15/30 AK

[benlj21]

1. yes, i think we have value in a raise, especially with the Ad.
2. we want them to call (although i think we are folding to a 3-bet?)

[jason_t]

[ QUOTE ]
1. yes, i think we have value in a raise, especially with the Ad.
2. we want them to call (although i think we are folding to a 3-bet?)

[/ QUOTE ]

wtf

[busted]

[ QUOTE ]
[ QUOTE ]
If people don't fold, raising is cutting down our own odds.

[/ QUOTE ]
On the turn, if people don't fold, you are getting 3:1 on your raise. You have 9 outs to the nut flush. 3 additional outs to a straight, but let's discount those in half because of the made flush possibility--1.5 outs. And, 6 outs to an over pair, which once again let's discount in half--3 outs. So, that's a total of 13.5 outs, which is roughly 27% equity. So, a raise could be for value if 3 opponents call.

[/ QUOTE ]
OK folks, help me out as I go through the numbers -- math was never my strong suit. Feel free to correct my calculations!

In counting discounted outs, I start with the following:
1. flush draw - 9 outs
2. OESD - 8 outs
3. gutshot straight draw - 4 outs
4. overcard - 1.5 outs
5. backdoor flush draw - 1.5 outs
6. backdoor straight draw - 1.5 outs
7. pocket pair or split pair - 2 outs
8. set or trips - 1 out

After the turn card, Hero's hand has: (1) a nut flush draw, (2) a nut gutshot draw, and (3) 2 overcards (A/K).

The flush draw is worth 9 outs. The gutshot draw is worth 3 outs because the Td is counterfeited by the simultaneous flush draw. The 2 overcards are worth 3 outs combined (50% discounting of 6 outs).

Thus, by my count, hero's hand has a total of 15 outs. 7stud suggests above that the 3 outs for the gutshot should be further discounted by 50% because of the made flush possibility -- fair enough. That results in a total of 13.5 discounted outs.

At the turn, there are 46 cards outstanding. Therefore, hero is an underdog by 32.5:13.5 = 2.41:1 (with pot equity of 1/3.41 = .293 = 29.3%).

When it's his turn to bet on the turn, hero has pots odds of 9 BB/2 BB = 4.5:1 if he raises, or 9:1 pot odds if he calls. However,the multiway pot with 4 players I think means that hero's outs should be further discounted. I don't know how to calculate this impact. But if he is better than a 4.5:1 underdog with the further discounting, then the turn raise is still justified.

I agree with 7stud that hero has plenty of pot odds to raise on the turn. But the key issue is whether each of his opponents have pot odds to call the raise. LMP and SB would have to CC 2 BB, with pot odds of 11:2 = 5.5:1. BB would only have to call a single bet, with pot odds of 15:1.

BB would call for sure (as he did). Whether LMP and SB have hands with odds that justify calling 2 bets is up for grabs (although my gut tells me there's a good chance that 1 of them would fold to a raise). If so, driving out the 3rd opponent significantly increases hero's chances of winning this big pot.

P.S. I'd definitely raise on the river, too. Basically for the same logic - the remote but non-negligible possibility that it may make one of your opponents fold. To quote SSHE again: "Even if your raise is not for value, in a large pot improving your winning chances by just a few percent can make your raise profitable. ... When you have a draw to a big hand (e.g., the nut straight or flush) the pot size determines how you should play it. In a large pot, improving your chance to win (with a raise) is more valuable (than just calling)." (pp. 159-160, Note - I added info. in parens)

[Hobbs.]

Below is a little EV calc comparing raising versus calling with a plot showing how our EV changes with the frequency that SB and LMP are willing to overcall.

EVr = b1 * (1bb) +b2 * (2bb) +b3 * (f1*(3bb) - (1 - f1)*(3bb))
+(a1 + b1)*(2bb)

EVc = f2 * (1bb) - (1 - f2)*(1bb) + (a2 + c2)*(1bb)

variables:
a1 = fraction of sb overcalling two cold
a2 = fraction of sb overcalling one bet
b1 = fraction of bb folding to river raise
b2 = fraction of bb calling our river raise
b3 = fraction of bb 3-betting
c1 = fraction of LMP overcalling two cold
c2 = fraction of LMP overcalling one bet
f1 = fraction of the time we are winning when bb 3-bets
f2 = fraction of the time we are beating bb when overcalling

