Too tight?

I'm about halfway through Winning Omaha/8 Poker by Mark Tenner and Lou Krieger. It's OK, but I'm wondering if it is practical to play the way the authors suggest, or whether it could even be profitable to do so.

Maybe I should put the caveats up front: I'm far from an expert in the field. In fact, it's precisely because I always feel that I'm just making it up as I go along when I play the O/8 portions of my almost nightly HORSE tournaments that I wanted to read something to give me a more solid grounding in the game. I therefore have no vast fund of experience against which to judge their advice. Yet I'm still finding myself skeptical that it makes any sense.

Let's begin with their starting hand requirements. On pages 82-84, they identify seven classes of playable hands:

1) Any A-2.

2) A-3, if the ace is of the same suit with any of the other three cards.

3) A-3, if the remaining two cards are 10-pointers (i.e., would count a 10 in blackjack--10s, jacks, queens, and kings).

4) Ace with two other "prime cards," which they define as aces, deuces, treys, fours, and fives. Exception: Do not play any A-5-5-x hand.

5) Suited A-x if two of the cards are 10-pointers.

6) Any four prime cards with no ace. (I assume that they would exclude trips or quads, though this is not explicitly stated.)

7) Any four Broadway cards, unless you are dealt trips. (Presumably quads, too, though again this goes unsaid.)

They have more details about which of these are stronger than others, which features within each category make a hand more or less favorable, which ones are weak enough that you shouldn't play them from early position or after a raise, etc. But that's the basic formula.

Incidentally, I note that this list, while tight, nevertheless was not completely consistent with their earlier dogma, quoted here, about not playing hands containing the dreaded sevens, eights, and nines. For example, A-2-8-9 obviously fits into their category 1. The list also violates their other earlier-stated dogma, quoted here, about avoiding "three-legged stool" hands. For example, A-2-3-9 fits category 4, but has an obvious "dangler" (and violates the other rule at the same time). On p. 81, they write--in italics, even--"[D]anglers are the kiss of death in Omaha starting hands." I have so far found no attempt to reconcile this assertion with the list of starting hand requirements that clearly allow for some dangler hands.

After studying this list for a while, I got to wondering what percentage of hands one would play if restricted to these rules. I attempted to work it out mathematically, and came up with 81,882 total hands, which amounts to 30.2% of the 270,725 possible starting hands. However, I was highly unsure of whether I had this right, because I found no easy way to account for hands that fell into more than one category. For example, Ad-3c-Kd-10s fits into both categry 2 and category 3. Ac-2c-3h-10d fits into categories 1, 2, and 4. Because I couldn't be sure how much double-counting was in my total, I had little confidence in my conclusion.

*******
A digression. Long ago I read somewhere (can't even vaguely remember where it was now) about a college physics professor who assigned his class to determine how long it would take for the center of a one-pound roast to reach 180 degrees in a 350-degree oven. One student reasoned that beef was mostly water, and therefore used the known thermal properties of water to make a theoretical calculation. Another went out and bought a one-pound roast, stuck it in an oven with a meat thermometer, and waited for the target temperature to be reached. A third student called his mother, an experienced cook, and asked her for the answer.

The philosophical question raised by the story is which student would make the best physicist? The field needs theoretically minded researchers. But it also needs good experimentalists. And, finally, it needs those with the insight to perceive shortcuts.

********

So, finding myself stymied by the theoretical approach, I decided to try the experimental. Yesterday I logged onto PokerStars and found an insanely loose $0.02/$0.04 limit Omaha/8 cash game, ten-handed, with 49% to 77% of players seeing the flop (according to the stats in the lobby, which I checked periodically). Over the course of about two hours, I played exactly 100 hands, checking each one against the Tenner/Krieger list. Only 10 of them would meet the criteria. (Specifically, one each from categories 2 and 7, two each from categories 1, 4, 5, and 6, and none from category 3. However, I did not check carefully for multiple qualifiers, just chalking the hand up to whichever criterion I first noticed it would meet.)

(Incidentally, since nearly every hand went all the way to a showdown--this being a no-folding group of players--I also kept track of how many non-qualifying hands would have turned into winners. The 90 "don't-play" hands would have won four scoops, 14 half pots, 2 quarter pots, and, one time, would have chopped the low three ways for one-sixth of a pot. The obvious implication is that despite the questions I'm raising in this post about how tight is too tight, it's clear that the great majority of "reject" hands as classified by Tenner and Krieger really are dogs, with only remote hopes of winning.)

