Buzz's LO8 article (Omaha 8 or Better Blind Defense 101)

[Nate.]

All--

Many around here are properly grateful for Buzz's unfailing politeness and his ability to stomach any amount of tedious Bayesian calculation. Unfortunately his recent LO8 article is severely disappointing. He recommends completing the small blind with "only excellent" hands.

There is no real analysis or calculation backing this claim up: I am reminded of the advice in Miller/Sklansky/Malmuth's book that any piece of poker "analysis" that does not attend to the expectation of a play should be handled skeptically. Moreover, this is a surprising claim, one that runs against the practices of expert players (and some other published opinions). Just to get an idea of how surprising this claim is, consider LHE: there it is correct to complete with quite a few hands, and LHE is a more positional game than LO8. The reasoning given in the article is reminiscent of LHE advice circa 2003: intuitive, eloquent, and logically empty.

--Nate


Link to article:

Omaha 8 or Better Blind Defense 101 by Frank Jerome

[Buzz]

Originally Posted by Nate.

Unfortunately his recent LO8 article is severely disappointing. He recommends completing the small blind with "only excellent" hands.

Yes. I was writing about full game play rather than heads-up play. I should have made that clear.

There is no real analysis or calculation backing this claim up:

True and a good point.

I am reminded of the advice in Miller/Sklansky/Malmuth's book that any piece of poker "analysis" that does not attend to the expectation of a play should be handled skeptically.

That advice makes good sense to me.

Moreover, this is a surprising claim, one that runs against the practices of expert players (and some other published opinions). Just to get an idea of how surprising this claim is, consider LHE: there it is correct to complete with quite a few hands, and LHE is a more positional game than LO8. The reasoning given in the article is reminiscent of LHE advice circa 2003: intuitive, eloquent, and logically empty.

I didn’t provide much reasoning for not defending the small blind.

The whole text of not defending the small blind is:

My best advice is don’t defend the small blind in Omaha-8.

Don’t even complete unless you have an excellent starting hand. The terrific odds you seem to be getting from the small blind are misleading unless you go “all-in” before the flop. Acting first on the rest of the betting rounds is a disadvantage.

If you have an excellent hand with a high potential to make the nuts, then that is a different story. Position is important in Omaha-8, but having the nuts is better than having position.

If everyone folds to you when you have posted the small blind, consider attempting a steal against the big blind. However, this is an offensive move rather than a defensive one and depends at least as much, probably more, on the skill of the big blind poster than on the cards you hold.

A good simple rule is “don’t defend the small blind in Omaha-8.”

I thought (and think) that good, general advice.

The main thrust of the article is to provide a logical rationale for defending your big blind.

Perhaps I can modify a section about odds I wrote for the big blind against a single opponent who has raised, with everybody else folding, to apply to small blind defense.

In a fixed-limit game when a single opponent raises, and everybody else has folded (with the big blind yet to act), if big blind is expected to fold, assuming the small blind is half the big blind, it initially seems as though you are getting seven to three (2.33 to 1) odds to call a raise. However, unless you or your opponent goes all-in, since your opponent is likely to be initiating a bet into the pot each betting round, you’re realistically gambling 6.5 small bets to possibly win 8.5 small bets (1.3 to 1 odds) if you scoop.

If you only realistically expect to win half the pot, when you defend your big blind against a single opponent, you’re gambling 6.5 small bets to possibly win 1.5 small bet (~0.23 to 1 odds). And you’ll be out of position for every betting round after the first.

I don't think those are very good odds, Nate, to get involved out of position with a hand you wouldn't voluntarily play from UTG. And you're not really sure what BB is going to do. Could be SB will be sandwiched between raisers on the first betting round.

Buzz

[Nate.]

Buzz--

You say that this even applies to completing the small blind when nobody raises. Agreed that you shouldn't call 1.5 bets willy-nilly; I had hoped to be clear that I was addressing the claim about completing (for .5 bets). Folding too much here is quite a bad mistake, and "don't even complete unless you have an excellent starting hand" is folding way too much.

(Anyway, if Villain is going to fire away on every street no matter what, I might defend (for 1.5) more loosely...)

All my best,

--Nate

[Buzz]

Originally Posted by Nate.

You say that this even applies to completing the small blind when nobody raises.

Hi Nate - Yes, I did write that.

You've got me because it's not absolutely correct. (Read on).

The article was about blind defense. Defending your blind implies someone is attacking it.

But I did interject the one sentence unrelated to defense:

Don’t even complete unless you have an excellent starting hand.

Since I interjected the sentence, although I cannot defend it (because I don't believe it, unless I qualify it), I'll try to make good sense of it.