Assumptions:
-we are winning 100% of the time against sb and LMP
-we are winning 100% of the time BB just calls our river raise

For a initial interation I made some assumption as to what some of these parameters should be:
b1 = 10%
b2 = 75%
b3 = 15%
f1 = 10%
f2 = 90%

(will make the independent variable the frequency at which LMP coldcalls)
a1 = 0.25 * c1
a2 = 0.5 * c2



For the graph I assumed that LMP or sb are fairly unlikely to overcall two cold and thus only extend that range from 0 to 0.2. OTOH, they are much more likely to overcall for one bet and thus we have a wider range on the x-axis. Since it is going to be less likely that LMP/SB overcall two cold lets set an semi arbitrary EV for the raising case at 1.4 (seen on the graph by the solid line) to get a feeling for the minimum frequency we need to pick up overcalls to make calling more correct in this situation.

Given all of these assumptions if get and overcall from LMP 40% of the time AND sb 20% then we are essentially indifferent to raising or calling. I would guess that given both of LMP and sb cold called on the flop and didn't raise when the diamonds got there on the turn that they are going to be more likely to overcall this river because they are holding top pairish type hands. Thus I think going for overcalls is going to be more profitable here.

[DeathDonkey]

Hi busted,

I like the style in which you post - you are here to learn and you are explaining your reasons quite well. That said, you have a couple ideas that are wrong and hopefully I can point them out and say why they are wrong.

[ QUOTE ]
hero has pots odds of 9 BB/2 BB = 4.5:1 if he raises

[/ QUOTE ]

[ QUOTE ]
But if he is better than a 4.5:1 underdog with the further discounting, then the turn raise is still justified.

[/ QUOTE ]

This is not how pot odds work. Pot odds are there to answer the question "should I call this bet or should I fold?" Your counting of outs and subsequent math look dead on to me, and you have answered the question of "should we call on the turn?" getting 9:1 of course the answer is yes.

But now if you want to ask "should I raise?" you can throw pot odds out the window and instead focus only on pot equity. You did that calculation as well but did not apply it - you calculated us to have 29.3% pot equity on the turn - with 3 opponents you have made an argument for raising for value. The problem with this is that all of your opponents would need to call your raise, and you would need to be correct about the number of outs that you have for this raise to be correct. Both contribute to the raise not being as much for value as it might first appear in my opinion.

There is more I am sure, but I wanted to clarify this difference between pot odds and pot equity and when to apply each.

-DeathDonkey

[7stud]

[ QUOTE ]
When it's his turn to bet on the turn, hero has pots odds of 9 BB/2 BB = 4.5:1 if he raises, or 9:1 pot odds if he calls.

[/ QUOTE ]
As DeathDonky pointed, that's not the correct way to look at whether you should raise or not. Calling and raising are two different betting propositions. If on the turn it was 3:1 to make the winning hand, and the pot contained 4bb, and it was 1bb to you, then you should call. That is the first betting proposition. The prior bets have created a pot which is offering you odds to take a shot at winning the hand.

The second betting proposition is whether to enter into an additional side bet with your opponents as to whether you will win the hand or not. However, you shouldn't conclude that if you raise, the pot will only be offering you 4bb:2bb = 2:1, and therefore you should forgo this second betting proposition. Whether you should raise or not depends only on how many of your opponents are willing to enter into this side bet with you. If you expect two opponents to call a raise, then you will be getting 2:1 on your raise when it's 3:1 against you winning the hand. If you expect 6 opponents to call a raise, then you will be getting 6:1 on your side bet, and you should make it, i.e. raise.

[7stud]

[ QUOTE ]
So, answer these two questions:

1. What is our equity likely to be? Do we have value in a raise?

2. How do we improve our equity by raising? Do we want the opponents behind us to call or fold?

[/ QUOTE ]

2a. By convincing the following hands to fold?
AA, KK, QQ, JJ, 44, AQ, AJ, A4, KT, and small made flushes

2b. It depends? We would like to increase our chances of winning when we spike an A, K, or straight, but if we currently have enough pot equity and no one folds, then our raise is for value.

[busted]

Thanks for your responses -- really appreciate it. I am interested in learning and plugging leaks in my game and developing a better understanding of basic poker principles.

