3-of-a-Kind
Join Date: May 2008
Posts: 78
|
|
Ok, there's something about this that's been bothering me for a while. While I think the practical conclusions of this idea are good, some of the theory doesn't really make sense to me. Since this has been bumped anyway I'll take this opportunity to be controversial and get massively yelled at by everyone:
Quote:
|
Originally Posted by Renton
Not only do you want to make the best play with THIS hand, THIS time, but you want to make the most money with your range, EVERY time
|
So what you seem to be saying is that you should make a certain play with a certain hand in order to be able to make more money with a different hand in the same spot. What I'm going to ask is: When do you make this money?
You obviously don't make it at that moment with that hand, that's kind of the whole point.
You don't make it in some parallel universe where you have a different hand because:
A. I don't really care how much money parallel-universe me has.
B. What I do in this universe doesn't effect what hand range my parallel-universe opponent puts parallel-universe me on.
Therefore the only time I can make this money is in future hands.
So lets assume for the sake of argument that you do the unbalanced thing and cbet 99 and somehow your opponent gets the idea that you're c-betting your entire range. He adjusts accordingly. Surely it's now possible for you to readjust to this? If your opponent thinks you are c-betting everything then surely the way he chooses to play as a result of this will also be exploitable.
I guess what I'm saying is that if you do something once that is profitable this time but would be exploitable if you kept doing it then that's probably not a bad thing. Your opponent will adjust and you can go back to doing the "balanced" thing and make more money from him.
The c-betting example is a pretty complicated one that is quite hard to analyse because of multiple streets of subsequent action, so I'm going to look at a couple of simpler ones where what I see as this same misconception about doing things "for your range" is bandied around.
Firstly, light 3-betting. People will often say things along the lines of "yeah, light 3betting's great because it means you get paid off on your big hands" ie. another way of saying that it's good for your range.
Again I'll hypothesise that 3betting 76s can't increase the amount of money you would have won if you'd had aces. So if we make the assumption that you're not 3betting 76s because it's profitable in isolation, you must be doing it to force your opponent to adjust such that you make more money in future on your big hands. The question I would ask though, is why would he want to adjust? If you're losing money 3-betting 76s without him adjusting and since he hasn't adjusted you're making the same amount from the rest of your range, he's now making more money than he had been without having to do anything. Of course in reality it probably is profitable to 3bet 76s in isolation, but you should understand that the reason you're doing this is to make money NOW. In all likelihood your opponent will adjust in such a way that he is actually less exploitable. It is possible that he'll over-adjust and start paying off your monsters with nothing, but in general, your exploitation is not going to make him more exploitable.
As a second example I'm going to consider a river situation in which there are two players, each with a pot-sized bet behind on the river. Player A's range is the nuts 50% of the time and air the other 50%. Player B's range is entirely bluff catchers. Both players know this.
So what is the most balanced thing for player A to do? Assuming that if he bets he shoves, then he should obviously push with all his nut hands. To balance this he should push with half his bluffs so that it makes no difference to him how often B calls. (Think about this for a few minutes, it does add up.) So if A does this then there is nothing B can do to exploit him.
Similarly B's most balanced play is to call 50% of the time. If he does this it doesn't matter how much A bluffs, he always makes the same amount. If both players are doing this then the system is dynamically stable, since neither player can make more money by changing how much they bet.
However, just because these would be the most balanced plays to make, doesn't mean they'd be the best ones. In fact doing this makes it impossible to exploit any mistakes of your opponent. Playing with perfect balance means that no matter how exploitable your opponent's decisions are, you only make as much money as you would if he were also playing perfectly balanced poker.
If player B is calling less than 50% of the time then the most profitable course of action is actually to bluff always. At least until he readjusts. Then when he starts calling more than 50% you should never bluff. If two players are constantly trying to readjust and exploit one another, and they are both fairly competent, then over a period it will appear as if they are playing in a balanced way, when in fact they are always trying to make the most profitable decision for that particular hand.
So I guess the main points I'm trying to make are:
1. Being "balanced" shouldn't be a goal in itself, but rather an inevitable conclusion of trying to make +EV plays (and i mean NowEV) on a hand by hand basis against competent players.
2. When you do something "for the sake of your range" you are actually doing it for one of two reasons:
2a Because making the less balanced play would actually be less profitable in this particular hand with these particular cards (often because it would turn your hand face-up.)
2b In order to make your opponent more exploitable in future.
3. For 2.b to be the case your decision has to be either a case of:
3a. Not exploiting a mistake he makes regularly for a small edge, when it will give you a bigger edge if he keeps making it.
(As an eg. think of a player who could be 3-bet with atc profitably. You don't want him to adjust so you don't 3bet him with atc, but only the ones with most EV vs his calling range and you call some hands that can also be played profitably in this way)
3.b Doing something that will tilt your opponent into adjusting badly (even though he doesn't need to) and make him more exploitable.
4. You should always know why you are really doing something (ie. 2a or 2b, profit now or profit later) and not just use the get-out "good for my range".
5. You shouldn't hunt me down and string me up for appearing to diss ABCD theorem, which I do mostly agree with.
|