[Shandrax]

[ QUOTE ]
[ QUOTE ]

Especially the hand grouping stuff is driving me crazy. I have the Sklansky groupings, then I found some study on the internet that attempts to prove that they were incorrect and then I got the Lederer groupings. Quite a dilemma...

[/ QUOTE ]

No dilemma. In all but the toughest games, suited cards that add to 20, AA-77 and AK-AJo + KQo are your bread and butter in unraised pots.

Other hands are situational. Those situations can include:

Improved position

As your position improves in unraised pots, suited aces and suited cards that add to 19 can be added even if the pot will not necessarily be multiway.

A multiway pot

If there are 2 or more limpers (plus the blinds), you can add all pairs. With 3 or more limpers, all but the smallest suited 0- and 1-gappers become playable. In late position, suited King-middle becomes playable multiway.

If you expect 5 people to see the flop because that's the way your table plays, then you can call with any pair in early position even before there are limpers.

A weak player has entered the pot

Usually, you want to raise and isolate. You don't always need to have even a normal calling hand, as you will be playing the board against this player if you are successful in isolating him.

Blind steal

If you are in late position, and the pot is unopened, unsuited big cards that add to 20 become playable. Since the pot will not usually be multiway but heads up, being suited is not crucial because you will not have the pot odds to draw anyway. Instead, you will be hoping to flop top pair and punish draws, dominated hands and middle/third button hands.

[/ QUOTE ]

That's the logical way to approach the problem, BUT sometimes computer simulations discover things that can't be explained logically at first (usually happens later when people are aware of the phenomenom).

Let's assume we follow the theory from the WLLH-thread and see the small stakes player as someone who more or less mechanically follows a good (but pure) strategy like a Blackjack-dealer. The Blackjack-dealer doesn't play optimal, but if you keep the counters away, he will make profit. In the same way over the long run the poker player will beat all players who play a different strategy that contains more leaks.

Now we have to identify possible leaks. If we take the starting hands, then we might have a reason for a huge swing in percentages already. If some computer simulation found out that playing T8s utg adds 0.01c to your bankroll on average, then folding it will reduce your winnings. That's a pretty ridiculous example, but I hope you get my point.

The initial hand selection builds the foundation for maximum profit and if there is a computer proven way to play, it will beat human aproximations in the long run.

Personally I would love to see our famous authors like Sklansky, Malmuth and Miller start a project with computer simulations to do research on the problem. First step would be to contact the guy who analyzed the old hand groupings and talk to him, maybe spot errors in his assumptions and so on.