simple push / fold question

[45suited]

Note that both poll responses stipulate that I am stupid in one form or another... /images/graemlins/tongue.gif


Seat 9 is the button

Total number of players : 7
Seat 2: NatetheNuts ( $615 )
Seat 1: Ultima785 ( $1410 )
Seat 6: HERO ( $845 )
Seat 4: brandonave ( $620 )
Seat 9: seahawks4412 ( $1955 )
Seat 7: bienpotron ( $1180 )
Seat 8: ByoCraig ( $1375 )

Blinds(50/100)

** Dealing down cards **
Dealt to HERO [ Jc Ac ]
brandonave folds.
HERO is all-In [845]

[tigerite]

You can push this face-up for +$EV. Not a lot, but still.

[bennies]

there are 1276 starting hands in poka. You would hate to face 42 of them (AQ+, JJ+) You have 5 opponents left to act.

[tigerite]

Put it this way even if they call with every pair, AK, AQ, AJ and KQs it's still +EV.. and even more so if they call with KQo, KT and QTs.

[45suited]

[ QUOTE ]
there are 1276 starting hands in poka. You would hate to face 42 of them (AQ+, JJ+) You have 5 opponents left to act.

[/ QUOTE ]

I posted this to see what the reasoning would be by those who don't like the push here.

By your above comment, are you advocating pushing or folding?

[AlcateL]

pooooooooooooooosh

[bennies]

I voted fold but I forgot you get 150 chips, not 100. This swings it, I'm sure it's a push, as tigerite said, even with your cards open...

[Ixnert]

I voted fold because I'm a douche and get nervous about pushing into that many people (and I hate AJ), thinking it was close, then ran the numbers, and it's not. Push.

Sorry I'm kinda a poker noob. Can someone explain how you guys are running these numbers? I understand the general idea, that even though hands like AK have him beat 70-30, the times that everyone folds increases the value of the push, but how exactly is this calculated? Thanks.

[Ixnert]

[ QUOTE ]
Sorry I'm kinda a poker noob. Can someone explain how you guys are running these numbers? I understand the general idea, that even though hands like AK have him beat 70-30, the times that everyone folds increases the value of the push, but how exactly is this calculated? Thanks.

[/ QUOTE ]

I did it kind of back-of-the-envelope, since I'm at work and don't have SNGPT at hand to do it automatically, but approximately:

Assume that only hands that beat you call you, but all of those do. (This is usually, though not always, the worst case scenario.)

There are 42 hands that beat you badly (AQ+, JJ+), and another 54 that are slightly ahead of you (22-TT). Just consider the 42, since the coinflips aren't going to change the numbers much. (They actually work slightly in your favor, since the blinds make that a +chipEV situation.)

42 hands out of 1276 total is about 3.3%. The odds that one of the four hands left to act is in this set is (1-0.033)^4 = about 12%.

So about 88% of the time, you get the blinds without a fight, +150 chips.

When you get called, you're, aggregate, about a 3:1 dog. So about 9% of the time (12.5% * 3/4) you get called and lose, and about 3% of the time you get called and win. -825 chips when you lose, at least +875 when you lose. More likely +975 or +925, but take the worst case scenario.

87.5% * +150 = +132
9% * -825 = -74
3% * +875 = +26

132 + -74 + 26 = +84 chips. This is your chip EV. In reality, it'll be slightly higher than this, because you might get called by some hands you beat (smaller aces, KQ), and if you get called by a smaller pair you still have positive equity because of the blinds even though you're a slight dog.

Your $EV is a bit harder to figure out, but usually when your chip EV is this positive, your $EV will be positive as well, at least away from the bubble. Look at the favorite links post (sort by most views, it'll be near the top) for links on ICM if you're interested in detail on this.

http://www02.clf.navy.mil/enterprise/Photos/page%203/push%20big.jpg

[stupidsucker]

What level?

I can see passing on this at the 10+1, but not if the blinds will rise before you are the BB.

This is a + EV push no doubt in my mind.

For the record... I push this, but I dont feel it is an autopush.

[stupidsucker]

your seats are listed out of whack... I think.

[45suited]

It was an 11. I'm trying to make plays like this basically auto-push or auto-fold, based strictly on the math.

I avoid tilt this way, since I can just push my chips in knowing that I did the "right" thing. What happens after that... meh.

[45suited]

[ QUOTE ]
your seats are listed out of whack... I think.

[/ QUOTE ]

I don't know why the HH lists them like that. Seat 9 was the button, 1 was the SB, and 2 was the BB. Sorry for the confusion.

[stupidsucker]

[ QUOTE ]
I can see passing on this at the 10+1, but not if the blinds will rise before you are the BB.

