To D.Sklansky - 88

[curtains]

I'm starting a new thread about the 88 hand with 1k - 2k blinds. I believe the question was "What should the stack sizes be relative to the blinds before it's incorrect to move allin with 88 preflop"

[David Sklansky]

It depends on how bad the other guy plays. The better he plays the bigger the stack. A flat call could be allright to induce a bluff even when stacks are smallish. A moderate raise generally can't be right unless the stacks are big enough that you can happily get away from the hand if he pushes all in on a reraise.

[ZeeJustin]

This question is much more complicated than it seems. It is relatively easy to figure out how big the blinds must be before shoving 88 becomes -ev. (This can either be done assuming your hand is face up, or putting your opponent on a range of hands. Both are relatively easy)

However, as the question is posed, you would also need to know the EV of all plays other than shoving. This of course depends greatly on how your opponent plays, and many other factors.

[TStoneMBD]

im glad you asked this question curtains. at least somebody is asking the right things.

that is all.

[kmvenne]

The more interesting question about this hand is the value of the check raise, more specifically calling it, since the move itself was cleary correct. Knowing Sklansky was waiting until these blinds to make his move for the win, surly the check raise is a lot more powerful then the normal large amount of strenght it shows. Should Phil lay this down getting 5-1? He'd have a lot of play left if he did, and knowing the chips he loses are worth a lot more then face value at these blinds, and assuming he had some play if he folds, which it sounds like he does, was calling a mistake here? I almost lean yes, perhaps not despite game theory considerations that might have made his choice moot assuming flops that kept 88 ahead were already decided upon for check raising.

[curtains]

I didn't originally ask it. I just copied it from someone else, whom David Sklansky asked to create a new thread on the subject.

[citanul]

For at least a partial answer, I think you should look up the karlson sklansky hand rankings. They may be called anything from "karlson sklansky" to "sklansky karlson" to "chubukov sklansky" to "sklansky chubukov." I honestly don't remember which got settled on in the end.

If searching becomes difficult, look through poster karlson's posts.

citanul

[That guy]

Phils stack was $25k and there was $3k in the pot ($1k/$2k) -- assuming no antes. So his stack/pot ratio is 8.33x. 88 is ahead of a random hand 97%+ of the time.

A moderate raise is problematic because of the implied odds you are laying.

(note: I don't know what Davids stack size was so I am just discussing this in theory)

If Ivey bets 1/2 his stack ($12,500), Sklanksy will have to call $12,500 - 2,000 or $10,500 to try to win $14,500. He has the odds if the cards are face up but should muck if he puts phil on 2 over cards -- based on pot odds. Based on implied odds, he could still call.

To me though, the real problem is what does Phil do when the board comes with over cards and 1/2 your stack is already in the pot? Now you have to push the rest in and hope your opponent missed the flop? That sucks.

But the conflicting problem for Phil is how to vary your pre-flop bets based on your opponents range of hands and still disguise what you have??

Phil probably raised to $5,500 so as not to give off any clues as to what he had... Still, Phil misplayed every street on this hand.

[maryfield48]

[ QUOTE ]
Phils stack was $25k and there was $3k in the pot ($1k/$2k) -- assuming no antes. So his stack/pot ratio is 8.33x. 88 is ahead of a random hand 97%+ of the time.

A moderate raise is problematic because of the implied odds you are laying.

(note: I don't know what Davids stack size was so I am just discussing this in theory)

[/ QUOTE ]

DS' stack had to be 12k - they started with 20k each, didn't they?

[WarDekar]

I don't think that's right because DS said Phil bet out 6k on the flop and DS check-raised him all-in. If DS only had 12k at the start of the hand, he wouldn't have been able to do that. Either way, it doesn't make much sense for Phil to raise 1/5 of his stack. Really the only way I see Phil laying down this hand post-flop is if DS moves in on the flop, but having position DS is at a huge advantage if he actually hits anything, especially hitting a board like this.

David - I have to ask, what would you have done if the flop had come different? Say you flopped top pair? Mid pair? JT8?

[valenzuela]

Pushing with 88 (showing them) is +ev with 42x the blinds or less. Now if you dont show them it could go to 60x the blinds maybe.

[J_V]

[ QUOTE ]
Pushing with 88 (showing them) is +ev with 42x the blinds or less. Now if you dont show them it could go to 60x the blinds maybe.

[/ QUOTE ]

I got about double your answer. Here's the simple math I did. Tell me if you spot any errors.

We are talking about opening all in on the button and flipping your hand up.

Chances of being dealt a pocket pair better than 88 = approximately 1/34. Thus, 33/34 times you will win the blinds. Let's assume blinds are 1 and 2. So you win 99.

1/34 times you are called by a better hand. You are approximately a 4-1 dog here.

So we solve for the break even point:

0= (blinds won) + suckout equity w/ 88 - equity loss w/88

0= 99 + .2x - .8x

x=165 Big blind =2. Breakeven point is 84.5x BB.

Edit: This seems to have major ramifications. 22 would be +EV up to 42xBB. Also, if your opponent will only call w/
QQ KK AA, it becomes profitable up through 165x BB or so.

*profitable, means not necessarily best, as ZeeJustin pointed out other plays may be more +EV, and that is hard to figure out.

[valenzuela]

deuces ( showing them) is 21 x the big blind, Aks is 32x, KK is 250, QQ 126, I have only worked out pocket pairs and ace/king, I might have a marginal error however, the most practical aplication to this IMO is against better players.( yes against negreanu im pushing QQ with 60 x the bigblind)

[MonkeeMan]

[ QUOTE ]
Tell me if you spot any errors.


