Wednesday, August 24, 2005

Wholeness and the Implicate Order

YTD: +$19784.89

Imagine you are on the turn of a PLO hand against two other players, A and B, to be original. There has been a some action on the flop, say a bet and some calls, and the turn presents you with the delight of a nut flush draw and top trips, and you are last to speak.

(Clearly this is a fictious example :-)

However, there is also a clear straight visible. The pot contains $100 dollars and you and your foes all have $1300 left.

What is your strategy? Assume A and B are not idiots.

What do you do if A bets out and B raises?

Is this any different from A checking and B betting?

If you could choose a betting order, what would you choose?

And lastly how would having bigger stacks change things?

35 comments:

Anonymous said...

A very short question, so I'll restrain (for the moment) from a very long answer. This will probably lead you to asking me "why"? And I may, or may not, elucidate at a later time.

If A bets and B raises, I reraise all in

If A checks and B bets, I flat-call or perhaps put in a mini-raise.

If the stacks are significantly bigger:

If A bets and B raises, I mini-raise.

If A checks and B bets, I put in some kind of medium raise. Not the max, not a mini-raise, but somewhere in between. It would depend on what level of tolerance I thought A and B possessed for raises.

Pete

Anonymous said...

The ideal scenario would be to get everybody all-in (giving you odds of 2/1).
So your basic strategy is to keep everybody in for as large a bet as possible.

Doing some figures on twodimes, against two opponents with the made straight, you’re between a 42% and 47% chance to win the entire pot (say 45% on average).

However getting all-in against one opponent you’re still about 45% but now you’re only getting just over even money which is very bad.

(i) so if A bets and B raises, then if I reraise A might fold if he has only the bare straight. This leaves me all-in against B when I am only 45% and getting just over even money.
So in this situation I just call B’s reraise, and hope that A will reraise all-in, or if not at least call.

(ii) If A checks and B bets, I will not reraise because of the possibility of A folding, and then B reraising me all in.
So here also, I just call. Again I hope that A will reraise, or if not at least call.

(iii) If I could choose a betting order, I would choose the first one
i.e. A bets and B raises. This gives a good chance that both have straights, and so a possibility that everybody will go all-in, or if not the strong probability that A will call B’s raise after my call.

(iv) I don’t think that having bigger stacks would change things too much. I will always just call any raise trying to keep everybody in.
The exception to this is if for example, A bets, B raises, I call, A reraises, B calls. I will then reraise again if I think that both A and B will stay in. But I won’t reraise if I think that B might drop out if A puts in a further reraise, as this might reduce substantially my good odds. I’m on slightly shaky ground here.

Anyway that’s my opinion. Let me know where I went wrong.

Kevin

Anonymous said...

Small stacks

A raise, B re-raise. Flat call, for me. Nuts str8 highly likely here, unlikely going to pass. As others have stated, you want dry str8 from A in or dominated draws.

A check, B bet. Raise max. A unlikely to check str8, chance B stabs at the pot.
Next best flat call to encourage dominated draws. Don’t like mini-raise since it will discourage dominated draws, but not dry nuts and you could get in a tangle.

Larger stacks: More passive I guess, aggression may lead you to pass if aggression is met by greater aggression.

Position: I suppose, when considering the value of position, you should consider the river too. I’m not sure I can see too much value in position wr.t. river: no chance to bluff; flush comes - non-nut flush unlikely to bet/ or call reraise. Similarly, for full houses.

Generally, I want the likely weaker (likely) drawing hand on my left so I can give the pot odds the player needs. It may also give me some shot of making extra on the river, I suppose.

Anonymous said...

I'm with Kevin on this one, his reasoning looks excellent.

Big Dave D said...

As a teaser, I think the ideal betting order is not one of the two i detailed.

gl

dd

SimonG. said...

Who said there was a made straight out there? :)

And if Dave named a sie this happened at how much would that affect your play?

I can think of a site where if the player is going to call for a dollar they will call for it all. When that is the case, it limits your thinking opportunities and therefore limits your opportunity to improve as a thinking player. But it's still a fucking great game to be in!!

Anonymous said...

