Folding AA

MGrah

New to the forum folks so forgive me if a similar discussion has been had recently.

Just looking for a couple of opinions on a strange hand.....
I was playing in a 5 seat STT last night, when a very strange situation came up. 3/4 hands into the game one player decides to push all in, gets two callers before it comes back to me on the BB.
I have AA.
I know pot odds scream for a call, and that you will be ahead of any hand (assuming no one else is holding AA) but I wonder how many people here would fold for the guarantee of being in the money, considering the approx 1 in 3 chance of going out.

Cheers

Lenny_mc

tis a tough one, but i think i may have called. Turns out you did right folding, but something like 70%* of the time you will win. *(I think, and this would go down the more players you are playing against)

oooh - i hate tough decisions.

NickyOD

MGrah wrote:

New to the forum folks so forgive me if a similar discussion has been had recently.

Just looking for a couple of opinions on a strange hand.....
I was playing in a 5 seat STT last night, when a very strange situation came up. 3/4 hands into the game one player decides to push all in, gets two callers before it comes back to me on the BB.
I have AA.
I know pot odds scream for a call, and that you will be ahead of any hand (assuming no one else is holding AA) but I wonder how many people here would fold for the guarantee of being in the money, considering the approx 1 in 3 chance of going out.

Cheers

Fold if you would rather finish 3rd than 1st.

Iago

First player was probably on AKs or any PP from QQ-88 and didn't really want a caller

First caller probably holding KK-JJ, maybe TT, but probably not..

Second caller I would normally put on AA, but if you're holding AA...I still think they have AA-KK.

preflop, I'd call here and then kick myself when I got outdrawn!

Lenny_mc

i suppose tho that you run the risk of being outdrawn, compared to a much less risk of finishing itm. tough call.

biteme

how can you fold this hand. Take out 3 players and have most of the chips on the table going into heads up. This is the hand that will win the STT for you most times

NickyOD

The only time you can ever justify folding AA is close to the money in a satellite or a tourney with a flat payout structure. Otherwise you play for first and make this call without even thinking about it every time. If your aces get cracked, so be it. Shrug and move on to the next one.

Lenny_mc

but if you have three players all in ahead of you, your going to have to assume that they have pretty decent hands, including some aces, so chances are you arent going to hit an ace. sure you have top pair, but your looking for the board to pair to get two pair, (possibly giving someone trips) your going to have to catch 4 cards for a flush or 4 for a straight. So, unless the board gives trips for a house (your still running the risk that someone has the 4th card especially if its a high card) you only have highest pair.

that would be my view of it anyway....

heads up or against two people in the long run you will win calling with aces, but against more than that and with three all in raises ahead of you our odds are going down. As i said, Its a tough call.

Turlock

So how did the hand actually work out?

henbane

This has been done to [post=2208136]death[/post] before. HJ's response is as correct as it gets. Especially given a normal payout structure for a 5 person STT. Do not fold this unless you can come up with a proper mathematical reason why based on the circumstances.

NickyOD

Lenny_mc wrote:

but if you have three players all in ahead of you, your going to have to assume that they have pretty decent hands, including some aces, so chances are you arent going to hit an ace. sure you have top pair, but your looking for the board to pair to get two pair, (possibly giving someone trips) your going to have to catch 4 cards for a flush or 4 for a straight.

That's not true. If the other two aces are not live then your chances of winning the hand are actually better because they have less outs for a straight.

AdAs: 60.03%
QQ: 16.86%
AhKh: 11.22%
AcJc: 11.87%

or....

AsAd: 59.39%
KsKd: 18.2%
ThTc: 17.15%
AhQs 5.25%

Try multiple combinations of 4 way all in hands and your aces will still be almost always be 55-61% favourite.

Calling here when you have the chance to bust 3 players is an easier decision than calling just one players all in with QQ.

MGrah

Turlock wrote:

So how did the hand actually work out?

Well - I was thinking something along the lines of Iago's post.
First player on ATs or better, second with big pocket pair KK - JJ, third on same with possible AA. Either way, I figured chances are there is at most one A in the deck, possibly not even that. So I folded - gutless I know, but with a full starting stack (short one blind) I would still have a chance.

