PokerStars.com Poker Stars Pokerstars.net

Omaha Strategy Tips

Rec.Gambling.Poker Posts

Raising Before the Flop in Omaha8
> John wrote...
> > I know people who never raise pre-flop and they do very well at O8.
Sean Duffy replied...
> It's possible to do very well with a major leak in your game.

Mary can still win if she never raises before the flop if she plays good starting hands and very well post-flop -- and her opponents are absolute dopes. But never raising before the flop is a sign of a bad poker player. If that bad player plays with *worse* players, then that bad player can still win. Omaha is to a large degree about starting hands. To state the obvious, starting hands exist before the flop. Putting in more money when you have a good starting hand and when your opponents have crappy ones is basic Omaha8.

Some people don't think starting hands matter in Omaha8, and God bless them. Also, many mediocre players who do understand Omaha is about starting hands don't "get" that starting hands only exist before the flop. If most of Omaha is starting hands, then aggressively betting your hands before the flop should be an obvious conclusion.

Sean Duffy wrote...
> In a typical low limit game I think failing to raise before the flop is a huge error.

To say the least. I don't expect to get a hand this good (AA45 double-suited) in any hour I play online, or two in a casino.

> Your low possibilities aren't great, but who wants to win half a pot?

Low strength is half pot strength. It's not the reason to play a hand. The low strength of AA45 is underrated though. It's as if people think there are a jillion low cards out there... if it comes an Ace, fine. If it comes 876, ok. If it comes either a deuce or a three, fine. What is the problem? If it comes 874, fine, muck if you want.

What does anybody out there who doesn't want to raise think those three limpers... or *anybody*... has that is better than AA45? The reasons to raise are many, starting with you have a much better hand than at least 2/3 of the current opposition, and including that you want to at least *try* to get rid of 23. Of course anytime a sucker wants to call a raise with 23 that is fine too. And if you got a sissy willing to muck A3, that's fine too.

It wasn't clear to me that Tony was asking about As4sAd5d, but if he was, in a nine-handed showdown simulation (10,000 hands) this hand scoops 1187 of the time, gets high 2219, and gets low (including scoops when there is no low) 2970. The random hands win something like 465/975/890. These are just showdown numbers, but it should show this is a fine hand.

A lot of players seem to think that to bet one bet you need a hand of X value, but to raise you need a hand of 5X value. They also seem to not think about what the other players have.
###

Monsters Under the Omaha Bed
Lee Munzer wrote...
> In the low limit games, approximately 1/2 my opponents will only raise with
> premium ace type hands, thus I can find better places to invest my money
> than to call or reraise with AA76.

You are making the argument that you would fold AK in a Holdem game because two opponents raise with AQ. AA76 becomes much better because of the two raises. In fact this is *ideal*. Your hand almost certainly dominates both opponents. There just are almost no better Omaha8 situations than this.

> What I've found is unless someone like you or Men "The Master" is
> playing in my low limit Omaha8 game, when two players raise, I can
> read at least one of these duel raisers for a premium (ace containing)
> hand -- if not both more readily and confidently than I can isolate raising
> hands in low limit HE, where, for example, two early raisers can hold 99
> and QQ. Thus, the AK is "live".

You are making my point. If both players have an ace, both are in deep crap against you. You, like most people, seem to look at your hand and somehow don't seem to think about what the other players have. What four cards are you giving these two players that make AA67 double suited a dog? If both have an ace, our hand is very dominating. This is good. This is what you want. You want to have AA against two players with dead cards in their hands. The one confrontation you fear is something like AA45 and KKQQ. But, if both players have an ace, say As2s3h4h and Ac2cKdKh, we beat the tar out of them. Also, there are the two blinds to chop up. You can manufacture scenarios where we are in trouble, but there are very few, and even the worst case isn't awful.

AA67 double suited LOVES to play against two raising hands *because* they have aces. This is a good thing. I guess you didn't get the Holdem example. Here is it more obviously: when you have AA in Holdem, you love to play against a raiser with AQ and a reraiser with AK. It is not bad at all for you that both opponents have aces. It is terrible for them.
---

Lee Munzer wrote...
> Badger please run "Poker Probe" on:
> As2s3h4h
> Ac2cKdKh
> Ah7hAd6d
> I'd take a wild guess at something like 34%-27%-39% (respectively).

Hand scoops high shares low share percentage
A234 12,985 23,616 37,866 30.74
A2KK 19,690 28,699 35,628 32.16
AA76 25,603 47,686 26,507 37.10

> In the quiz example it's not clear whether there will be three, or perhaps,
> four/five players taking the flop. Yes, if two opponents hold hands like As2s3h4h
> and Ac2cKdKh and they are your only two opponents, I like Ah7hAd6d.
> So, you're right! If I'm close, you'll make approximately
> 18% on every dollar you invest.

You should love this hand in the situation you weren't liking it... against two quality hands that typical players would raise and reraise with. In the 100,000 hands above, it wins in half of them. The other two hands get clobbered since their "low shares" are quartered except those times it comes a deuce. The Probe percentages can be misleading if you don't recognize that the low shares are commonly money-losing shares.