Let's say you're in the small blind, there are three limpers ahead of you and you expect BB to check. In that case you're getting 9 to 1 odds to see the flop. And those do seem like very fine odds. But if someone with the best flop fit pushes and everybody else folds, and then the push continues on the rest of the betting rounds, it's going to cost 11 times that completion bet to see the showdown. SB will be getting odds of about 19 to 11 (1.73 to 1) for the whole pot, and 4 to 11 (0.36 to 1) for half the pot.

On the same deal, UTG is getting 4 to 1 odds to see the flop, which under the same conditions become 18 to 12 (1.5 to 1) for the whole pot and 3 to 12 (0.25 to 1) for half the pot. The point is those seemingly juicy odds SB is getting are not very much better than the odds UTG would be getting for limping under the gun with trash.

Can you see there's not really much of an odds advantage to completing from the small blind, compared to limping from UTG? And that's why completing from the small blind in Omaha-8 is not generally a good play, despite those juicy looking pre-flop pot odds.

Here's where you have me trapped: What to do, for me, always depends on my opponents. And in truth, I can envision playing a less than stellar hand from the small blind (or anywhere). Indeed, in truth, I might complete from the small blind if I, for example, wanted to be in the hand with a loose cannon who was throwing his money away. (You can't get any of it if you're not in the hand when he's throwing it away).

I really was thinking of defense, but then I did write that one sentence about not even completing from the small blind.

I don't want to retract the sentence, because I think it's good advice to not complete from the small blind unless you have a hand you'd voluntarily play from UTG or unless you have some special reason, related to your opponents or an opponent, to play the hand.

Thanks.

Buzz

The point is the implied pot odds or implied half pot odds you're getting for completing from the small blind are no where as juicy as the pre-flop odds (unless you go all-in). Since the betting isn't over on the first betting round the odds you seem to be getting on the first betting round are very deceptive - and that's especially true from the small blind.

And, of course there is also the danger BB will raise and someone else will limp/re-raise.

And Hero will be out of position for the whole rest of the play of the hand.

[Nate.]

Buzz--

Wow, that's a heck of a response. To write all this stuff day after day displays a true love for the game. The forums owe a lot to you.

I'm about to head to bed for the night, but I want to quickly point out that I think your line of explanation there is wrong, and not just because of the weird border cases you think are causing me to have you "trapped." I think your whole strategy of assuming that we need to calculate as if we were calling all the way down immediately, is simply unsound, for all the obvious reasons. Again, if your argument were correct, it would apply to LHE too, where many millions of hands of data have shown the conclusion to be empirically false. (Heck, it's hard to see how it wouldn't also translate, mutatis mutandis, into an argument against limping the button!)

I would bet $30,000 against your $10,000 that an impartial expert panel would disagree strongly with the statement that "it's good advice not to complete the SB for half a bet with any hand you wouldn't limp UTG in a ring game, unless playing conditions are unusual."

All my best,

--Nate

EDIT: Man, my gambler's instincts took over there. I sometimes have a hard time remembering that it can be considered rude to offer to bet. I just wanted to show that I'm serious about my opinion, and not just picking an argument.

I also agree that the BB's tendencies matter here. But not too many opponents I've faced have raised aggressively enough to influence my SB decisionmaking here.

[Buzz]

Originally Posted by Nate.

I think your whole strategy of assuming that we need to calculate as if we were calling all the way down immediately, is simply unsound, for all the obvious reasons.

Excellent point.

Indeed, where five players will end up seeing the flop for one small bet each, the initial pot odds SB is getting to complete and see the flop are, 2.25 times better than the initial pot odds UTG is getting to call and see the flop. On the surface, it looks like completing to see the flop is a bargain.

However:.....
The fly in the ointment is: the betting is not over on the first betting round. There still are three more betting rounds to go. And it's a split pot game. You can easily get more or less stuck calling at the showdown, hoping to salvage half of the pot, actually getting about 3.0:1 half-pot-odds to call the last bet at the time you call the last bet, but if you consider all the money you will have put into the post since that first pre-flop completion, your implied half-pot-odds were closer to 0.4:1.

And although if it's a coin flip it's correct (getting 3 to 1 half-pot odds) to call that last bet, it would be a terrible call at 0.4 to 1 half-pot odds. If you could have looked ahead, considering all the money you would invest in the next three betting rounds, you would realize you didn't actually have favorable implied pot odds to make the initial completion of the seeming bargain half small bet.

Once you make a bad decision to continue in a hand when you should fold, in subsequent decisions you will face, it may actually be correct to continue. As a result, especially in a fixed-limit game, you may get stuck in a hand you shouldn't have been playing from the outset.

