Starting hand selection in LO8

[RoundTower]

OK I wasn't going to start a thread on this topic, but this makes more sense to reply to what Buzz wrote in the "grade the December magazine" thread.

[ QUOTE ]
[ QUOTE ]
Obviously a fair bit of research has gone into this article.

[/ QUOTE ]Yes. [ QUOTE ]
But I'm suspicious of "point count systems", etc, as a way to govern your starting hand selection,

[/ QUOTE ]If anyone really wanted to use a "point count system," I'd advise them to use Hutchison's system.

I tried to come up with a nitty gritty way to quickly select hands with which one should probably be seeing the flop for the December issue. Seems to me where people who are new to limit-Omaha-8 stumble most is with regard to what starting hands to play.

What is presented in the article is a simple three step mechanism to quickly evaluate the playability of a starting hand.<ul type="square">
1. flush value
2. high combo value
3. low combo value
4. And then you also consider position.[/list]You're "suspicious"?

You can easily alleviate your suspicion by trying the system for yourself. See if it works for you or not.

Try this: Deal out nine or ten four card Omaha-8 hands and then quickly evaluate each as described in the article (without regard to position) for<ul type="square">1. flush value
2. high combo value
3. low combo value.[/list]About one out of every four hands, on the average, will add up to a total of five or more points. Put the hands the system considers "playable" to one side.

Then deal out a five card board and see which hands of the nine or ten dealt would end up as winners for high, low, or both (scoop).

Do this a hundred times, keeping track of how often the 5+ hands end up as winners in each category (high, low, scoop) and how often the other hands end up as winners in each category. That process should at least remove your suspicion and will not cost you any money in the bargain. You'll see for yourself whether the system is a good predictor of success or not.

But remember, if you decide to use the starting hand selection system in a real game, you still have to play good poker with regard to other decisions. I advise you (or anyone) to read or re-read Ray Zee's classic. Success playing the game involves much more than just good starting hand selection.

I remember the first time I sat down in a casino Omaha-8 ring game. I didn't know which starting hands to play and which ones to fold. As a result, I played too many hands and lost. It was obvious to me at the time that I was seeing the flop with more hands than my opponents but it was not clear which ones should be folded until I saw the flop.

[ QUOTE ]
and I don't think the article has much practical value.

[/ QUOTE ]That's ironic. I actually took your criticism from last month to heart and tried to write something for December that had very much practical value.

Buzz

[/ QUOTE ]

First I should note that I'm not an expert at all in this game, I rarely play it and haven't consistently beaten any level above 5-10 or so. So don't take anything I say as gospel.

The system has been developed by running simulations based largely on your hand's equity all in preflop. This would be great to answer the question "my opponents all move all in blind, what do I need to call?" but not so great to answer "three guys limp to me in the CO, what hands are profitable to limp?" which are the practical questions you have to answer.

To some extent you acknowledge this, but even based on what you wrote in this post (deal out a load of hands and see how often the high-scoring hands win). But more important is which hands win more money in game situations, I don't believe this is accurately modelled by all-in simulations.

Also, some of the point count scores seem arbitrary or just wrong. For example A[img]/images/graemlins/heart.gif[/img] 4 [img]/images/graemlins/spade.gif[/img] K [img]/images/graemlins/heart.gif[/img] K [img]/images/graemlins/diamond.gif[/img] outscores A[img]/images/graemlins/heart.gif[/img] 2 [img]/images/graemlins/heart.gif[/img] J [img]/images/graemlins/heart.gif[/img] J [img]/images/graemlins/spade.gif[/img], I would normally prefer the latter.

Finally, just because I said I'm suspicious of point count systems in general doesn't mean they aren't useful. I'm also suspicious of HUDs, one-party government, and people who wear headphones at the table.