> Now, let's "devil advocate" the situation to see how much of a favorite
> we are when there are five players (the other two defend blinds with
> Jc10c9d7d and KsQs8h8d). I realize there is a good chance a reasonable
> player would muck these blind hands in the face of heavy raising.

Another thing to consider in plugging these probe numbers in is, yes if you add a hand like the JT97 that happens to have a richly favorable deck available, that crap hand now can becomes a favorite. And, the JT97 gets most of it's value at the expense of the AA76 because it gets scoops and high shares of the pot.

But before people start running to play this crap though, they need to consider that if their crap is not taking up virgin territory -- for instance, if the AKK2 was AJT2 -- the crap hand is pitiful. AA wants these randomish crap hands out. If they play, they might have the best of it, but they might be enormous dogs.

You can tweak these hands many ways... turn the AKK2 into AK23 or AKQJ or whatever.... and then add randomish hands and get wildly different Probe results, some of which make the AA67 much worse, some make it better.

I think though you have to look at this mostly from two perspectives. First, the AA67 loves the confrontation with two good hands. Second, AA67 double-suited would LOVE to play against a whole field of random hands every time. A less important thing is the AA67 can face some situations where it isn't nearly as good, or even any good at all. To oversimplify it, the AA67 is a bit like QQ in Holdem. You almost always just play it and have a solid expectation, but once in awhile you are a dog (though the AA67 double-suited will never be as big a dog as QQ will be to AA or KK).
---

Jonathan Kaplan wrote...
> Does the dominance (three-handed) of Ah7hAd6d over As2s3h4h and
> Ac2cKdKh require BOTH other hands to have A2 exactly? What if the 2c
> was the 3c, how much difference does that one change create?

I don't see that making a difference. The AA67 was getting low about 1% of the time, when it came a 3, a 4 and one of the two remaining deuces. If you change one of the deuces to a 3, it just makes one of the two other hands worse, and doesn't hurt or help the AA67 any (except that when there are two ace deuces, rather than an A2 and an A3, there is going to be a bit more betting for the AA67 to have to face).

> Of probable played opponent hands, do you feel that the AA67 hand
> will "get better" a larger or smaller part of the time, or stay about the
> same, in an "average" $10/20 Omaha8 game? If the word "average",
> or the limit specificity means anything. (Just curious how much variance
> there is going to be, if one can't specifically conclude that the opponents
> hold two such pretty hands, simultaneously.)

If two players hold such pretty hands, I would imagine adding any more players would hurt the AA67 -- except players with KKxx and 2345 hands. The configuration of the Aces really hurts the A234 and AK22 against the AA67, but with more players in the pot than this, the AA67 is hurt by both aces being dead. ANOTHER consideration though is, any player who would jump in cold for four bets with any hand except four picture cards is likely to be such a very poor player that even though his hand is in good shape with a rich deck, he is likely to play the situation rottenly.
###

Omaha HiLo Issues
If your Omaha edges are usually only small, you are playing too many hands. I think it was Sean Duffy who wrote that playing Omaha8 becomes tolerable online because the game is so much faster. I think this is an excellent observation.

Spencer Sun wrote...
> He certainly has calling odds on the turn, but I'm not sure about betting.
> If last to act (i.e. "only" 5 opponents) I'd be inclined to take the free card.
> The flush is only good for half the pot, and while A3 might be good and
> a deuce might come, I don't think that's enough to bet for value.

There are a few reasons why betting is likely right. Most important, players in this game are horrible. They are capable of making a ton of errors when someone bets (that they can't make for a check). For example: A dry A2 might checkraise, driving out straights that have Sean tied... A nut straight might checkraise, driving out any other A3, giving Sean a terrific freeroll... A bet might freeze an A2, whereas a a check will bring the A2 to life on the river.

Sean's bet is correct because of what actually happened: he got to see the showdown, with a decent hand and a nut draw, for one single bet (with the option to have bet the river if he made his hand). If he checks the turn, there will almost certainly be one or two bets on the river, which he should call both because the players pay so horrible and because the high hand is so easy to make and be bet.

Bets like this on the turn make money even when he loses. He saves bets when he misses his hand, but makes quite a bit more when he makes his hand. This is the type of hand I love to say thank you for online. To oversimplify it, any time I got (a losing) second nut low on the river and it's checked around, this is a good thing.
---

Edward Hutchison wrote...
> I did make a quick check of AA76 (where both aces are suited) using
> Mike Caro's Poker Probe and found that the hand has a win rate of
> about 16-17% in a ten handed Monte-Carlo simulation. As the cut-off
> for play under my system is 15%, this hand would qualify for play.
> I will, therefore, give the matter some study and at the risk of complicating
> the system I will amend the posted version (again) making some minor
> award for the A-6 combination.

It's not a flaw exactly, but an inevitable problem with something like Poker Probe is assigning winning percentages in HiLo split games. I just ran this hand, and it's assigned a win percentage of 17.05%. In a run of ten random hands they would each get a value of 10%. So 17% is way above average.