I should admit my limitations. I got started with HE a couple years ago and used Jones (WLLHE) and Sklansky’s first book on HE (the book was about 25 years old but had some good stuff). When SSHE came out, that became my bible and my game improved a lot as I tried to follow Miller’s basic approach.

I guess I’ve known for a long time that I don’t have a deep understanding of many critical poker concepts. However, SSHE gave me a reliable, consistent, easy-to-understand approach that works fairly well, for me at least.

Also, I should mention that I focus first on table selection. I play 5/10 or lower and 90% of my hours are on Party on weekend nights. My winrate seems to be highest for tables that contain 2-3 low-to-moderately aggressive loose players, and the remainder of the table filled with loose, passive calling machines. This is followed by tables with mostly LP players, followed by LAG tables. I don’t do too well with TP players, and I leave tables quickly if it looks like it’s pretty TAG.

I guess that shows my level of confidence about playing with really good players. (But it helps my winrate a lot.)

Coming back to the hand being discussed in this thread, my overall perspective is as follows. First of all, I would have done exactly as hero for PF and flop betting – basically betting and raising until I’m called. There were 16 SB in the pot at the end of flop betting – which is a little over my personal criteria of 15 SB for identifying a VERY LARGE pot in the making.

With 3 opponents betting and 16 SB already in the pot, my mouth starts to water, especially if I’ve got a good hand. And I consider AdKs followed by QdJd4s to be a very strong hand – you’ve got a gutshot nut straight, the 2 highest overcards, and a backdoor flush draw.

At that point, I’m definitely pot committed. I’m going to be betting and raising until the showdown. When the 7d shows up on the turn, I’m estatic – man, now I’ve got a nut flush draw, in addition to the AK overcards and the nut gutshot draw.

That’s why I originally questioned just calling on the turn. I’d have been throwing money in that pot as fast as my opponents would allow me – right up to showdown.

However, my problem is that I don’t know or understand the theoretical poker reasons why I should be doing this. All I know is that this play is EV+ in the SSHE games I play. Sometimes I bust and am left with a hand with AKQJ high. Sometimes I hit the flush. Sometimes I hit the straight. And sometimes I hit one of the overcards that gives me TPTK and wins a fair proportion of these pots.

I think my approach is consistent with Miller’s approach. And it works well in the games I play. However, as is obvious from my previous posts, I don’t really understand the theory behind it. Maybe it is not correct from a theoretical poker perspective. Maybe it just works in the selective games I play and against the players that are at most of my tables.

Help – all comments and thoughts are welcome. I really want to plug any leaks in my game and work towards improving so I can be competitive at higher levels, where the competition is much tougher. FWIW, I’m extremely confident that at least 1 and probably 2 of my opponents would have folded before showdown (if I kept betting like a maniac). But this hand was from 15/30 which is way above my usual level, and all 3 of these opponents may have stuck until SD. I can’t really say because I don’t play at that level.

P.S. My gut feeling is that the approach I outlined works at low stakes because it’s related to “protecting your strong drawing hand” and “trying to drive out opponents by betting and raising” that is discussed at length in SSHE. But I truly don’t understand the theory behind it, or if I’m incorrectly applying the basic principles.

[Elevens]

Another great thread for the digest...

[7stud]

[ QUOTE ]
The problem with this is that all of your opponents would need to call your raise, and you would need to be correct about the number of outs that you have for this raise to be correct.

[/ QUOTE ]
As you probably are aware, the idea is that when some opponents fold, even though it will reduce the value your raise is getting, your winning chances will go up enough to offset that. In dollar terms, when opponents fold, what you gain is your increased winning chances mulitplied by the pot size.

However, I don't have enough experience to know where that balances out in this situation. I suspect that W.Deranged believes that those that fold will not have any winning chances to give you, and those that call will have you soundly beaten, so a raise won't work for either reason: it won't improve your winning chances enough and it won't be for value.

Are you so soundly beaten that your raise won't be for value? If you just look at the nut flush outs, you have 9 outs if an opponent has a made flush, which is roughly 20% equity. Additionally, if someone else has trips, they could make a full house on the river. Someone with trips has 9 outs to a full house, so you have to make your flush and they have to miss their full house. In fact, the 4d will not be an out for you if it makes someone a full house. That means you might only have 8 outs, and you could have equity as low as 8/46 x 37/46 = .14 or 14% (37/46 is the probability someone with trips misses their 9 outs to a full house).