World-class players may be able to pull off plays that are ill advised for ordinary players. Daniel Negreanu or Phil Ivey may be able to get away with playing hands that would better be folded by most of us. But they may be using all their considerable skills to do so, skills some of us may lack.

The bottom-line point I was trying to make is: If you wouldn't play a particular hand from UTG, then I don't think you should play that same particular hand from SB.

Again, if your argument were correct, it would apply to LHE too,

Not quite, because of (1) the split pot nature of high/low games.

In addition, many conclusions derived from Texas hold 'em do not apply well, or as well, to Omaha or Omaha-8 because of (2) the greater likelihood (because everyone has more cards), when you can't make a good five-card hand at the showdown, an opponent can.

where many millions of hands of data have shown the conclusion to be empirically false.

I can't comment intelligently about how Texas hold 'em data might be applied to Omaha-8, or vice versa.

(Heck, it's hard to see how it wouldn't also translate, mutatis mutandis, into an argument against limping the button!)

Not the topic here, but I don't think it's generally wise to limp from the button with trash in Omaha-8. I do tend to play more hands from the button, but they're mostly what I consider "marginal" hands. (I have a wide range for "marginal" hands).

I would bet $30,000 against your $10,000 that an impartial expert panel would disagree strongly with the statement that "it's good advice not to complete the SB for half a bet with any hand you wouldn't limp UTG in a ring game, unless playing conditions are unusual."

You've paraphrased what I wrote. That's not quite how I put it. I wrote:

"I think it's good advice to not complete from the small blind unless you have a hand you'd voluntarily play from UTG or unless you have some special reason, related to your opponents or an opponent, to play the hand."

Even so, maybe you're right. Maybe impartial experts would "disagree strongly" with that statement of mine.

We're talking Omaha-8 ring games here, not Texas hold 'em games, not Omaha-8 tournaments, and not short-handed or heads-up contests.
We'd need a panel of fixed-limit-Omaha-8-ring-game experts.

I'm not quite sure how we'd decide who the "experts" are. I don't think many such "experts," aside from published authors or tutors who are looking to turn a profit, want to be well known or share their opinions on how best to play.

EDIT: Man, my gambler's instincts took over there. I sometimes have a hard time remembering that it can be considered rude to offer to bet. I just wanted to show that I'm serious about my opinion, and not just picking an argument.

No problem.

Although I'm not interested in such wagers, I'm also serious about my opinion.

Interesting idea, though: a panel of Omaha-8 ring game experts.

We may already have that on the 2+2 Omaha-8 forum. But there's great diversity of types of games favored and also playing styles. Sometimes we're mostly in agreement while other times we disagree.

Regards

Buzz

[Nate.]

Buzz--

One way to tell what good players think about a situation--especially a situation as very common as the one we're discussing--is to go play with them, when their own dollars are on the line and they express their commitments through actions you can easily see. I've played many hours of ring-game O8 with some very very good players, all of whom (that I can remember, and my memory's pretty good for things like this) complete the small blind with a range far wider than anything you could possibly be advocating.

Well, I'm not sure that I have much productive to add to the discussion at this stage. Barring a sudden desire to spend tons of time theoretically analyzing SB preflop play in O8 (which my schoolwork probably won't allow for quite a while), looks like I'll just have to take a hit to my winrate if I'm wrong...

All my best,

--Nate

[Buzz]

Originally Posted by Nate.

One way to tell what good players think about a situation--especially a situation as very common as the one we're discussing--is to go play with them, when their own dollars are on the line and they express their commitments through actions you can easily see. I've played many hours of ring-game O8 with some very very good players, all of whom (that I can remember, and my memory's pretty good for things like this) complete the small blind with a range far wider than anything you could possibly be advocating.

Do they also play a similar wide range of hands from UTG?

Buzz

[Nate.]

Originally Posted by Buzz

Do they also play a similar wide range of hands from UTG?

Buzz

Buzz--

No.

--Nate

[Buzz]

Originally Posted by Nate.

No.

And from what you have written, they're limping, rather than attacking (at least pre-flop).

The implication, then, would seem to be that they feel they're getting favorable odds, or at least good enough odds, to continue.

The situation is somewhat different for a straight high game than for a high/low split game. In a high/low split game, "experts" play starting hands that have a good chance to scoop, and then play in such a way as to try to promote one way hands to scoopers - but against tenacious opponents, there's no way of getting around split pots. And that changes the odds.

I want to make sure you're talking about expert fixed-limit full-table Omaha-8 ring game play, rather than something else.

