SSHE Book Club Discussion - Part Three: Postflop Concepts (1st Half)

[johnnybeef]

TT,

great post explaining the discounted outs section. it is my belief that the section on hidden outs is somewhat of an fyi section (i.e. heres what a hidden out is yada yada). in the text, mr. miller describes how hidden outs can be calculated into a weighted average giving you a palpable way to account for hidden outs. in my opinion this is impractical, as calculating a weighted average is a somewhat lengthy calculation (especially when done mentally). i think the thing you need to take away from this section is that every once in a while there is a possibility that your opponent may be counterfeited. because of this it is sometimes correct to call when your odds to improve are slightly less than those the pot is offering you. (in fact i just went back and reread the section and ed states on pg 110 something very similar to what i wrote above.)

good luck everyone,
johnny

[darvon]

OK. I understand that I should read TOP. It's on order. I have read WLLH and ITH and sections of others.

And I appreciate that people are trying to explain some nuances of Counting Outs.

But I would like to hear some opinions on my basic question. Let me restate it and focus it on Counting Outs.

Is SSH attempting to teach me Counting Outs or is it assuming I know how to Count Outs and is only discussing nuances of it?

[mistrpug]

[ QUOTE ]
Is SSH attempting to teach me Counting Outs or is it assuming I know how to Count Outs and is only discussing nuances of it?

[/ QUOTE ]

Well what is there to know other than the "nuances of it"? I mean, the basic way to count out is "Nine cards make my flush. I have 9 outs." What else is there to know?

Ed teaches you to think more along the lines of: "It's not a very high flush, so maybe I'll discount my 9 flush outs to 8, and I have an overcard that might be good, I'll count that as another 1.5. Add my backdoor straight(1.5) and I have about 11 outs..."

[BradleyT]

[ QUOTE ]

Ed teaches you to think more along the lines of: "It's not a very high flush, so maybe I'll discount my 9 flush outs to 8, and I have an overcard that might be good, I'll count that as another 1.5. Add my backdoor straight(1.5) and I have about 11 outs..."

[/ QUOTE ]

That right there is the golden part of counting outs that most of us never practiced before.

[darvon]

What there is to know is a couple of things.

You use "Counting Outs" as a quick and easy method of estimating the probability of your hand (or an opponents) winning.

Basic question 1: When is Counting Outs going to be close enough to be useful and when is it not? What method do I use when it is not?

Basic question 2: People use the phrase "Counting Outs is the way of estimating the probability of improving your hand". While that is exactly accurate, it is not useful. When you figure to call/fold with pot odds or pot equity et al you are comparing the expected values of the pot (or betting cycle) to the amount of the call or bet, thus you need the probability of winning, not improving. So we make the simplifying assumption that for cases not covered in Q1, improving=winning. Is this correct?

Basic Question 3: When you flop no pair, is Counting Outs still germaine?

Basic Question 4: When you have Pocket Pairs is Counting Outs still germaine?

Those are the types of questions that where I am still unsure of the answer.

[AKQJ10]

[ QUOTE ]
Basic question 1: When is Counting Outs going to be close enough to be useful and when is it not? What method do I use when it is not?

[/ QUOTE ]

I don't think it's ever just plain useless, but the cases you mention below are cases where it's probably more trouble than it's worth. If you flop a strong hand then you're more interested in whether to built a pot or protect your hand, and how to protect your hand, than you are in counting the number of cards that will retain your advantage.

[ QUOTE ]
Basic question 2: People use the phrase "Counting Outs is the way of estimating the probability of improving your hand". While that is exactly accurate, it is not useful. When you figure to call/fold with pot odds or pot equity et al you are comparing the expected values of the pot (or betting cycle) to the amount of the call or bet, thus you need the probability of winning, not improving. So we make the simplifying assumption that for cases not covered in Q1, improving=winning. Is this correct?

[/ QUOTE ]

Nononononono.... a thousand times no. This may be where some of the confusion is arising. In every other poker book I've read about figuring pot odds, this simplifying assumption is made, or the whole topic is glossed over. But emphatically the point that the SSH authors are making is that improving your hand does NOT equate to winning the pot, and that you're far more concerned with the latter. That's the whole point of the discussion above about discounting outs -- those cards that improve your hand, but only stand a 50% chance of winning you the pot, should be counted as only half as valuable as those cards that improve you to the unbeatable nuts.

[ QUOTE ]

Basic Question 3: When you flop no pair, is Counting Outs still germaine?

