Limit O8 regarding when to draw to bare runner runner nut low.

[yukoncpa]

8 players. Absolute poker. Limit Omaha 8. Active Players in this hand are unpredictable, loose and bad.

I have Ah2d8c9d on button. Utg limps in, mid position calls, I raise, Blinds fold. Both limpers call. 7 bets in pot.

Flop is 6hQcKs

UTG bets, mid pos. calls, I fold with 9 bets in pot

When hand was over, UTG scooped with KhJc5d6d ( I did not make a note of the other persons hand). UTG flopped two pair and I was satisfied with my fold.

However I ran my hand and his hand through an all in simulator and for the flop of 6QK, I got the results: me - 23.72% villain - 76.28% giving me the impression that all I needed was 4-1 on the flop to check out a turn for runner runner. ( I realize that if I improve on the turn, I’m going to have to spend more money, lowering my implied odds, but of course, if I don’t improve I just fold )

Can this be right? How big in general should the pot be to peal a card on the flop for a runner runner nut low when you are last to act, and have no high or low backup to speak of?

[Buzz]

[ QUOTE ]
However I ran my hand and his hand through an all in simulator and for the flop of 6QK, I got the results: me - 23.72% villain - 76.28%

[/ QUOTE ]Yukon CPA - I get pretty close to that. My simulator (only 10,000 deals) has it 1076 scoops for Hero, 6501 scoops for Villain and 2423 split pots.

So if hero puts one more bet into a 9 bet pot, and that is the end of it, then
1076*9+2423*4-6501*1 = +12875
12875/10000 = an e.v. of +1.3 bets per try. But if Hero has chips left, that's not the end of his investment.

Are you going to have Hero fold if the turn card is unfavorable, but continue if the turn is a good low card, say the 3[img]/images/graemlins/diamond.gif[/img]?

If so the turn card will be unfavorable 23/41 and favorable 18/41.

Then Hero (as simulated with the 3[img]/images/graemlins/diamond.gif[/img] as the turn card) never scoops and splits 5021. Villain scoops 4979 and splits 5021.

Thus Hero loses the one bet 23/41, loses two bets 18/41*4979/10000 and wins four bets 18/41*5021/10000.

-23/41-2*18/41*4979/10000+4*18/41*5021/10000 = -0.561-0.437+0.882 = -0.116

And Hero ends up with a negative expectation. Seems like I wrote a recent article related to this for the 2+2 on-line magazine.

At any rate, the odds you seem to be getting for calling one bet are deceptive (unless you go "all-in"), because you'll get stuck for more.

[ QUOTE ]
Can this be right?

[/ QUOTE ]It sort of is. See above.

[ QUOTE ]
How big in general should the pot be to peal a card on the flop for a runner runner nut low when you are last to act, and have no high or low backup to speak of?

[/ QUOTE ]I don't know. I think with just one pre-flop raise it probably will be big enough to continue. There's also the possibility of running into another nut low who will be drawing for runner-runner, perhaps also with a better draw for high. That didn't happen in this simulation because we chose cards such that it wouldn't. But in a real game it would be a factor.

I believe the correct odds play, assuming no pre-flop raise, is to fold the low draw when the flop only has one low card. There are, of course some other considerations.

What about if the betting was capped before the flop and then you expected no more raising? In that case, I'd take off one more card. No question about it.

What about if there was simply a double bet on the first betting round? Without figuring it, I'm guessing then the odds would be favorable for taking off another card, assuming there would not also be a raise on the second betting round.

(I've seen good arguments for raising before the flop with A2XY hands that cite having more favorable odds to continue after flops with only one low card as one of the reasons to raise before the flop with A2XY hands, but then there come to be other considerations).

Buzz

[NickMPK]

[ QUOTE ]
[ QUOTE ]
However I ran my hand and his hand through an all in simulator and for the flop of 6QK, I got the results: me - 23.72% villain - 76.28%

[/ QUOTE ]Yukon CPA - I get pretty close to that. My simulator (only 10,000 deals) has it 1076 scoops for Hero, 6501 scoops for Villain and 2423 split pots.

So if hero puts one more bet into a 9 bet pot, and that is the end of it, then
1076*9+2423*4-6501*1 = +12875
12875/10000 = an e.v. of +1.3 bets per try. But if Hero has chips left, that's not the end of his investment.

Are you going to have Hero fold if the turn card is unfavorable, but continue if the turn is a good low card, say the 3[img]/images/graemlins/diamond.gif[/img]?

