14 replies to this topic

[ziggy587]

In tournament NL hold em. How many outs should I have to call an all-in on the flop? Just generally speaking of course as I realize it would depend on chip stacks, players, payout structure, etc. How many outs should I be looking for as a guideline?

[benhoug]

In tournament NL hold em. How many outs should I have to call an all-in on the flop?

Let the pot odds tell you... There is no such thing as a definitive answer to this question.

[The Bwaves]

In tournament NL hold em. How many outs should I have to call an all-in on the flop? Just generally speaking of course as I realize it would depend on chip stacks, players, payout structure, etc. How many outs should I be looking for as a guideline?

You want the pot odds to outweight your outs.

[ziggy587]

You want the pot odds to outweight your outs.

So is it outs divided by 45 cards? or 47 ?

9 outs = 20% so I have to be getting 5 to 1 ?
14 outs = 31% so I have to be getting 3 to 1 ?

Is this right?

[CobaltBlue]

So is it outs divided by 45 cards? or 47 ?

9 outs = 20% so I have to be getting 5 to 1 ?
14 outs = 31% so I have to be getting 3 to 1 ?

Is this right?

You know your cards and the 3 on the flop. So it's 47 for the turn and 46 for the river.

Assuming that you're calling an all-in...in order to be a favorite with your outs (with two cards to come), you need 14 outs. You'll be 51% at that point. So technically, you can call regardless of pot odds if you have that many.

The easiest way to figure out rough percentages (with two cards to come) is to multiply your outs by 4. If there's only one card left, multiply by 2.

As for the odds, you're translating them wrong. 20% is 4 to 1. 33% is 2 to 1. Think of the percentages as fractions...80% versus 20% -> 80/20 -> 4/1

[MasterLJ]

So is it outs divided by 45 cards? or 47 ?

9 outs = 20% so I have to be getting 5 to 1 ?
14 outs = 31% so I have to be getting 3 to 1 ?

Is this right?

No, that's not right.

The rule of thumb is take the number of outs and multiply it by 4 to get an *APPROXIMATE* percentage of winning by the river.

Flush draw = 9 outs, 9 x 4 = 36%
OESD = 8 outs, 8 x 4 = 32%

The rule is fairly accurate until you get above 13 outs.

You need 13 clean outs to be ahead of someone with a draw. Dirty outs are those that improve your hand, but improve your opponent's better. And example... if your opponent as a straight flush draw, then there are cards that can give you your straight, but your opponents straight flush, etc etc etc

[The Bwaves]

So is it outs divided by 45 cards? or 47 ?

9 outs = 20% so I have to be getting 5 to 1 ?
14 outs = 31% so I have to be getting 3 to 1 ?

Is this right?

Very close, see LJ post he nailed it.

[mk]

The rule of four and the rule of two are nice for quick estimations, but it really isn't that hard to work out the actual numbers and commit them to memory.

To calculate the actual number, you simply take the ratio of the number of cards that don't help you over the number of cards that do help.

An example:

You hold the 9 :icon_suit_diamond: 7 :icon_suit_diamond: in the BB and the flop comes A :icon_suit_diamond: J :icon_suit_club: 2 :icon_suit_diamond: . So you've seen 5/52 cards. There are 9 diamonds left in the deck. So that's 9 out of 47 that help your hand. So you subtract the 9 from 47 to get 38 cards left in the deck that don't help your hand. You take 38/9 and get 4.22. That means it's 4.22:1 against you making a flush on the turn only. On the river, you've seen one more card, so it's just 37/9, or 4.11:1.

In order to determine the probability of making the flush with two cards to come and hypothetically assuming you were all in on the flop (guaranteeing that you get to see both cards), you multiply the probabilities against hitting on the turn and river and subtract this from 1 (100%).

The probability against making a flush on the turn is 80.85%
The probability against making a flush on the river is 80.44%

Multiplying them, you get 65.03%; 1 -0.6503 = .3497, or 34.97%

[Bizzle]

The rule of four and the rule of two are nice for quick estimations, but it really isn't that hard to work out the actual numbers and commit them to memory.

To calculate the actual number, you simply take the ratio of the number of cards that don't help you over the number of cards that do help.

An example:

You hold the 9 :icon_suit_diamond: 7 :icon_suit_diamond: in the BB and the flop comes A :icon_suit_diamond: J :icon_suit_club: 2 :icon_suit_diamond: . So you've seen 5/52 cards. There are 9 diamonds left in the deck. So that's 9 out of 47 that help your hand. So you subtract the 9 from 47 to get 38 cards left in the deck that don't help your hand. You take 38/9 and get 4.22. That means it's 4.22:1 against you making a flush on the turn only. On the river, you've seen one more card, so it's just 37/9, or 4.11:1.