[/ QUOTE ]

Absolutely! A two-overcard hand has six outs to top pair, but those 6 need to be discounted drastically because top pair is such a precarious hand. Of course how much you discount them depends on several factors -- the level of coordination of the flop, the betting, your knowledge of your opponents, etc. We can debate just how much to discount certain hands.

[ QUOTE ]

Basic Question 4: When you have Pocket Pairs is Counting Outs still germaine?


[/ QUOTE ]

Possibly, but in the case of small pocket pairs that don't flop a set, you have two outs to improve, even before discounting for the possibility that a set may not win. Unless you can expect the pot to pay you off at 22 to 1, it's not even worth doing the math.

In the case of large pocket pairs, you're estimating your probability of winning, which involves pot odds but not really out-counting. If overcards to your pair flop, your probability of winning may be small. However, suppose you hold JJ, the flop is Q-high, and an opponent bets into you. If you estimate that there's a 25% chance your opponent would bet with less than top pair, and the pot's offering you good enough odds you might want to stay in. That said, the pot would have to offer you MUCH better than 3 to 1 because you can anticipate having to call double-sized turn and river bets. If you figure you'll have to call a total of at least 5 more small bets (one small + two big), you want the pot to contain more than 10 small bets in the example I gave (because your opponent will put in 5 more as well, giving you 15 to 5 pot odds by the river).

Anyway, i invite others to critique that example but my point is it's not usually a matter of outs because the 1/23 probability that the next card will make your set is so small.

One case with a pocket pair where outs do matter is when you flop a set but think a straight or flush is out against you. In that case you have seven outs to a full house or quads on the turn and ten on the river. If your set is low you might want to discount those somewhat for the possibility of an opponents' higher full house. If you have the top set then you need only to consider the possibility of quads or a straight flush, which of course are both pretty remote.

Hope this helps. I'm certainly not an authority on Ed's book, but i've read and studied it, and i think i have a pretty good idea about these concepts. Of course criticism of my understanding is most welcome.

[MEbenhoe]

This hand is a hand I played at 4/8 at Canterbury a couple months ago. I feel its a good hand for demonstration purposes.

Hero is on the button with A [img]/images/graemlins/diamond.gif[/img] 5 [img]/images/graemlins/diamond.gif[/img] in a 9 handed game. UTG folds, UTG+1 folds, MP1 calls, MP2 folds, MP3 calls, CO calls, Hero calls, SB completes, BB checks.

6 players see a flop of J [img]/images/graemlins/club.gif[/img] 8 [img]/images/graemlins/diamond.gif[/img] 5 [img]/images/graemlins/heart.gif[/img].

SB checks, BB checks, MP1 bets, MP3 and CO call.

Hero has how many outs?

Hero should... fold, call, or raise?

Part 2 to follow as soon as a consensus answer is reached on this one.

[BAS]

I'd count 1.5 outs for the runner runner nut flush.
The 2 5's should be good for 2 more.
Someone could have AJ or A8, so I would discount the 3 potential A's to 1.5 outs.

I'd count 5 outs, and I'd raise.
Maybe you knock out the blinds with something like A8, and potentially you get a free card on the turn.

-Brent

[darvon]

First of all, thank you for responding.

And let me try to focus my questions. Although I might want to discuss it, what I am looking for is a TEXT that will give me a methodology that will answer these questions. I raised it as I wondered if SSH was trying to be that text or not.

[ QUOTE ]
[ QUOTE ]
Basic question 1: When is Counting Outs going to be close enough to be useful and when is it not? What method do I use when it is not?

[/ QUOTE ]

I don't think it's ever just plain useless, but the cases you mention below are cases where it's probably more trouble than it's worth. If you flop a strong hand then you're more interested in whether to built a pot or protect your hand, and how to protect your hand, than you are in counting the number of cards that will retain your advantage.


[/ QUOTE ]

That sounds reasonable. I would like to see a text which discusses it in depth.


[ QUOTE ]


[ QUOTE ]
Basic question 2: People use the phrase "Counting Outs is the way of estimating the probability of improving your hand". While that is exactly accurate, it is not useful. When you figure to call/fold with pot odds or pot equity et al you are comparing the expected values of the pot (or betting cycle) to the amount of the call or bet, thus you need the probability of winning, not improving. So we make the simplifying assumption that for cases not covered in Q1, improving=winning. Is this correct?