Ten percent was way under what I had estimated mathematically. Even with a fairly small sample, that was too far off for comfort. It was also taking too long. So I broke out a deck of cards and dealt myself another 100 Omaha starting hands. I found five in category 1, eight in category 2 (of which three would also have qualified in another category, but I lumped them all here), two in category 8, and none in categories 3, 4, 5, 6, or 7. That's a total of 15 "playable" hands out of 100, with 85 rejects.

This combined sample is now big enough for me to believe that the listed guidelines would actually result in playing about 10-15% of starting hands. I think that I must have messed up the math somewhere in my first accounting, but it's too much of a chore to go back and try to learn what went wrong.

I don't think there's anything intrinsically wrong with recommending a starting hand spectrum that tight--especially for beginners, as this book is explicitly aimed at. In fact, the authors trouble themselves to point out that more experienced players can profitably play a somewhat wider range of starting hands, because they are less likely to get trapped with good-but-not-quite-good-enough hands in a large pot. Fair enough. I certainly believe that to be true for hold'em, so I find it perfectly plausible for Omaha as well.

But the authors' obsession with avoiding second-best hands extends past their starting requirements, and this is where I began to wonder whether their full set of guidelines were really feasible.

They are positively fanatical about getting away from straight draws and even made straights if you don't also have a solid low to accompany it, or in the face of a possible flush. It's not hard to see why: straights can be beaten by higher straights, flushes, and full houses, and even if they hold up you may end up with just half of a pot. So I see the underlying wisdom and reasoning.

Still, their straight and straight-draw rules strike me as draconian and timid: When three low cards flop and you don't have the nut or second-nut low to go with your straight draw, fold it. If two low cards flop, draw to the nut straight only if making your straight won't produce a possible low, unless you have a strong low draw, too--and even then you should play it passively, checking and calling, rather than aggressively. With no low cards on the flop, don't play a straight draw unless you have at least seven outs to the nut straight or you have additional outs such as top two pair or top or middle set. Non-nut straight draws with bottom two pair or bottom set should be mucked. If the flop has two suited cards, you must have at least eight outs to the nut straight to continue, and play it check-and-call. A straight draw with two suited cards and two low cards on the flop should be mucked, unless you have at least the second-nut low draw to accompany it. If you flop a straight with three low cards on board, check/call with three or fewer opponents, and muck against four or more; with two low cards, also just check/call. If you flop the nut straight with no low cards on board, never raise without a redraw. If you flop the second-nut straight, fold unless you have at least six outs to the nut straight or a full house. Muck the nut straight with three suited cards on the flop; just check/call when two suited cards flopped.

You get the idea. Basically, they're trying to get you to recognize situations in which you could get trapped for a lot of bets which will prove to be the second-best hand, even if it is the best hand on the flop.

No doubt there is a lot of hard-won, painful, expensive experience speaking here. But it leaves me wondering what fraction of hands one would end up actually playing to the river under their guidelines. And I haven't even gotten to their chapters on playing the turn and river, where I assume there will be yet more advice to muck, muck, muck just about anything short of the mortal nuts. I get the feeling that, especially in a limit game, the very few pots eventually won will not be large enough to justify the large number of times one is folding out of fear of running into a better hand or winning just half a pot. I certainly don't know this to be so, and I am given pause in my criticism by the authors' vastly greater experience. But offhand it just doesn't strike me as right. My limited experience--admittedly against really bad opponents--is that it often does not take the nuts to win. When you have many opponents willing to go to the mat with 10-high and jack-high flushes, it seems OK to chance it with the king-high flush rather than always living in fear that the nut flush is lurking.

I am also concerned by the number of situations in which they advocate completely passive checking-and-calling play, rather than taking the initiative. That obviously goes against the most basic and universal advice of good poker: being selectively aggressive, and putting your opponents to the decision, rather than reacting to their actions.

All in all, I am left wishing that I could watch one or both authors play through an extended session and see whether they actually play as snugly as they are laying out in their book. My hunch says no. But I could be wrong. Lord knows that has happened once or twice before in my life.