Board came T68xx first raiser had unbelievably T8os! didn't get to see the other hands but I would expect they were slightly better.

Needless to say his chip lead didn't last, got to Heads up with the other player and lost with 9s over 8s against 9s over As.

henbane

MGrah wrote:

Well - I was thinking something along the lines of Iago's post.
First player on ATs or better, second with big pocket pair KK - JJ, third on same with possible AA. Either way, I figured chances are there is at most one A in the deck, possibly not even that.

All of which made you a massive favourite for the hand.

Board came T68xx first raiser had unbelievably T8os! didn't get to see the other hands but I would expect they were slightly better.

The result is totally irrelevant when it comes to making the decision preflop.

MGrah

henbane wrote:

All of which made you a massive favourite for the hand.
The result is totally irrelevant when it comes to making the decision preflop.

All of which puts me at about 55% to win, which against each hand individually is fine, but I'm playing against all three - not so massive of a favourite, when you'll lost almost half the time.
Obviously the consensus seems to be call, which answers my question. I think I'd still fold. My reasons are as follows:
let's say the payout is $100, $150, $250
you can't measure pot odds based on your stack here, regardless of what any book tells you because it's not real money it's chips you may end up with 4 times as many chips as the other guy heads up, but you can only win 1.66 times his winnings and 2.5 times the 3rd place.
Let's say you make this play 100 times and you're of a level standard to the already folded player:
55 times you win the pot, let's say you win out 50 times and finish 2nd 5 times that's $13,250
If you fold 100 times you win at least $10,000
that leaves you needing to come second 65 times to catch up, or win only 22 times.

Depends on how you do that math really.

I realise the result is completely irrelevant to the decision, just answering how it turned out.

Anyway, as I said apologies to those who have been through this a million times already.

NickyOD

MGrah wrote:

Depends on how you do that math really.

Download pokerstove, work out your fold equity and see if it changes your mind.

Rodge

Where do I find this fabulous 5 handed STT where the top three get paid? That has to be +EV.

Fatboydim

IMHO I think you were right to fold - I'm not saying it's what I would have done / hey I'd have probably called with 78 - But I think there's one simple rule in poker - if you're gut says "I'm not happy about this call" walk away from the hand and wait for a better opportunity. The purists will say that there is no better opportunity than getting dealt AA in the pocket. I asked Donnacha O'Dea the question is there ever a right time to fold aces. And his reply was the standard one that you read here... The only correct time would be if you have a low positive EV to win the tournament and you move up a few places in the finishing order thereby gaining more money - It's in Slansky's book on Tournament poker too. - Undoubtedly the correct play is to call.- But the little man says no - so fold and don't feel bad about it.

ionapaul

The little man says a lot of things in my ear while playing...'you haven't seen a flop in aaaaages, come on, live a little and limp in with your lovely 9 5 suited from early position!'...suffice to say I often ignore the little man in my head and think my play benefits from this

henbane

I really should be studying.

Assuming a set up similar to below...

pokerstove

equity (%)      win (%) / tie (%) 

Hand  1:	  15.3549 %   [  00.15   00.00  ]    { random }
Hand  2:	  13.3336 %   [  00.13   00.00  ]    { AA-22, AKs-ATs, KQs-KTs, QJs-QTs, JTs, AKo-ATo, KQo-KTo, QJo-QTo, JTo }
Hand  3:	  13.9227 %   [  00.13   00.01  ]    { AA-JJ, AKs, AKo }
Hand  4:	  57.3888 %   [  00.56   00.01  ]    { AA }

        	  equity (%)      win (%) / tie (%) 

Hand  1:	  11.3282 %   [  00.10   00.01  ]    { AA-TT, AKs-ATs, AKo-AJo }
Hand  2:	  12.8650 %   [  00.12   00.01  ]    { AA-JJ, AKs, AKo }
Hand  3:	  12.8650 %   [  00.12   00.01  ]    { AA-JJ, AKs, AKo }
Hand  4:	  62.9419 %   [  00.61   00.02  ]    { AA }

5 person STT with 3 places paid $250, $150, $100(bit weird but that's what you've posted). Within 3-4 hands of the first level so I'll assume all have starting chips.

Assuming no ties on the big pot, you're guaranteed $100 for folding but you and another player will be on 2000 starting chips and the other player will be on 6000.