[/ QUOTE ]

I found myself to be 100% wrong here. This hand with a tighter calling range behind you makes your EV less then loose callers. This makes sense because your EV comes mainly from Ax callers behind. the 150 chips really isnt a big deal for $EV, but are gold for just adding to your stack to maintain it for future FE.

Thanks! That was a very enlightening and informative post, thanks for taking the time to help. Is the SNGPT you refer to the SNG power tools? Would you recommend buying that for a beginning SNG player?

[schwza]

hahahaha

[schwza]

[ QUOTE ]
Just consider the 42, since the coinflips aren't going to change the numbers much. (They actually work slightly in your favor, since the blinds make that a +chipEV situation.)


[/ QUOTE ]

this is a touch shady. you'd prefer that 33 folds rather than calls because even though your situation for 33 calling is +EV, it's more +EV to steal blinds. barely matters - i'm just nitpicking.

speaking of nitpicking, your result against the calling range will be less good than what you described. your analysis assumes that if QQ calls, there's no chance AA overcalls. i.e., your range should not be randomly distributed among the calling range. but this is tiny. nice analysis overall.

Yeah, I would definitely recommend it. When it's used well, it's immensely helpful and insightful. And even used badly it's still +EV.

Do searches on ICM to get an idea of what you're getting into. If you buy the program, my advice is to look at it more as a teaching tool, less as the Rosetta Stone. In other words, it's at its best when the user is thoughtfully experimenting with it instead of just mindlessly jamming in numbers and cards expecting an infallible answer to pop out like toast. Mmm, toast.

eastbay doesn't need me to shill for him, but I'm very happy I bought it.

[jdl22]

[ QUOTE ]
Sorry I'm kinda a poker noob. Can someone explain how you guys are running these numbers? I understand the general idea, that even though hands like AK have him beat 70-30, the times that everyone folds increases the value of the push, but how exactly is this calculated? Thanks.

[/ QUOTE ]

get this program (http://www.sitngo-analyzer.com/)

[Ixnert]

[ QUOTE ]

this is a touch shady. you'd prefer that 33 folds rather than calls because even though your situation for 33 calling is +EV, it's more +EV to steal blinds. barely matters - i'm just nitpicking.


[/ QUOTE ]

This is true; an artifact of the "back-of-the-envelope"-ness of it. Simplifying assumptions rule. /images/graemlins/grin.gif (At least when they don't overly much affect the result, as I'm convinced they don't here, though I worded my disclaimer to that effect badly in my original post.)

[ QUOTE ]

speaking of nitpicking, your result against the calling range will be less good than what you described. your analysis assumes that if QQ calls, there's no chance AA overcalls. i.e., your range should not be randomly distributed among the calling range. but this is tiny. nice analysis overall.

[/ QUOTE ]

I don't feel too bad about this assumption, because I'm pretty sure SNGPT does the same thing. (Assumes that you'll get at most one caller, I mean.)

[schwza]

[ QUOTE ]
I don't feel too bad about this assumption, because I'm pretty sure SNGPT does the same thing.

[/ QUOTE ]

that's pretty distressing to hear. the # of times you see aces will sometimes be way higher than the number implicitly assumed by sngpt if that's the case.

[Ixnert]

[ QUOTE ]
Thanks! That was a very enlightening and informative post, thanks for taking the time to help. Is the SNGPT you refer to the SNG power tools? Would you recommend buying that for a beginning SNG player?

[/ QUOTE ]

SNGPT = SNG Power Tools, yes.

Absolutely useful, if used thoughtfully. And definitely worth buying for anyone but the most casual of players; easily pays for itself. I'd even say it might be better to buy it sooner rather than later as a beginning player, to reduce the bad habits you might have to unlearn.

I still don't push as much as SNGPT tells me I should, but I think it has improved my instincts for such situations quite a bit over what they were.

[Ixnert]

[ QUOTE ]
[ QUOTE ]
I don't feel too bad about this assumption, because I'm pretty sure SNGPT does the same thing.

[/ QUOTE ]

that's pretty distressing to hear. the # of times you see aces will sometimes be way higher than the number implicitly assumed by sngpt if that's the case.

[/ QUOTE ]

It hardly seems worth worrying about in most cases. Taking the original hand (AJs, 4 to act), if you assume that someone has a hand that we're worse than a coinflip against (AQ+, JJ+), the odds that someone ELSE has exactly AA are about a quarter of a percent. Given that we're talking about that on top of something that only happened one time in eight or so (to be pedantic, they're not independent events, but close enough given that we're talking in approximations), it'll only be a factor in really marginal cases.

It's much more significant with more hands yet to act, of course, so you'd want to be much more careful with, say, 7 to act, but then you're not going to be pushing much in that situation.

[curtains]

I would NEVER in my wildest dreams fold this. I don't think its at all close.