But what does that 17% represent? In 10,000 trials it's 774 scoops, with 2124 hi shares and 1286 low shares. (When a player scoops and there is no low, the program still assigns the low share as won by the scooper.) The average statistics for the hands are 458 scoops, 904 high shares, 868 low shares.

What people should be taking out of these numbers is against a full field of players going to the river, the hand scoops the pot 7.74% of the time.

AsJsAdTd is either the best pure high Omaha hand or the second best (after AsAdKsKd), depending on how you measure things. This hand scoops at almost the same rate as As6s7dAd. *And* given that AA76 can often also win low, it's win percentage is then higher than AsJsAdTd, which clocks in at about 15.42%.

Scooping is how you win money in HiLo games. AsAd6s7d is an absolute powerhouse in any sort of game, especially very loose or very tight ones. I don't see it as at all "close".

And it's a much better hand than As2d3c4h, which scoops a pitiful 3.28% of the time, but has a win percentage of 24.3%. The win percentage number is not the important one.
###

Online Omaha Games
plplaya wrote...
> I sat down in a paradise 5-10 Omaha8 game this morning and the average
> pot was $50, with flops seems at 35%, would you sit down at this game?

Gleefully.

> I have never seen an Omaha game this bad!

Why do you call it bad? It's nearly ideal.

> It was like playing mid-limit Holdem where post flop is almost always a
> heads up or 3 way situation. I got A2Q3 double suited on the big blind
> hoping to raise it up, and everyone folded to me!!! I left after the only
> chaser/fish I could identify got busted.

Fish come in different forms, and the chaser isn't even the preferred.

> Am I correct in a assuming this is an unbeatable type of Omaha game?
> When does an limit Omaha8 game become unbeatable?

If you can't beat this game for a LOT, your Omaha game blows chunks, period. Again, these games are almost ideal. A bunch of passive players who misplay 80% of the hands they play, plus one or two loose "play crap cards" liveones.

> When I play California limit Omaha, it's always 7 way action preflop,
> with 5 seeing the river (an impossible game to lose long term at).
> I never have encountered a game like today's 5-10 at paradise.

The California games are fine too, but they are looser and more aggressive. Bad players include chasers and aggressive bettors, but very few pansies. The online Omaha games have some of the most passive, weirdly backward-playing just awful players that God saw fit to create. The betting strategy employed by most of these players is as bad conceptual poker as I've ever seen.

These games are very different than California games, and it can be a jarring thing to see, but they are much softer (I know that sounds impossible), have absurdly low variance, and are extremely profitable. The Paradise 3/6 is only a middling game (usually), but every other limit at the online cardrooms I've played at offers about the best earn I've seen in any limit games I've ever played in. Of course, you have to play properly, but that is a different issue.

Stephen Jacobs wrote...
> I'm real interested in what playing Omaha properly means. I gather
> from your posts that the first factor is playing the correct starting hands,
> with absolutely no compromise (that should be no more than 1 voluntary
> flop in 2 circuits).

Not at all. Online Omaha games offer the opportunity to push the envelope of playable hands. Given the passivity of the games, if you play a little under 1.5 hands a round in the blinds (usually getting a free ride in the big blind) and a little over 1.5 hands a round outside the blinds, you are doing just fine.

> Even then, you seem to recommend mucking if anyone else might plausibly
> have liked the flop more than you do. I get the impression that you like to play
> straightforwardly when you do play, but I don't remember anything specific.

There is a big difference between the sort of loose California games the original poster described, where the game mostly is "the best hand wins" and online games where if you say "boo" half the players fold hands that should not be folded. Straightforward play should not be used online (usually). Most players are clueless about the action, in large part because many are playing two games, so it is important to not be obvious in your actions. In other words, loose-aggressive game Omaha is very different than tight-passive game Omaha.

> So let's try an example: AQQT (suited once--you're invited to comment on
> where the A vs non-A suit makes a difference) in late-middle position with
> an early limper. Raise, fold or call?

Normally I'd call. I want players. I would raise specific players and situations. Folding is silly.

> You see the flop with those cards 3-handed. Comes A97 rainbow,
> one of your suit. Checked to you (we're giving you weird passive
> opponents). What's the plan?

Check.

> Might you ever continue against an A86 flop?

Almost certainly not, not in such a small pot.

> I'm probably your ideal victim now, but I want to learn.

An ideal victim sends money directly via PayPal.... :)
---

plplaya wrote...
> Which is then?

Are you asking what is an unbeatable game or the preferred fish? People who fold when they are getting excellent odds to call or raise are preferred to opponents who chase with marginal pot odds. People who raise and bet idiotic draws head-up (like raising one player with 2347 offsuit on a As8Js flop) are better to play against than those who call with that hand.

> I just had never seen an Omaha game that tight.
> What adjustments should I make to beat it?

Think of AK more in terms of its Holdem value than its value in a loose Omaha game.

> In this game, since people would call preflop raises, I raised when I could
> with good hands, normally I limp along like everyone else in the CA style
> game as to not raise suspicion of an A23 being out. I want the A-3, and 23
> calling me on the turn and river.

A23 hands are rare so aren't a terribly important aspect of the game.