[/ QUOTE ]

Nononononono.... a thousand times no. This may be where some of the confusion is arising. In every other poker book I've read about figuring pot odds, this simplifying assumption is made, or the whole topic is glossed over. But emphatically the point that the SSH authors are making is that improving your hand does NOT equate to winning the pot, and that you're far more concerned with the latter. That's the whole point of the discussion above about discounting outs -- those cards that improve your hand, but only stand a 50% chance of winning you the pot, should be counted as only half as valuable as those cards that improve you to the unbeatable nuts.


[/ QUOTE ]

Thanx again. I are correct but I have poorly stated my question. Assuming that an improved hand is a winning hand is the assumption of "Level 1" Counting Outs. "Level 2" and above use more sophisticated techniques like discounting to get a good estimate in situations where Level 1 breaks down. But the goal of Level 1 and Level 2 is to give me the probability of winning, so that I may multiply that probability by the size of the pot (or betting round) to allow me to compare it to the size of a bet to make my decision. The questions are when to shift from Level 1 to Level 2 to Level 3 etc.. and whento abandon Counting Outs altogether.

[ QUOTE ]

[ QUOTE ]

Basic Question 3: When you flop no pair, is Counting Outs still germaine?

[/ QUOTE ]

Absolutely! A two-overcard hand has six outs to top pair, but those 6 need to be discounted drastically because top pair is such a precarious hand. Of course how much you discount them depends on several factors -- the level of coordination of the flop, the betting, your knowledge of your opponents, etc. We can debate just how much to discount certain hands.


[/ QUOTE ]

That is what I understand from reading various texts, but what I haven't found is THAT discussion in a text. Where are they?

[ QUOTE ]

[ QUOTE ]

Basic Question 4: When you have Pocket Pairs is Counting Outs still germaine?


[/ QUOTE ]

Possibly, but in the case of small pocket pairs that don't flop a set, you have two outs to improve, even before discounting for the possibility that a set may not win. Unless you can expect the pot to pay you off at 22 to 1, it's not even worth doing the math.

In the case of large pocket pairs, you're estimating your probability of winning, which involves pot odds but not really out-counting. If overcards to your pair flop, your probability of winning may be small. However, suppose you hold JJ, the flop is Q-high, and an opponent bets into you. If you estimate that there's a 25% chance your opponent would bet with less than top pair, and the pot's offering you good enough odds you might want to stay in. That said, the pot would have to offer you MUCH better than 3 to 1 because you can anticipate having to call double-sized turn and river bets. If you figure you'll have to call a total of at least 5 more small bets (one small + two big), you want the pot to contain more than 10 small bets in the example I gave (because your opponent will put in 5 more as well, giving you 15 to 5 pot odds by the river).

Anyway, i invite others to critique that example but my point is it's not usually a matter of outs because the 1/23 probability that the next card will make your set is so small.

One case with a pocket pair where outs do matter is when you flop a set but think a straight or flush is out against you. In that case you have seven outs to a full house or quads on the turn and ten on the river. If your set is low you might want to discount those somewhat for the possibility of an opponents' higher full house. If you have the top set then you need only to consider the possibility of quads or a straight flush, which of course are both pretty remote.

Hope this helps. I'm certainly not an authority on Ed's book, but i've read and studied it, and i think i have a pretty good idea about these concepts. Of course criticism of my understanding is most welcome.

[/ QUOTE ]

I have no criticism of your understanding. It matches and exceeds my understanding. What I am having trouble finding is a text with a methodology on when to use/modify/abandon Counting Outs in enough depth to apply to all of the hands I see.

[darvon]

Thank you. That is an excellent demo.

You ask "How many outs do you count?" I will answer that but let me ask a different question. "How many outs does SSH count?"

SSH teaches hidden outs but the amount of possible hands of the raiser and 2 callers is daunting. Opponents could have gutshot, or open-ended or one pair with Ace kicker or one pair with non-Ace kicker or two pair or very remotely a set or a backdoor flush no pair
.

gutshot is too small, ignor it.
OESD is 1.5 outs you have no outs to beat it. It will occur 3* 4/47 * 4/46 or 1/47 of the time
pair with Ace kicker will occur...

STOP! My time just ran out at Party and it folded my hand. [img]/images/graemlins/frown.gif[/img]

RATS!

Well if I don't have the time to do all the combos for Hidden Outs, I find no method in SSH that will help here.

So let me go back and just discount some outs.

The two 5s give me trips which will probably hold. 2 outs.

3 Aces will give me two pair, if someone is holding an Ace and a board, then I won't be happy. I have 3 people who probably have a pair, I estimate that all 3 won't hold an A about 3/4 of the time. call it about 2 outs.

I have 4 outs. about 8.5% chance, 13 SB in the pot....call.