Player A 6000
Player B 2000
You 2000

You will win the tournament one fifth of the time.

You will finish 2nd in two cases:

(a) Player A wins 3 fifths of the time and either you or player B will finish 2nd in half of those instances - (3/5)(1/2) = 3/10

(b) Player B wins and you finish 2nd. Player B will win one fifth of the time and you will finish 2nd two eigths of those times - (1/5)(2/8) = 2/40 = 1/20

You will finish 2nd (1/20) + (3/10) which is 7/20

You will finish last 1 - (7/20 + 1/5) = 9/20

EV = 1/5(150) + 7/20(50) + 9/20(0) = $47.50 [assuming buyin is $100 which makes sense from the given payout structure on a 5 person MTT]

If you make the call, you will win ~60% of the time. In that case

Player A 2000
You 8000

There are two outcomes here:

(a) You win 4/5ths of the time.
(b) You come 2nd 1/5th of the time.

EV = 4/5(150) + 1/5(50) = $130

Your $EV on the call is 60% of $130 which is $78.

Make the call even in an STT with a weird payout structure like this one. Even at 50% of $130 it's better to make the call.

I really should be studying.

Feel free to point out the numerous mathematical errors in this post

NickyOD

EDIT. Damn henbane why'd you have to do that same time I posted this. I don't have pokerstove at work. This is what I came up with in my head.

Ok. so lets say you are a 55% favourite to win and get 4-1 chiplead HU. Assuming you and your HU oponent are of equal abilities you are a 3-1 favourite to beat him at that point and win $250. .55*.75=41.25%. There is a 13.75% (.55*.25) chance you will win with your aces but still finish second for $150

You have a 45% chance of calling and winning nothing.

I don't know the stacks but assuming you fold and the pot is not split you have 100% chance of winning $100. At a guess 1 player will have 60% of the chips and you and the other shorty with each have 20%. Again assuming you are all of equal ability you now have a 25% chance of winning $250 (I think) and a 50% chance of winning $150.

Looks like folding was indeed the right move. Sorry I doubted you.

Fatboydim

I'll tell you something else about the little Man Paul - When he says "fold" he's always right - when he says "All in" he's always wrong. I'm thinking of giving him the sack.

Hitman Actual

Henbanes maths is pretty spot on, as far as I can see, so it's an obvious call. I worked it out a lot more generally as follows:

If you call, 60% of the time you will be heads-up with 8K against 2K, so you will win $250 four times out of five, and $150 the one other time. 40% of the time you lose and get nothing. So in 100 hands, that totals [(250x4) + (150X1)]x60 - (40 x 0) = $69000

If you fold, and its down to 3, then the chip ratios are 1:1:3, so you will win $250 one time in five. Here's where I generalise it to winning $150 two times in five for finishing second, and $100 two times in five for third. I know this isn't correct, but I can remember exactly how to do the 3-way maths!

So in a corresponding 100 hands where you fold, you win [(1x250)x20 + (2x150)x40 + (2x100)x40] = $25000.

I know my 3-way maths isn't strictly correct, but I don't think it's too far off (at least I hope not!). But anyway, this is always a call.

MGrah

Cheers Nicky - no worries. It's a marginal decision either way.

Henbane - don't think your maths work because you're not adding in the fact that you will lose $100 40% of the time. Do it again with just the payouts rather than the gain, and you'll get a different result. (If you just look at the gain - you have to add in the loss) On second thoughts don't - you should be studying

For those interested these are the payout structures on Pacific Poker for a 5 seat S&G. 1st 50% 2nd 30% 3rd 20%.
The game in question was a $10 buy in, $25, $15, $10.
Just made it $100's for easy math (and to make myself look big :cool: )

For big stake (IMO) players there's a $200 buy in game with the same pay structure ($500, $300, $200)

Culchie

1,2,3 - 50,30,20 % .... Wow !