Usually the small blind post, in fixed-limit Omaha-8 ring games, is 1/3 or 1/2 of the big blind post (which is one small bet). I've done my reasoning in this thread for a 1/2 small bet small blind post, but my arguments and conclusions would be even more applicable for a 1/3 small bet small blind post.

Rightly or wrongly, I think those deliciously juicy odds you seem to be getting to complete from the small blind are very deceptive. I don't think the price of completing from the small blind, at least in a fixed-limit Omaha-8 ring game, is worth it after all. I don't think, in general, if you wouldn't limp with a particular hand from UTG in such a game, then you shouldn't complete with the same hand from the small blind.

I'm not talking about attacking the big blind (attempting a steal) from the small blind in a tight game. Rather it's a matter of completing or not for half of a small bet. (Looks like a bargain, but I don't think it actually is).

If we're talking about the same thing, you're telling me a panel of "experts" would disagree with me.

I don't know how to respond. If that's really true, I guess, rightly or wrongly, I have to disagree with your "experts."

Ray Zee has a well-written and well-thought-out article in the January 2009 issue of the 2+2 on-line magazine regarding hanging tough when you post the blind. I agree with him. And if it seemed like someone was somehow taking advantage of my not completing from the small blind, hopefully I'd react in some way so as to eliminate that slight opponent edge. I suppose that could end up amounting to playing more hands from the small blind than from UTG.

Or, in a tight game where I ended up heads-up against the big blind, I'd play more hands, probably usually attacking and especially with a good one-on-one hand. And where I didn't attack, if it came down to just me as SB against the big blind, with everyone else folding before the flop, I probably would complete more than I'd fold (probably actually mixing my play up in such a game, limping with some attacking hands, attacking with some limping hands, and either limping or attacking with some folding hands). And all of that would amount to playing more hands from the small blind in such a game than playing hands UTG.

So actually, your disagreement with that statement I wrote in the article, "Don’t even complete unless you have an excellent starting hand." makes very good sense to me, especially if your experience is mostly at very tight tables where the small blind poster often ends up one-on-one with the big blind, with everyone else getting out of the way. For me, except in tournaments, that's very rare. Usually when it ends up one-on-one before the flop in a full ring game in a brick and mortar casino, the participants chop (where they can get away with it) to save on the rake and keep the game friendly and interesting.

I still don't want to retract the statement, because I think it's still generally good advice for play in most Omaha-8 ring games. But at the same time, more important and over-riding advice is to adapt to conditions in the game in which you're playing.

However, truthfully, I can see situations where completing from the small blind would be a better play than never or rarely completing. Since the small blind poster is the second to last to act on the first betting round, it makes sense to me that, opportunistically, one would actually end up playing more hands from the small blind than from under the gun.

But except for these opportunistic times, or special considerations, I'm sticking to my position of at least not routinely completing from the small blind, just because you already have half of a small bet in the pot and because putting in the other half seems like a bargain.

Buzz

[Buzz]

Originally Posted by Buzz

I don't think, in general, if you wouldn't limp with a particular hand from UTG in such a game, then you shouldn't complete with the same hand from the small blind.

should read:

"I think, in general, if you wouldn't limp with a particular hand from UTG in such a game, then you shouldn't complete with the same hand from the small blind."

Sorry about the unintentional double negative.

Buzz

[jj179]

I disagree with the OP, although I do thank him for directing me towards an article I might otherwise have missed. I'd like to cut through the hyperbole a little - if this article was solely focusing on small blind play, not providing empirical evidence to back up a revolutionary new play in this position would be 'severely disappointing'.

However, sb play is not the main focus of the article and I would argue that suggesting you should tend to play only good or excellent hands in the small blind (full ring LO8) is not a matter for great controversy. I'm afraid it certainly does not follow that this suggestion ought to extend to LHE or further to button play in either game and there are no references made in the article offering this line of reasoning. This advice does not extend to LHE because the schooling effect that occurs in that game is either absent or drastically reduced in LO8 - although it's true any four cards can win, in my opinion you are going to need a hand with nut potential UTG or in the sb to make your call +EV. Of course if there are a lot of players already in, this initial call may be only slightly -EV, then again this first call may lead you to make greater -EV decisions later OOP in the hand.

I can't prove this opinion empirically and am just speaking from experience and drawing on that of other experienced players. I can't claim either to have played many hands against 'very very good' LO8 players (I play online only), I've rarely come across players I judged even deserved one 'very' and I think I'd find another table if I really felt my opposition really excelled at the game.

[Nate.]