I didn’t do enough twodimes examples above ! Against 2 opponents I said that you were between 42% and 47%. I forgot to include some of my suits in the opponents hands and also that they may have paired some of the board cards ! So you’re between 31% and 47% (say 39% on average).
Also against one opponent I said you’re about 45% on average. You’re probably more like 37%.
My excuse for this sloppiness is that I’m at work !
Anyway this doesn’t change my play – it just isn’t as good an overlay as I had thought.

As to the ideal betting order, I give up.

Kevin

Anonymous said...

I must be missing something here. Dave usually asks interesting questions and PeterB usually says something sensible.

But in this case:

A bets, B raises - I call.
A checks, B bets - I call.

Why? Well, I'm unlikely to be winning. I want to play it multiway (for value now, and to increase chances of getting paid on the end). And I want some money left to bet on the end if I hit.

If I could choose a betting order? A bets, B calls, I call. Much bigger chance of a call from B on the end if I hit.

If the stacks are deeper? No difference.

Deal'em said...

I want both guys in for whatever new money invested on turn. If successful, I get 2 to 1 odds on new money in versus draws to nuts 1.5 to 1 against. If A bets out and B raises, I smooth call if stacks deep (still a probable dog if it gets down to heads up and I want a shot at big implied payoff if my card comes with both guys in)and maybe smallish raise with 1300 stacks to build the pot.

I guess the betting order I would prefer is to be in the middle of A and B. If A leads out, I smooth call and hope B likes odds to continue on or even raise. If A checks, I lead out with medium raise.

Big Dave D said...

Just to further clarify the betting order thing, i meant that it wasnt one of the ones i had listed, but u still are last to speak...which makes it obvious really.

Check-check to you.

But why?

Anonymous said...

I think folding is also an option here, not a good one, but an option all the same.

Didn't have anything of value to add to the discussion, just wanted to join in.

:-)

Anonymous said...

Regarding the anonymous poster who said folding is an option, you have clearly met one of the most timid weak-tight omaha players you will ever meet. Whenever you can have a 3 way situation in which you have 45% equity, you should be looking to play that situation as many times as possible and for as much money as possible. If you are afraid to make these monstrously +EV plays because you fear the potential variance when you don't hit, then you have no business playing plo. Or poker for that matter.

The only question here is what action allows you to make the most money in this situation. If you can manipulate the action by min-raising and inducing a reraise that allows all 3 players to get allin I would do that. If I thought I might be chasing out a player would would be giving me more odds on my money when I am currently beat, then I would just smoothcall. In the betting sequence that goes A check, B bet, I would thus probably just smoothcall because if A doesn't have a straight or another draw he is willing to call with, then he will fold anyway and you will just be getting more money in against B when you are beat, although doing such does guarantee getting paid if he is tight enough to fold if the flush comes or the board pairs.

Glad to see you seem to be doing a little better Dave. I've seen you at the tables a couple times this week, and the action has been good on both party and stars. It's also ok on ub if you like playing real short.

Your fellow countryman ribbo sure is a prick isn't he?

BluffTHIS!

Big Dave D said...

Bluff and Gerg - let's play nice with the newbie anon guy...he might be new to PLO too...this isnt 2+2 u know :) Let's only insult people we know...hey where's pete fab when i need him!

But let me say, passing is clearly a monsterous error.

Bluff...i hate the fact you know who i am but i dont know who r...go on, gimme a treat!

I think my views on Ribbo are well known. What I would say is that he is much milder than he used to be.

gl

Dave

Anonymous said...

with stacks like this, i call no matter what, because my money street is the river. with much bigger stacks i put in at least one raise, for the same reason.

Anonymous said...

by the way, "fold equity" is a stupid term. it doesn't exist.

Anonymous said...

to gergery,

no, getting everyone all-in is profitable, but is far from ideal. ideally, you get everyone all-in only those times when you end up with the best hand. remember, this is a multi-street game.

steve said...

I'm not gong to get into the omaha thing except that I'm surprised that 45% is brandished around so much. Surely it's going to be closer to mid thirties - once you discount the likelihood of your opponents (& you) holding some of the 45% outs.

AKsu, you do make me laugh, get yourself a blog.

re the newbie: it's a blog not a ring game you know, there's no need to merciless exploit a fresh who simply wants to put in appearance and learn. Don't be so dramatic, lighten up with the: ' if you think this then you've no business playing poker..'.