If so the turn card will be unfavorable 23/41 and favorable 18/41.

Then Hero (as simulated with the 3[img]/images/graemlins/diamond.gif[/img] as the turn card) never scoops and splits 5021. Villain scoops 4979 and splits 5021.

Thus Hero loses the one bet 23/41, loses two bets 18/41*4979/10000 and wins four bets 18/41*5021/10000.

-23/41-2*18/41*4979/10000+4*18/41*5021/10000 = -0.561-0.437+0.882 = -0.116

And Hero ends up with a negative expectation. Seems like I wrote a recent article related to this for the 2+2 on-line magazine.

At any rate, the odds you seem to be getting for calling one bet are deceptive (unless you go "all-in"), because you'll get stuck for more.


[/ QUOTE ]

I don't have an answer for the OP, but doesn't this shorthand underrate the value of the OP's hand for at least 8 low turn cards? A 3 isn't really what he's hoping for. Instead, if the turn is a 5 or 7, the OP has at least three (maybe four) outs to the nuts on both sides. Is this enough to make the expectation positive?

[yukoncpa]

Hi Buzz,

I’ve only recently taken an interest in O8, So I’m afraid I missed all of your internet magazine articles, unless you’re writing under a name other than Buzz.

Do you have a website that you’ve collected your writings on?

[Fiasco]

I think Buzz writes under the name "Frank Jerome" in the 2p2 magazine.

As far as the original question goes, for some reason I have the number 12 to 1 in my head as the odds you need, but if somebody told me I was wrong Id believe it.

[Buzz]

Fiasco - With 12 bets already in the pot, for a favorable low card on the turn and no chance to make high (not exactly the situation here, as pointed out by Nick)

18/41*4979/10000 and wins 5.5 bets 18/41*5021/10000, loses one bet.

-23/41-2*18/41*4979/10000+5.5*18/41*5021/10000 = -0.561-0.437+1.212 = +0.214

If we specified different cards for Hero and Villain, then I think the simulator results would be a bit different. With these particular cards, it doesn't look like you need even 12 bets. Looks like 11 bets would do the trick here. (And that's not taking into account the better chances Hero has if the turn is a seven rather than a trey).

Buzz

[Buzz]

[ QUOTE ]
Do you have a website that you’ve collected your writings on?

[/ QUOTE ]Yukon CPA - No. I just (finally) upgraded my computer. The old one I was using was fine for me but didn't interface with the newer modified systems. Does anyone else hate upgrading and feel used when forced to do so.

At any rate, the previous published articles are not even all on the hard drive of the computer I'm currently using.

(I suppose I could figure out how to transfer them).

Buzz
(Frank Jerome)

[1MoreFish4U]

[ QUOTE ]
[ QUOTE ]
Do you have a website that you’ve collected your writings on?

[/ QUOTE ]Yukon CPA - No. I just (finally) upgraded my computer. The old one I was using was fine for me but didn't interface with the newer modified systems. Does anyone else hate upgrading and feel used when forced to do so.

At any rate, the previous published articles are not even all on the hard drive of the computer I'm currently using.

(I suppose I could figure out how to transfer them).

Buzz
(Frank Jerome)

[/ QUOTE ]

Hey Buzz - you could just email them to yourself, or of course you could copy them to a CD so that you would have a backup of them and then you could load them on any computer you like.

[Buzz]

[ QUOTE ]
I don't have an answer for the OP, but doesn't this shorthand underrate the value of the OP's hand for at least 8 low turn cards? A 3 isn't really what he's hoping for. Instead, if the turn is a 5 or 7, the OP has at least three (maybe four) outs to the nuts on both sides.

[/ QUOTE ]Hi Nick - You're right. A seven on the turn is a much better card for Hero than a trey.

[ QUOTE ]
Is this enough to make the expectation positive?

[/ QUOTE ]Probably is.

Sorry I missed that the first time through. Thanks for the correction.

Buzz

[Buzz]

To do the thing up right, we'd have to figure (or simulate) separately for each different low rank, also taking flushes into consideration, and then tabulate.

It would take an extensive tabulation. I actually have done that, and as recently as this month for an article I'm currently working on. But it's very tedious work and there is other stuff I should be attending to instead at this time.

I tried to short-cut that process by just choosing a trey as the turn card. I vaguely (without thinking a lot about my choice) purposely avoided a card for Hero that would also give him high, because I just wanted to look at the low.

But Nick is right.

Buzz