hand ranges and making $$$ article

[robilaruk]

this is a very long article, but has some really good stuff in it IMHO (tho I have to admit I don't understand half of it!) well worth a read and would be interested to know what the more maths minded among us think to the theories put forward cheers Damo http://sports.espn.go.com/espn/poker/columns/story?columnist=bluff_magazine&id=2817110

[GaF]

Re: hand ranges and making $$$ article Never heard of Sklansky dollars before - but isn't it just ev? :unsure Will read the whole article properly at another time :ok

[BAM]

Re: hand ranges and making $$$ article It seems that Galfond is explaining SHAL or EV in a different way. Things like this need to be drilled into yourself until it becomes automatic but the most difficult thing is surmising your opponents hand ranges. For the pro who is playing pro everyday it becomes easy but try putting that into practice in a 1000 runner insidepoker freeroll and you may come undone :lol Great article though Damo:ok

[slapdash]

Re: hand ranges and making $$$ article Well, I think I understand what he's getting at, and he's right, though as GaF and MCFC say, it's nothing mind-bogglingly original. I'm a bit sceptical about the mathematical framework, though. I'll explain why later in this post. Here's my restatement of what he's saying. How do you tell if you're playing well? The easiest way is to see if you're making money. In the long run, this will work, although it won't give you much idea about which parts of your game are profitable. And it has the disadvantage that there is a lot of variance in poker, so it will take a long time before long term trends dominate short term random fluctuations. If you use "Sklansky dollars" (a.k.a. working out your EV, as GaF says), then this reduces the variance, and so gives you a better idea in the short term of how successful your plays will be in the long term. OK, but that still leaves a lot of variance. Consider the following, somewhat artificial, situation. On the river, you "know" that your opponent has the second nuts. You have just enough chips left to make a pot-sized bet. For simplicity, let's suppose that you'll either check-fold or go all-in. Obviously, you go all-in if you have the nuts. But if you are predictable enough to go all-in only if you have the nuts, then your opponent will fold, and you'll just win the existing pot. Your best strategy, against a perfect opponent, is to go all-in with the nuts and also to bluff sometimes, so that when you go all-in you have the nuts twice as often as you bluff. Then, in the long run, it makes no difference whether your opponent calls or folds. If he calls, then two out of three times he loses an extra pot-sized bet, and one out of three times he wins twice the original pot. Suppose he calls, and it happens to be one of the times you're bluffing. Then in "Sklansky dollars", your bluff has cost you the size of the pot. But that doesn't mean you've played badly. If you calculate your EV for the whole range of hands that you would bet with, then you have made money. And calculating your "G-bucks" will show this. Where I'm sceptical is that if you calculate what your "G-bucks" would have been if you never bluffed (when your opponent has in fact called), you'd have made even more G-bucks. But of course this doesn't mean that never bluffing is the right strategy (unless you know that your opponent will always call a bet). So what I'm sceptical about is that the concept of G-bucks take into account some of the ramifications of varying your play, but only one part.