Of course he's got business playing poker, it's not an elitist game - it can be fun.

Be nice.

Anonymous said...

Sorry Dave, no treats ( :)) ). I have gone to great lengths to keep my screen names anonymous, and thanks to relatives with different town names have multiple screen names/neteller accounts to fool my regular opponents. And you have only yourself to blame for conclusively spilling the beans by posting unconverted hand histories in the past, and for giving the hint in your profile w/2 city names that allowed me to identify both of your party accounts. Nonetheless, because you do typically play in a very aggressive style, your hands often aren't that easy to read.

BluffTHIS!

P.S. Saying that ribbo is milder now is like saying that 3 day old shit don't stink as much.

Anonymous said...

Dave says that check-check to you is the best betting order.
If that’s the case, then you cannot call, because then bet-call-call would be better EV.

So check-check to you, and then you bet the pot. But then if A reraises pot i.e. makes it 400, B will probably fold and you will call (or A folds and B reraises pot).
So how is this better than the A bets, B raises, you call scenario ?
Is it because of the possibility that both will fold to your turn bet ?
I still reckon A bets, B raises is better EV.

So Dave, when are you going to share your thoughts with us ?

Kevin

Anonymous said...

Check-check to you is better because it indicates you are unlikely to be up against a made straight. So you bet, and if you are called they are drawing pretty-much dead... And they'll probably have a stab at it on the end if they hit.

Back to the original question. A bet and a raise with all that money left indicates, in sensible company, that A has the made straight and that B has the made straight with a redraw. So you've got a couple of outs missing. Lets say the equity is 30%, 30%, with 40% for you.

I'm not a chicken but I don't see the need to attempt an all-in coup at this stage -- for a start it might not happen and you're left holding the wrong end of it, and secondly, what is is worth?

There is $100 in already. OK, we get it all in. So there's now $4000 out there. And you have 40% = $1600, or +$300ev on your turn bet.

But if you just call the bet and raise (both pot-sized, and the original bettor calls), then you have $1300 out there on the turn, or +$120ev.

So, by avoiding the all-in, you are giving up $180.

But if you take it easy and see the last card, there is plenty of scope to recoup that if you hit. If a bet of $900 stood a 50% chance of being called on the end for example, then there it is.

Anonymous said...

Dave/chaos,

I think folding is also an option here, not a good one, but an option all the same.

Thanks for dousing the 2 flamethrowers.
It still amazes me that some poker people don't do irony, to me its an essential part of the game.

"Irony is a form of speech in which the real meaning is concealed or contradicted by the words used. Irony involves the perception that things are not what they are said to be or what they seem."

Some great ideas here on how to play this one, without going over it again I would try to play it like Kevin.

:-)

Anonymous said...

I don’t agree with JQ.
He states “Check-check to you is better because it indicates you are unlikely to be up against a made straight. So you bet, and if you are called they are drawing pretty-much dead... And they'll probably have a stab at it on the end if they hit.”

Suppose the board is 9 T 4 6 with 2 hearts and you have 3 tens and nut flush draw. You opponent could have a big wrap like K Q J 8, so how is he drawing pretty-much dead ? He has from 12 to 16 outs.
I agree that check-check might usually indicate that you are unlikely to be up against a made straight, but in my game, it would not be unusual for A to check the made straight in the hope that he could check-raise. I’m not arguing for or against this move, I’m just saying that you can’t discount it.
And if say an offsuit Q comes on the river, they will “take a stab” at it. They may even do it with the Jack blockers.

Anyway as Jq states if you can get it all in with 2 opponents your EV is $300.
Or $120 if it goes bet, raise, call, call.

I would take either of the above over the check-check-bet option, because in this case (i) you may get reraised by A, (ii) either opponent may call and bet to you on the river when you haven’t filled up, (iii) I don’t think there is much scope to recoup if you do improve.

Kevin

Big Dave D said...

Pete,

Glad to see I provoked you out of the sidelines :)

Cant think of any Simpsons jokes for now!

Interesting points, although I don't necessarily agree. I think a key point was that I said these players were not idiots. Your chance of getting paid off on a good river are slim.