In order to determine the probability of making the flush with two cards to come and hypothetically assuming you were all in on the flop (guaranteeing that you get to see both cards), you sum the probabilities of each event (the turn and the river).

The probability of making a flush on the turn is 9/47, or 19.149%
The probability of making a flush on the river is 9/46, or 19.565%

Adding them, you get 38.714%, or about 1.6:1.

Subtract the ~3% you make a flush on the turn and another diamond hits the river, and this looks pretty good.

Quick edit: Make it 3.330% and that looks good

[mk]

Subtract the ~3% you make a flush on the turn and another diamond hits the river, and this looks pretty good.

Quick edit: Make it 3.330% and that looks good

Right, I was obviously discounting bd straights as well.

[zipper]

Subtract the ~3% you make a flush on the turn and another diamond hits the river, and this looks pretty good.

Quick edit: Make it 3.330% and that looks good

Subtract the ~3% you make a flush on the turn and another diamond hits the river, and this looks pretty good.

Quick edit: Make it 3.330% and that looks good

When figuring the odds with two cards to come, you do not add the odds of hitting the turn card to the odds of hitting the river card. That overstates the odds of hitting on one or the other. The proper math is to take the odds of not hitting the turn card, 80.85% and multiply it by the odds of not hit the river card, 80.43%. This results in 65.03% which is the odds that you will not hit by the river. Then subtract this number from 1 to get 34.97% and that is the odds that you will hit the flush on either the turn or the river. So it is better than 2:1 to hit a flush when you are all in after the flop.

[mk]

When figuring the odds with two cards to come, you do not add the odds of hitting the turn card to the odds of hitting the river card. That overstates the odds of hitting on one or the other. The proper math is to take the odds of not hitting the turn card, 80.85% and multiply it by the odds of not hit the river card, 80.43%. This results in 65.03% which is the odds that you will not hit by the river. Then subtract this number from 1 to get 34.97% and that is the odds that you will hit the flush on either the turn or the river. So it is better than 2:1 to hit a flush when you are all in after the flop.

You're right, my mistake. I will edit my o.p.

[copernicus]

The rule of four and the rule of two are nice for quick estimations, but it really isn't that hard to work out the actual numbers and commit them to memory.

To calculate the actual number, you simply take the ratio of the number of cards that don't help you over the number of cards that do help.

An example:

You hold the 9 :icon_suit_diamond: 7 :icon_suit_diamond: in the BB and the flop comes A :icon_suit_diamond: J :icon_suit_club: 2 :icon_suit_diamond: . So you've seen 5/52 cards. There are 9 diamonds left in the deck. So that's 9 out of 47 that help your hand. So you subtract the 9 from 47 to get 38 cards left in the deck that don't help your hand. You take 38/9 and get 4.22. That means it's 4.22:1 against you making a flush on the turn only. On the river, you've seen one more card, so it's just 37/9, or 4.11:1.

In order to determine the probability of making the flush with two cards to come and hypothetically assuming you were all in on the flop (guaranteeing that you get to see both cards), you multiply the probabilities against hitting on the turn and river and subtract this from 1 (100%).

The probability against making a flush on the turn is 80.85%
The probability against making a flush on the river is 80.44%

Multiplying them, you get 65.03%; 1 -0.6503 = .3497, or 34.97%

There is absolutely no reason to go beyond the rules of 4 and 2 during live play.

The biggest misunderstanding about pot/implied odds is that they act as a "stop light". If you have pot odds (even by 1 TC) you go, and if youre short (even by 1 TC) you fold.

The errors in the rules of 4 and 2 (except at the extremes) are insignficant to your overall results, and other consderations are far more important than being even 5% off in your estimate of winning chances.

[mk]

I agree that you don't often need to go beyond the rules of 4 and 2 in NL tourneys, but in limit cash games, for instance, it's nice to know where you stand because you obviously get into a lot of situations where pot odds make or break your decision.

[copernicus]

I agree that you don't often need to go beyond the rules of 4 and 2 in NL tourneys, but in limit cash games, for instance, it's nice to know where you stand because you obviously get into a lot of situations where pot odds make or break your decision.

You arent likely to play enough limit hold em hands in your life where the EV difference between the rule of 4 and 2 and exact probabilities are significant enough to outweigh the variance in perfect calculations. The exception might be at 13 outs +, but in limit holdem in particular you are always playing those hands anyway.