Chances of being dealt a pocket pair better than 88 = approximately 1/34. Thus, 33/34 times you will win the blinds. Let's assume blinds are 1 and 2. So you win 99.

[/ QUOTE ]

You're assuming that overcards to the 8's will always fold. That's incorrect when you have a small stack. (e.g. JTs is damn near even money vs. 88) Also, some just like to gamble.

[That guy]

yes against negreanu im pushing QQ with 60 x the bigblind

against Negreanu, don't look at your second card until after you move all-in.

[citanul]

as i mentioned earlier, this work has been done for every starting hand, assuming the guy gets to see your cards before making his decision.

Link to whole thread (http://archiveserver.twoplustwo.com/showflat.php?Cat=&Board=genpok&Number=382772&Forum =,,All_Forums,,&Words=&Searchpage=1&Limit=25&Main= 382772&Search=true&where=&Name=5&daterange=&newerv al=&newertype=&olderval=&oldertype=&bodyprev=#Post 382772). Karlson made a mistake I believe in his first response, but if you read on, he eventually gets a whole sheet together.

This is a good start for talking about such things, prior to considering that your opponent might fold a hand he shouldn't, given that he doesn't see your cards.

citanul

[valenzuela]

Even if he reads me for queens is still +ev for me

[J_V]

You're assuming that overcards to the 8's will always fold. That's incorrect when you have a small stack. (e.g. JTs is damn near even money vs. 88) Also, some just like to gamble.

This doesn't matter because as the big blinds increase, calling with overcards is +EV for the 88. And as my answer indicates I'm not assuming a small stack.

[ QUOTE ]
Also, some just like to gamble.

[/ QUOTE ]

Uh, why did I respond to this post.

[J_V]

Could you please show your work. Your 88 answer looks off to me. I never do this math, since I play limit, but I'm pretty sure about my answer.

[valenzuela]

you show 88, you get called only by 99+, the chance of a pocket pair high than eights is:4*6=24 24/50=0.48....3/49=0.062.....0.48*0.062=0,029.....1/0.029= 34.5....your oponent has a better hand than yours 1 every 34.5 times..if you add up some suckouts ...34.5/0.81= 42.5

If you want more details on my math Ill gladly explain it( I like numbers, I even won some math contests until algebra arrived and ruined my life)

[sammysusar]

phil was probably making similar raises with alot worse hands than 88 so i assume he wanted david to go all in with something like a weak ace or a weak king. i guess he was trying to end the match right there. i dont see why it is such a terrible play 88 is fairly powerful heads up.

[That guy]

it is bad because he is letting David make a hugely profitable call with a hand he would have folded for a bigger raise. implied odds...

[That guy]

Do I have this right?

You bet up to the amount listed in the Sklansky-Karlson NL Hand rankings because you are trying to win the amount of the blinds. It is the summation of 1) all the blinds you win plus 2) the times you suck-out as the underdog which will mathematically be +EV -- even if your opponent knows what you hold???

Second question, according to the numbers listed... K2-offsuit has a ranking of 20... but this is based on a $2 blind. So it is 20/2 = 10x,

If your stack is exactly 10x the size of the blinds, you can move all-in with K2o when in the SB and this will be a +EV play? Is that a mathematically correct interpretation of these numbers??

this is quite enlightening... better than sliced bread... I have been wondering about this stuff as I often sit bewildered at the end of Sit N Gos...

[J_V]

[ QUOTE ]
if you add up some suckouts ...34.5/0.81= 42.5

[/ QUOTE ]

I can't figure out what you are doing with this, but it doesn't tell the whole story. Are you assuming only one blind? What about all the blinds you pick up?

This answer is just not right.

[MonkeeMan]

[ QUOTE ]
Uh, why did I respond to this post.

[/ QUOTE ]

You responded in an attempt to discredit my post. You failed.

I can tell you why I responed to your original post. You invited replies if anyone spotted any errors in your example. Don't ask if you don't mean it.

[valenzuela]

If im in the Sb and you are on the bb, and the blinds are 50/100 and either of us have less than 4250 its +ev for me to show you the eights and go all-in.

[J_V]

The problem we are referring to is theoretical. The fact that "some people like to gamble" does not matter and shows that you do not understand the problem at all. I believe the word "theoretical" would be lost on you in general.

[J_V]

Stop saying the same freaking thing over and over. I get the problem and your statement is true. However, it is true with much larger stacks as well. Like on the order of double the stack sizes.

Let's try this again. This is the last time I'm doing this.

In you example: 4250 stack sizes. Blinds 50-100.

You win 150 33/34 times. and then 1/34 you see a flop. You have 20% equity in that pot.

33*150+.2(4250) -.8(4250)=

4950+850-3450=2350 - not zero like it should be.

Try these numbers w/ 84bb and you'll see that I'm right.

Done with this thread now.

[MonkeeMan]

[ QUOTE ]
The problem we are referring to is theoretical. The fact that "some people like to gamble" does not matter and shows that you do not understand the problem at all. I believe the word "theoretical" would be lost on you in general.

[/ QUOTE ]

Wrong forum.

Ooh, a $5 word. I believe it's found in the dictionary after "pompous". Let's see if I can use it in a sentence:
"Theoretically, a player would fold JTs against an exposed pair of 8's because J_V asserts they would only play higher pairs."