Care to expand on not liking the term fold equity?

gl

dd

Big Dave D said...

Aksu is funny :)

Its watching all that English comedy.

Chaos, you would be surprised what twodimes shows...40% plus is very likely.

Bluff, I seem to have changed style a bit...tight passive preflop and my usual LAG on the flop. Seems to work well in the 2-4 and 3-6 as raising preflop doesnt isolate as much as it does at 5-10 and 10-20.

gl

dd

Big Dave D said...

JQ,

If you tweak my equity up just a little, then the EV hits $400. As I said, these players arent idiots. I think the chances of getting much out of them on the river if you hit is tiny.

Kevin,

It was me that said check-check was the surprising better betting sequence. But I will leave why to my "answer". Aksu got it right by answering the wrong question :)

gl

dd

Anonymous said...

Kevin -- You're right of course, I had been assuming a top straight for some reason.

Dave -- True, you can get up to $400. But you still have to engineer the 3-way all-in which I don't think is trivial. If you put in the 2nd raise there is a real risk of bet-raise-raise-fold-raise-call, whereby you have put $100 in 3-way with a further $1200 on the side against player B whilst getting much the worse of it.

So I definitely check.

If I was given a choice between a guaranteed 3-way all-in on the turn versus seeing the river with money to bet, then, yes, I would take the all-in.

Are you going to put us out of our misery?

Anonymous said...

Dave gotta dash (live poker!), but does 2 dimes go all Bayesian? Surely in this instance the chances are high that your opponents (as well yourself!) hold a number of your outs - if they've got wrap arounds, sets, flush draws etc they're causing you grief - you're can't be assessed as, say, X/44 (X your theoretical outs), surely?

chaos

Naturally, though, I give deference on all things omaha.

Anonymous said...

Some brief reasons for my own thought processes (many many posts ago).

My first thought was that, if you hit, these people are not going to pay you off. So, I want to get both people in for all my money on the turn.

But if I fail, getting one person all-in and the other person folding is not the end of the universe. It's turned a good plus EV into an only marginally minus EV. I think that the raise has a better EV than calling, hitting, and then not getting called on the river.

My second line of thought was that people have a lot of trouble getting away from the nuts, even the bare nuts, in a three-way hand. If A has the straight, most PLO players (even good ones), seem to find it hard to walk away with one card to come.

My third line of thought was that, if I reraise all-in, what will A think if he has the straight? My guess is that he would place me with the straight as well, and B with the set. Or he might think that we all have the straights. But we do not know that A has a straight, and we do not know, if he does have the straight, that it is a naked straight. The worst scenario is if A has a lower set and B has the straight. But would A bet out with the lower set? I wouldn't have thought so. Any other combination, and I see A calling your re-raise. Since we strongly suspect that B also has the straight, he is now in the situation where he knows what is happening, but he is still getting the odds to call.

With the "bigger stacks behind" scenario, things are tougher. I suggested the mini-raise here because I thought that this was the best way to show a hand that was trying to keep the pot as small as possible (whereas in fact you want a three-way pot as big as possible). I'm not too sure of the follow-through scenarios here and there's a lot of fuzziness about the whole situation. In fact, I doubt that you can categorically state that there is a "right" answer, unless you can predict with a fair degree of certainty what the other two participants will do in a wide range of possibilities. And, if you can do that, you should be playing at far far higher stakes than either of us can dream of. Players act unpredictably. In this kind of situation with big stacks, they are likely to act even more unpredictably. It's a lot easier to see how the hand is going to pan out when the opponents have smaller stacks.


Pete

Anonymous said...

Against non-idiots and idiots alike, the river is still my money street. Consider, there will only be considerable action on the river if:

A) One or more of these players has the nut straight now and a blank comes, or

B) One or more of these players hits a flush or better

C) One or more of these players tries to represent a flush or better, or

D) The river is a blank and it is checked to me.

If they are clowns ahead of me, all the more likely they pay me off when I hit and they don't. If they aren't clowns ahead, all the more likely they pay me off when they don't hit and I do.