... Then again 3rd = your money back as there's only 5 players (less reg fee) ... so still only 1st and second to make any money

MGrah

lenny_leonard wrote:

Henbanes maths is pretty spot on, as far as I can see, so it's an obvious call. I worked it out a lot more generally as follows:

If you call, 60% of the time you will be heads-up with 8K against 2K, so you will win $250 four times out of five, and $150 the one other time. 40% of the time you lose and get nothing. So in 100 hands, that totals [(250x4) + (150X1)]x60 - (40 x 0) = $69000

If you fold, and its down to 3, then the chip ratios are 1:1:3, so you will win $250 one time in five. Here's where I generalise it to winning $150 two times in five for finishing second, and $100 two times in five for third. I know this isn't correct, but I can remember exactly how to do the 3-way maths!

So in a corresponding 100 hands where you fold, you win [(1x250)x20 + (2x150)x40 + (2x100)x40] = $25000.

I know my 3-way maths isn't strictly correct, but I don't think it's too far off (at least I hope not!). But anyway, this is always a call.

Lenny - this doesn't work either, you're not applying the same calc to both situations try this....
1 [(250x4) + (150X1)]x60 - (40 x 0) = $69000
2 [(250x1) + (150x2) + (100x2)]x100 = $75000

but really it should be:
1 [(250x48) + (150x12) + (0x40)] = 13800
2 [(250x20) + (150x40) + (100x40)] = 15000

NickyOD

henbane wrote:

Player A 6000
Player B 2000
You 2000
You will win the tournament one fifth of the time.

Player A 2000
You 8000

There are two outcomes here:

(a) You win 4/5ths of the time.
(b) You come 2nd 1/5th of the time.

How did you work these out? I don't think having 1 5th of the chips means you will win 1 5th of the time. By the same logic if its 6 handed and you have 50% of the chips you would win 50% of the time. That can't be right.

For example in a HU where the stacks are 75% and 25%. Assuming equal ability. Player A has 66% chance of winning and Player be has 33%. I can't find the god damn equation for this but I read it on Andy Bloch's site.

Hitman Actual

MGrah wrote:

Lenny - this doesn't work either, you're not applying the same calc to both situations try this....
1 [(250x4) + (150X1)]x60 - (40 x 0) = $69000
2 [(250x1) + (150x2) + (100x2)]x100 = $75000

but really it should be:
1 [(250x48) + (150x12) + (0x40)] = 13800
2 [(250x20) + (150x40) + (100x40)] = 15000

Sh*t, you're right... I'm all over the place with that maths. It looks a lot more marginal than I first thought.

NickyOD wrote:

How did you work these out? I don't think having 1 5th of the chips means you will win 1 5th of the time. By the same logic if its 6 handed and you have 50% of the chips you would win 50% of the time. That can't be right.

AFAIR, it's generally accepted that your chance of winning a tournament is directly proportional to your chip stack in relation to the other stacks. Heads-up is the easiest example, so here it's 4/1. This was first introduced in TPFAP (at least I think so anyway ???). There were some good threads on 2+2 about this, but I dont have them bookmarked.

NickyOD

lenny_leonard wrote:

Sh*t, you're right... I'm all over the place with that maths. It looks a lot more marginal than I first thought.
AFAIR, it's generally accepted that your chance of winning a tournament is directly proportional to your chip stack in relation to the other stacks. Heads-up is the easiest example, so here it's 4/1. This was first introduced in TPFAP (at least I think so anyway ???). There were some good threads on 2+2 about this, but I dont have them bookmarked.

You're right. The result of the poll I read were incorrect. Sklansky has a "proof" in the Tournament Poker book.

Anyway the official answer from a pro is definite FOLDING OF ACES PREFLOP here.

Here was his reply.

"Wow its a fold.

I'm making an assumption that you have 55% equity in the four way pot with AA, which is just an estimate.

To figure out the tEV of folding I used the formula from here :

http://www.livejournal.com/users/extempore/91711.html

When you fold you will have 2,000 and another player will have 2,000 and the winner of the all in will have 6,000.

He will win 60% of the time and each of the small stacks will win 20% of the time.

He will finish 2nd 30% of the time (.75*.2 + .75*.2), therefore we each finish 2nd 35% of the time.

He will finish 3rd therefore 10% of the time, so we each finish third 45% of the time.

So tEV folding =

.2*250 + .35*150 + .45*100 = $147.50.

tEV calling =

.55*(.8*250 + .2*150) = $126.50.

