200-400 Hand.

[TheRedDragon]

The amount of "outs" we have here isn't really important. The question is whether the profitability of our turn improvements is sufficient to justify the call. How often we win the pot is information we need to get our answer, but there's more to it than that.

[d10]

RedDragon, I might be wrong, because I haven't calculated the value of anything in terms of "outs" in a long time, but I think most people who use an out as a unit of measure are considering all of the factors that give a hand value and adding or subtracting partial outs as necessary. So it's pretty much the same thing you're saying, just different words.

[stinkypete]

okay, here's a simplified analysis of the backdoor outs. too tired to do any more with it right now. maybe someone else can pick it up. hopefully it makes some sense. for now i'm not considering cases where more than 2 bets go in on the turn, or where we make a flush and lose, which is obviously wrong.


6/47 of the time we're going to turn a flush draw.
-> 9/46 on turn.

3/47 of the time we're going to turn an open-ended straight draw.
-> 8/46 on turn.

6/47 of the time we're going to turn a gutshot straight draw.
-> 4/46 on turn.

1/47 of the time we're going to turn an open-ended straight and flush draw.
-> 15/46 on turn.

2/47 of the time we're going to turn a gutshot straight and flush draw.
-> 12/46 on turn.


(47*46) = 2162

paying only 1 bet on turn, we can draw to:
6*9/2162 = 2.498% (flush)
3*8/2162 = 1.110% (oesd)
6*4/2162 = 1.110% (gutshot)
1*15/2162 = 0.694% (oesd + flush)
2*12/2162 = 1.110% (gutshot + flush)
TOTAL = 6.522%

payign 2 bets on turn, we can draw to:
6*9/2162 = 2.498% (flush)
3*8/2162 = 1.110% (oesd)
1*15/2162 = 0.694% (oesd + flush)
2*12/2162 = 1.110% (gutshot + flush)
TOTAL = 5.682%


how many bets are we paying on average?

check through: 10% * 0.5 bets
turn 1 bet: 60% * 0.5(29/47) + 1.5(18/47) = 60% * 0.883
turn 2 bet: 30% * 0.5(35/47) + 2.5(12/47) = 30% * 1.011

average: 0.05 + 0.5298 + 0.303 = 0.883 bets

WIN % = 70% * 6.522 + 30% * 5.682 = 4.5654 + 1.7046 = 6.27%

bets won:
check through: 10% * 8.5 = 0.85
turn 1 bet: 60% * 10.5 = 6.3
turn 2 bet: 30% * 13 = 3.9
total: 11.05 bets

6.27% * 11.05 = 0.693 bets

So we're paying 0.883 bets to win 0.693 bets on average.

[The Bryce]

What you're forgetting SP, is that bets only go in on the turn when we improve to a better draw. Once we improve by picking up a diamond, for example, we are no longer a 25:1 dog to make that hand.

[stinkypete]

[ QUOTE ]
What you're forgetting SP, is that bets only go in on the turn when we improve to a better draw. Once we improve by picking up a diamond, for example, we are no longer a 25:1 dog to make that hand.

[/ QUOTE ]

i considered that.

[ILOVEPOKER929]

[ QUOTE ]
i used to autocall here. my natural tendency is to play loose and aggressive so this call comes naturally for me. now i play significantly more thoughtfully and cautiously when up against preflop raisers and reraisers. live those raises just tend to be so much more meaningful. it's easy to overestimate implied odds and underestimate reverse implied odds in the heat of battle. in aggressive games against experienced or tough players it's better to think the other way around. people suggesting the button might frequently check behind on the turn here are being way too optimistic. 200-400 players have a tendency to bet when checked to.

[/ QUOTE ]

I see what youre saying mike, but I do think if David knows his opponents well, this will take away some of the reverse implied odds aspect of this hand. I still believe that, if David can hit a Jack or a King on the turn or river and play this hand nearly perfect becuz he knows his opponents well, the flop play can go from a close call/fold situation to an easy call. Lack of opponent knowledge certainly increases RIO, which is an important issue in this hand.

[mike l.]

"For all I know, improving my hand could easily improve an opponent's hand further."

some real deep thinking there. this site is great now.

[Turning Stone Pro]

[ QUOTE ]
Reads are always nice, but this one is all math.

[/ QUOTE ]

Les, I agree with this, and that's why I feel knowledge of the opener is critical.

Let's say the turn improves your draw, or even gives you top pair or a pair or jacks. Whether the opener is going to check-raise the turn or not is paramount, since this is the difference between seeing the critical river card for 1 BB, or as many as 4. Huge mathematical difference.

Remember, the key to this problem, IMO, is that we have to figure the + or - ev knowing we need to hit twice. What it costs to get to the river is the most important factor.

TSP

[Peter_rus]

[ QUOTE ]

6/47 of the time we're going to turn a flush draw.
-> 9/46 on turn.


[/ QUOTE ]

There are ten outs on a flop to flush draw 7 to simple flushdraw, 1 to flush+openended, and 2 to flush+gutshot. You counted only 9. Is this because you discounted flush outs that pair the board?

[ QUOTE ]

So we're paying 0.883 bets to win 0.693 bets on average.

[/ QUOTE ]

Seems you're missed implied odds on a river which are positive even if you're including rare cases when our flush/straight will be dead. If you do - there will be more accurate ev-estimation from only backdoor outs.

Anyway - good job.

[stinkypete]

[ QUOTE ]
[ QUOTE ]

6/47 of the time we're going to turn a flush draw.
-> 9/46 on turn.


[/ QUOTE ]

There are ten outs on a flop to flush draw 7 to simple flushdraw, 1 to flush+openended, and 2 to flush+gutshot. You counted only 9. Is this because you discounted flush outs that pair the board?

[/ QUOTE ]

i actually just took these from an earlier post. i guess i should have checked them.

discounting for flush outs that pair the board is something that needs to be done though, and i'm not sure that reducing it like this is enough.

[ QUOTE ]


[ QUOTE ]

So we're paying 0.883 bets to win 0.693 bets on average.

[/ QUOTE ]

Seems you're missed implied odds on a river which are positive even if you're including rare cases when our flush/straight will be dead. If you do - there will be more accurate ev-estimation from only backdoor outs.

Anyway - good job.

[/ QUOTE ]

yep, i was lazy and tired and decided not to include that since it really complicates things. i was somewhat conservative with the implied odds partly because of this, but i probably didn't discount enough. you can lose quite a bit when you make your flush and lose, even though that's going to be a rare case. i'm not really sure how to calculate the odds there, since estimating the frequency at which that happens is really difficult.