And fold equity is oxymoronic. Calling or raising are (almost) always positive equity plays, and the relevant calculation is whether your equity exceeds the cost. Folding is a 0 equity play, and should only be done when your positive-equity plays are all unprofitable. (Certain game-theoretic possibilities excepted)

The one time the term makes a little bit of sense is when you're talking about a situation where most people incorrectly call, like with KQo in the BB when UTG has raised and it is folded around in a typical 20-40 LH game. But even there it is just a convenient misnomer.

Big Dave D said...

Pete,

The key phrase, as Gerg points out, is that I did say upfront that these players are not idiots. So although the chance of them passing a bare straight is small, it is real. And the chance of them paying off when you have absolutely acted like you are drawing is small again.

I use fold equity in another sense. Basically I take it to mean the extra equity you can get from a raise or bet play when a player may fold his hand, this winning the pot by default. His fold has increased your equity.

Gerg, I've actually got an example of a tourney hand where I pass AA when I am first to act! A later entry though.

cheers

Dave

Big Dave D said...

Pete B,

I will cover off most of your points when I finally write up the "answer". What I would say is that,to quote :"But if I fail, getting one person all-in and the other person folding is not the end of the universe. It's turned a good plus EV into an only marginally minus EV." To my mind this is a disaster. Allin coups are all about EV. I *could* be getting as much as $400 for my $1300 bet. Turning this into a negative number is "the end of the universe", PLO speaking.

gl

Dave

Big Dave D said...

Chaos,

You can just run multiple twodimes sims! Run a very favourable one, a medium one, then a very bad one.

Enjoy that funny live stuff.

gl

dd

Anonymous said...

First off, who says Big Dave is a good player? Second of all, it is rarely a wise play to try to get someone to lay down the nuts in this game, particularly on the turn. Among other reasons, it is too likely they have a redraw themselves -- either two pair, a small set, a non-nut flush draw, or higher straight possibilities.

And again, I think you are missing my point. I am not suggesting that a bare straight will pay you off when the board pairs or a flush hits, but a smaller full house or a smaller flush usually will. When was the last time you saw someone check-fold a made hand they hit on the river? A bad player will check-call. A good player will bet-fold. Either way you get paid.

And ok, I withdraw the objection to fold equity. See, I'm not always snarky.

On a side note, this whole conversation reminds me of one of the main problem with weak poker players, which is a failure to risk one profitable situation for a better one. A classic example being in nl hold'em when you have a set on the flop against an aggressive player with a likely flush/straight draw. Most players just always raise to "protect their hand," end up with all the money in, get sucked out on, and pat themselves on the back for getting it all in with the best of it. Better players will sometimes risk getting sucked out on and wait until the turn semi-bluff to pounce. Like the PLO play in question here, defering action until a street where the money is bigger and your hand is better defined can be riskier but is much more profitable.

Big Dave D said...

Pete,

You should stop by more often...just for you:

Hutz: Mr. Burns, we've got witnesses, precedent and a paper trail a mile long.
Burns: Yes. But I have ten high-priced lawyers.
Hutz: Ya, ya, yaaa!!! [runs out of office]
Homer: He left his briefcase. Hey, it's full of shredded newspaper.

(Don't change your degree, or I run out of jokes.)

I understand your argument but I'm not awfully convinced. You have your equity now on the turn. You may be able to pick some of it back up again on the river, but that's a big if. You say it often happens they hit a smaller redraw themselves, and pay you off. That probably happens about as often as they don't have a redraw and can't pay you off, stupidity outstanding.

I'm no NL expert. Or PLO necessarily either. But I don't think the two situations you compared are analagous. In the NL one your are simply infront, and worried about being outdrawn and how to play a tricky river. In the PLO one you are only "infront" if you can get both players to go allin with you. Otherwise you are not even +EV. The beauty of PLO, and one of the reasons I've stuck with it, is that it is often much clearer how "right" decisions are. The game is built with that much more information in it. And I think its pretty clear that if you *could*, and here is the real problem, the best situation would be for everyone to be allin on the turn.

good to hear from you,

DD

Anonymous said...

just for fun: what's the worst equity you could possibly have in this spot? It's fairly easy to think of some 5-outers. (e.g. board of 2s 7h 8h 9d and hands Ah 9c 9s 3h, Th Jh 8c 8d, 5h 6h 2d 2c) Are there any 4 outers or less?