Originally Posted by jj179

I disagree with the OP, although I do thank him for directing me towards an article I might otherwise have missed. I'd like to cut through the hyperbole a little - if this article was solely focusing on small blind play, not providing empirical evidence to back up a revolutionary new play in this position would be 'severely disappointing'.

However, sb play is not the main focus of the article and I would argue that suggesting you should tend to play only good or excellent hands in the small blind (full ring LO8) is not a matter for great controversy. I'm afraid it certainly does not follow that this suggestion ought to extend to LHE or further to button play in either game and there are no references made in the article offering this line of reasoning. This advice does not extend to LHE because the schooling effect that occurs in that game is either absent or drastically reduced in LO8 - although it's true any four cards can win, in my opinion you are going to need a hand with nut potential UTG or in the sb to make your call +EV. Of course if there are a lot of players already in, this initial call may be only slightly -EV, then again this first call may lead you to make greater -EV decisions later OOP in the hand.

I can't prove this opinion empirically and am just speaking from experience and drawing on that of other experienced players. I can't claim either to have played many hands against 'very very good' LO8 players (I play online only), I've rarely come across players I judged even deserved one 'very' and I think I'd find another table if I really felt my opposition really excelled at the game.

jj--

Good to have others chime in to this thread.

I certainly didn't mean to suggest that Buzz offered any kind of analogy between O8 and LHE in defense of his claim. That was entirely my doing. But I think it strengthens certain intuitions that let us understand the situation better.

I also agree that this was something of a quick side comment, and that Buzz wasn't trying to be controversial or do any heavy mathematical lifting. However, I do think that being so tight from the SB is a significant leak, for all the reasons I mentioned... limit O8 is a positional game but not nearly much as a game like NLHE, in which it's frequently correct to make immediate pot odds play a very minor role in one's decisionmaking. (For whatever it's worth, this is something I always found mystifying about the Harrington books, at least the first volumes. Dan is a smart guy, and of course he doesn't think that preflop pot odds are the whole story, but he sure emphasizes them.)

Oh, and as for why I was playing against experts: in mid- / midhigh-limit O8 and mix games, in live settings, even very good games often include a few of the best mix-game players in the room. This can be an excellent chance to learn a thing or two while experiencing a very good earn.

All my best,

--Nate

[jj179]

Hi Nate,

Well I'm grateful to you and to Buzz for starting this debate in the first place, it's always refreshing to have one in which ego and the need to be correct are not the true reasons for the dialogue and where we are all open to changing our positions.

Actually I think the reason you and I disagree on the play here could be at least in part due to where we play. I never play live and if I did I would certainly play higher than online, where I can keep loading tables until concentration or reads are an issue. That would certainly explain why I rarely encounter great players as no doubt the online experts play higher, and I also switch tables when they are not quite as juicy as I'd like. That being the case, I can only make some guesses about how some of the issues we've mentioned will differ from your experience to my own, the reason why we genuinely like a different play in the same spot.

One factor would be in, my opinion, that the penalty for playing OOP live is probably smaller than online. From past history I've found the small blind to be the place where you are most likely to get drawn into an unexpected (and inexplicable) raising war where, having made the initial marginal call you later get stuck for many more bets you have become forced to call. I'd guess these wars are less frequent live for fear of the participants making themselves look idiotic, and easier to avoid in the first place given the extra cues of live reads.

I also think that live players are far more likely to notice if you are playing super-tight and that would detract from an ideal table image - from what I've heard you don't want to seem overly serious or focused entirely on winning. It could well be though that there is a right or wrong answer here but I can't think of a theoretic proof, and being a bit on the lazy side with regards to tracking my LO8 games I can't offer empirical evidence either. I'd be interested to see what you make of the game online though and whether your preference changes, I'm personally thinking if I do play live your looser approach might indeed be better, the reverse being true online.

[Buzz]

Hi JJ and Nate.

I feel that I badly botched this thread.

And it's at least partly because of the way I wrote the sentence in the article that led to this.

If I had just included the word "usually" and written, "Don’t usually even complete unless you have an excellent starting hand." then the sentence would express my opinion. Rightly or wrongly, that's my opinion. But that opinion is based on typical fixed-limit full-game play, as I encounter it, in brick and mortar casinos.

Rightly or wrongly, I don't think it's usually worth while, in a typical fixed-limit Omaha-8 full-table casino game, to complete from the small blind just because you seem to be getting juicy odds. I think those odds are very deceptive and a lure, especially for loose players. I regularly watch opponents completing from the small blind and then getting in big trouble with hands they shouldn't have played to begin with.

Different perspectives may lead to (at least slightly) different opinions.

Thank you both for your excellent, well-written responses.

Buzz