You'd need 64.1% equity in the four way pot to call, and I don't think you do have that (although pokerstove couldn't tell me because it was taking too long)."

henbane

MGrah wrote:

Henbane - don't think your maths work because you're not adding in the fact that you will lose $100 40% of the time. Do it again with just the payouts rather than the gain, and you'll get a different result. (If you just look at the gain - you have to add in the loss) On second thoughts don't - you should be studying

Damnitall, you've found one of the few situations where the fold of AA is correct under certain assumptions.

$EV of folding = 1/5(150) + 7/20(50) + 9/20(0) = $47.50
$EV of calling = 0.60[4/5(150) + 1/5(50)] + 0.40[-100] = $78-$40 = $38

Tis pretty marginal either way as once AA is 65% for the hand (this depends on your opponents hand range), the call is profitable. There's also the consideration that it's possible for a call from AA to get knocked out but to get the 3rd place money in the tournament which leads to another round of similar math for the 40% side of the equation which might make it profitable - I'm definitely not doing this.

Feel free to point out any more mathematical errors in this post

Nicky wrote:

How did you work these out? I don't think having 1 5th of the chips means you will win 1 5th of the time. By the same logic if its 6 handed and you have 50% of the chips you would win 50% of the time. That can't be right.

I'm just using one method of determining likelihood of winning. I think you will find sklansky/malmuth use this one. There are other methods. All are faulty as they don't take into account ability or the impact a well-played large stack can have on a situation and a number of other factors. As long as you're comparing like with like the math should hold up.

NickyOD

I think this deserves a........

"NLD!"

henbane

Who is this pro you mentioned? Someone you know or did you email paul phillips?

Jaden

I know that maths made sense, but folding AA pre-flop - Never. My brain would go "fold, fold, fold", but by the time it regained control of my hands, the chips would be in there.

AA stands for Automatic Action. I'm never playing a 5 STT again, just in case I might come across this situation. Folding AA pre-flop would just hurt too much. I did it with KK a month ago (turned out to be the correct choice), but I haven't played since. Left, a bad, bad taste.

Interesting topic though.

NickyOD

henbane wrote:

Who is this pro you mentioned? Someone you know or did you email paul phillips?

I don't want to disclose his name but he's a friend of mine who plays on the WPT. Not Paul Phillips.

RoundTower

NickyOD wrote:

For example in a HU where the stacks are 75% and 25%. Assuming equal ability. Player A has 66% chance of winning and Player be has 33%. I can't find the god damn equation for this but I read it on Andy Bloch's site.

Suppose both players are so equally skilful that they both stick it all in every hand. Then Player B has a 50% chance of surviving the first hand and doubling up, and a 50% chance of winning the tournament if he does, so he has a 25% chance of winning.

I'm not surprised you couldn't find an equation that made up some other numbers.

If you have x% of the chips, and all players are playing to win the tournament, and all players are of equal skill, and the blinds are inconsequential, you should have x% chance to win the tournament.

Of course those 3 conditions are unlikely to be met, but this is a good approximation. I agree exactly with Nicky's "pro" maths, and exactly with henbane's maths up until the point where you go crazy and forget to add in some numbers, so I got:

EV(calling) = P(230) where P is the chance that you win the pot if you call, and

EV(folding) = 147.50.

If P>64.1 and you are exactly as skilful as your opponents you should call. Therefore you should probably fold.

RoundTower

Actually there's one other possibility - that the third player to go all in is not a moron, or that he is a moron who is lucky enough to have been dealt AA.

Then you still have only (probably) about a 60% chance to win the hand, but you will only be winning half the pot. I make your EV here $125, as compared to $147.50 by folding. This brings it from a marginal fold to a clear fold.

NickyOD

There's one thing that was forgotten and that's that if you call and lose, assuming all stacks are equal its a 3 way split for 3rd of $33. If you have slightly more then your equity increases quite a bit which means the equity in calling and folding are actually very close indeed.

Hectorjelly

I have used Andy Wards EV calculater for the EV calculations. Its not perfect but its roughly the same as the CEV calculater used by 2+2. You can download it here http://www.pokersoft.co.uk/freeware.htm


I assumed that it was a $25 stt, with first prize of $75, and a second of $50. Third gets nothing which makes the calculations easy. If there is some other prize structure let me know.

