaticus" >> <eleaticus@bellsouth.net> wrote: >> >> >"martin" <martinchaide@hotmail.com> wrote in message >> >news:1158247667.625017.289020@d34g2000cwd.googlegroups.com... >> >> hello >> >> >> >> If someone knows how to calculate if pushing a starting hand is >> >> +EV(chips) when you are against more than one possible caller ( By >> >> example you want to steal blinds) or even better someone knows an >> >> algebraic formula to do so I will be infinitely grateful for my hole >> >> life to that person provided he was so kind to explain it here. tyvm in >> >> advance >> > >> > >> >Even assuming you mean pushing allin, it can't really be done, martin. >> > >> >> You are now going to make a very long post that says "poker is a game >> of incomplete information". Well yes. Let's assume the OP already >> knows that though. >> >> >Not even against just one possible caller unless you know him/her >> >thoroughly, and then the true/exact formula would be fairly complicated >even >> >on the surface. >> >> >> Yes, but that doesn't mean he can't make some sort of calculation. >> >> It's rather like in limit. You raise before the flop, another guy >> threebets. >> >> You have to put him on a range of hands. You can't know what he has >> exactly, unless he only threebets with one hand. >> >> So you put him on a reasonable range and make your decisions based on >> that. >> >> When the flop comes AK7 and he bets out, you can work out what your >> equity is if you hold QQ. It's not an absolute figure, not what your >> equity *is*. It's what your equity will tend to be in the long run. >> >> You do the same with push decisions in NL. >> >> > >> >Will you be called with 77? AJo? etc. >> >> You just assume a likelihood based on what you know, just as you do in >> any other spot. >> >> >You have to assume there is some range of hands that one possible caller >> >might call with. >> > >> >More than one possible caller? No way. >> >> This is just wrong. You can calculate it for a bunch of callers. It's >> not easy though. > >Sure, just assume a bunch of things that won't be anywhere near right. Welcome to poker. >> >> > >> >The best approach is a bunch of simulations for each of the hands you >would >> >go allin with. >> >> Well, okay. I guess that's what PokerStove does, after all. >> >> > >> >AA against every hand someone might call with, one at a time, played to >the >> >end. >> >Ditto KK >> >etc >> >down to the least hand you would consider acting with this way. >> > >> >Then, you could take the each of the hands a given player might call >with, >> >adjust for the hand's natural frequency given your hand, etc. >> > >> >But note what you'd have to do versus both blinds to do an 'exact' >> >simulation. >> >> >> Only if you wanted an *exact* answer. >> >> >Your AA not only changes the likelihood that the 'first' player has >certain >> >hands, but that player's hand changes the likelihood of what the second >> >player might have. >> >> Yes, but a calculator like PokerStove takes that into consideration. >> >> >And you'd need - for 'exactness' - to pair each hand the first might call >> >with, with each hand the second might call with, and run your AA against >> >every one of those pairs. >> >> Yes, but an ICM calculator will not make the mistake of doublecounting >> cards. > >You read with animus and no understanding if you think anything I said there >included double counting. I don't have any animus whatsoever. I'm simply pointing out that an ICM calculator will take into account each range. > >> >> >> > >> >Given that, a formula/table c/would follow. >> >> Yes. >> >> So basically you're saying it can't be done but this is how you do it? >> Erm, yes. This is how you do it. > >BTW, you are at essence saying I was wrong, so where is the formula martin >requested? I gave it in my other post. It has a lot of xs and ys, so to speak. > >hmmmmMMMM? if x+y-z>w, EV>0. You're welcome. -- Dr Zen King of the wild pixels. http://gollyg.blogspot.com
|