clarkmeister theroem

[Asim]

i have read this in a few posts now.. i was wondering if someone has a link that explains it.. I did a search but couldnt find it.


what is it?

[jrz1972]

You're HU and OOP. The river card puts four to a flush on the board. According to the Clarkmeister Theorem, you should bet with any two cards.

[MrWookie47]

I disagree. You don't bet with any two. You bet for value. If you have top pair or something, bet so as not to let villain check behind. If you somehow made it that far w/ just J high, I think I like check/folding.

[therockofgibraltar]

I have not read the original post but I have understood it the same way MrWookie: you don't bet with any two cards but with something (top pair, middle pair etc.)

[Paxosmotic]

How about bet with any two as long as you can fold to any raise without concern for your precious table image? Bet and call with a decent hand. I think the pot is generally going to be large enough on the river that if the bluff succeeds even 1 in 8 times it's going to be profitable.

[Rev. Good Will]

clark discussed this in a previous thread not too long ago

http://forumserver.twoplustwo.com/sh...mp;sb=5&o=

[bosoxfan]

this was a pretty good discussion


thread

[detruncate]

[ QUOTE ]
How about bet with any two as long as you can fold to any raise without concern for your precious table image? Bet and call with a decent hand. I think the pot is generally going to be large enough on the river that if the bluff succeeds even 1 in 8 times it's going to be profitable.

[/ QUOTE ]

Quotes from Clarkmeister:


"Bet 100% of the time..."

"bet almost all the time"

And one more useful bit from a "theorem" thread:

" ...liberally betting the river out-of-position in pots where you come into the river with the initiative is a good strategy. Sklansky's "55% chance of being ahead when called" rule for a value bet does not apply in situations like this, when frequently, most worse hands will check behind, but almost all better hands will bet. You could need as little as 10-15% chance of being ahead when called and still be correct to find a bet." Link.

There have been tons of these threads. Basic explanations are easy to find. Here, if nothing else (since I don't want to go through it again). If you can't find what you're looking for, check the micro/ss archives. <----- "older archives" link on the green monster, then "click here to read recent archives."

It's the combination of value betting + taking the pot down with a second-best hand that is the basis for the play -- we don't have to win via bluff very often for our EV to be significantly improved since we're dealing with the size of the pot rather than just winning or losing some % of our river bet. However, it is heavily slanted in favour of value betting rather than bluffing. We expect to lose fewer extra chips to better hands when we bet than we make from worse hands that would otherwise be checked behind. Winning the whole pot sometimes might tip things from marginally profitable to clearly profitable, but it's still the value bets that get us most of the way there.

When the value bet component of the equation is very small, the bluff equity bit needs to be at least big enough to compensate. Just bear in mind that your bluff equity is often not as large as it might appear at first glance, as average opponents will usually call any flush + almost any pair. It's this very thing that makes the value bet profitable, but it also means that you have to target your bluffs carefully.

General rule of thumb: if you're going to call a bet, you should definitely bet instead in these situations. That takes care of nearly all the hands you see a river with, as we have to remember that we're HU and should therefore call with most hands that have at least some showdown strength (relatively speaking). The opportunity to bluff our opponent out of the pot will take care of some of the leftovers, leaving relatively few check/folding situations.

Bottom line is that betting 100% of the time is probably not a big error since it will almost always be correct to do so if you're getting to the river correctly, but failing to bet when you should is a leak. Only check/fold hands with absolutely no showdown value. It's clearly best, of course, to be able to correctly recognize when you have no showdown value... but err on the side of betting if you're not sure.