Calling and winning leaves 2 players, you will have 8000 chips and 2nd place locked up. This gives you a EV of $70.
Calling and losing leaves you with an EV of 0 as there is no third prize.

Folding leaves you with 3 players, you have 2000 chips and an EV of $33.

Using poker stove and some hand range guesses, your equity is around 60%.

Text results appended to pokerstove.txt

5,521,264,672 games 19.954 secs 276,699,642 games/sec

Board:
Dead:

equity (%) win (%) / tie (%)

Hand 1: 14.1454 % [ 00.14 00.00 ] { AA-22, AKs-A9s, KQs-KTs, QJs, AKo-A9o, KQo-KTo, QJo }
Hand 2: 14.0763 % [ 00.14 00.00 ] { AA-22, AKs-ATs, KQs, AKo-ATo, KQo }
Hand 3: 12.0091 % [ 00.11 00.01 ] { AA-77, AKs-ATs, AKo-ATo }
Hand 4: 59.7692 % [ 00.59 00.01 ] { AA }



So that means that the true equity of calling is 60% of $70, and 40% of $0. This is $42 which is greater than the equity of folding, which is $33.

RoundTower

Yes, good point, in that case it entirely depends on whether you have slightly more than the other players who have gone all in (even an extra $5) or slightly less. Let's assume you have probability P of winning the pot, the pot will not be split, and they all have an equal chance of drawing out on you.

If you call and win, this is more or less irrelevant, although one player may have a tiny chance of getting back into the game from a miniscule stack. If you call and lose:

If you cover everyone then you are guaranteed third place, so you have an added EV of (1-P)100. About $40.

If you cover only two of them, however, for you to finish third the slightly bigger stack has to win the pot, so you have added third-place equity of (1-P)33. About $13, and probably not enough to change from a clear fold to a marginal call (especially when there's a good chance the third player has AA)

If you cover only one of them you cannot finish third.

It now looks to me like if you have the biggest stack, you should call, and if you have the smaller stack, you should fold.

RoundTower

That was in response to Nicky. Hector, the payout structure is completely different from the one you assumed.

Here are my complete calculations:

Fold:
stacks 6000, 2000, 2000.
EV = 250(4/20) + 150(7/20) + 100(9/20)
EV = $147.50

Call and win:
stacks 8000, 2000. Prizes $250 and $150.
EV= 250(4/5) + 150(1/5) = $230.

Call and lose:
EV = 0. (may not be correct depending on exact stack sizes)

Call and split: Prizes $250, $150, $100.
stacks 4000, 4000, 2000
EV = 250(8/20) + 150(7.5/20) + 100(4.5/20)
EV = $179.

Suppose you win the pot P% of the time and don't split it.
Call: EV = P($2.30). If P is greater than 64, you should call.

Suppose you split the pot S% of the time and never win it outright.
Call: EV = S($1.79). If S is greater than 82, you should call.

Using Hector's results from pokerstove, you should fold in this case. However it is important to know the exact stack sizes, because you may have some chance of calling, losing, and finishing third.

NickyOD

Hectorjelly wrote:

equity (%) win (%) / tie (%)

Hand 1: 14.1454 % [ 00.14 00.00 ] { AA-22, AKs-A9s, KQs-KTs, QJs, AKo-A9o, KQo-KTo, QJo }
Hand 2: 14.0763 % [ 00.14 00.00 ] { AA-22, AKs-ATs, KQs, AKo-ATo, KQo }
Hand 3: 12.0091 % [ 00.11 00.01 ] { AA-77, AKs-ATs, AKo-ATo }
Hand 4: 59.7692 % [ 00.59 00.01 ] { AA }

lol. We could be here all day.
Personally I think these ranges are too high.
Player 3 is only going to be calling there with AK/KK/AA/QQ possibly JJ/TT

Hectorjelly

NickyOD wrote:

lol. We could be here all day.
Personally I think these ranges are too high.
Player 3 is only going to be calling there with AK/KK/AA/QQ possibly JJ/TT

Player 3 should only call with AA or KK, but then in any game where somebody open pushes in level 1 and gets 2 callers I would suspect that they arent the tightest or best players around. Also tightening up player 3 hand range makes very little difference to AA's equity.