all-in pre-flop KK vs. QQ deal

[Jive Dadson]

Post deleted by Jive Dadson

[luckychewy]

ur not in the latb thread guy.

[Zag]

[ QUOTE ]
[ QUOTE ]

By the way, another way to reduce the variance, one that is always (nearly) fair, is to split the pot into two or three roughly equal sub-pots. (It doesn't matter if they are exactly equal.) Then deal out a different board for each sub-pot, awarding each one separately. This actually turns out to be a slight advantage for the hand that is ahead, unless you gather up the cards and reshuffle for each separate board you deal out. But the difference compared to true odds is slight.



[/ QUOTE ]

That's not so. Whether you shuffle the cards back in or not, the EV (expected value, or "true odds" as you call it) is unaffected. You just have to decide before hand which way you are going to do it. Not shuffling the cards back in reduces the expected variance more than shuffling back in does, so no one ever shuffles them back in.

Trust me. I am a master of mathematics. [img]/images/graemlins/grin.gif[/img]

[/ QUOTE ]
Well, I don't trust easily. In fact, I shouldn't, because you happen to be wrong. The difference is introduced because the hand that is behind might waste an out by winning one of the sub-pots "twice." Since he doesn't get paid double for winning twice, his overall EV drops very slightly.

Let's take an extreme case: Ks Qs vs Jh Jc on a board of Jc Js Ts. So the KQ has exactly 2 outs (As and 9s), and two cards to hit one of them.

With reshuffle (or only drawing once): EV of the quads is (43*42)/(45*44) times the size of the pot.

However, if you don't reshuffle, the quads still has exactly that chance to win the first sub-pot. HOWEVER, there is a tiny chance 2/(45*44) that BOTH the As and the 9s will fall on that first pot, leaving the quads absolutely guaranteed to win the second (and all subsequent) sub-pots. Obviously, if he sometimes has a slightly higher EV and never has a lower one, then his overall EV has gone up.

[Requin]

lol why would you bump it after this long, just to make an argument that's wrong.

[Zag]

Sorry, I've been away and was just catching up with old threads that I had been in. Please explain why my argument is wrong.

[tewall]

Another possibility (agreeing with your argument) is that exactly one of the As or 9s will come fall in the first sub-pot, which kills any future sub-pots. With reshuffling, the behind guy always has a chance of winning.

[good2cu]

[ QUOTE ]
It's easy enough to calculate. Just go to twodimes.net

http://twodimes.net/h/?z=171416

You had 82% pot equity, villain had 18%

Overall pot was $380. 18% of $380 is $68.40. You made a good deal.

By the way, another way to reduce the variance, one that is always (nearly) fair, is to split the pot into two or three roughly equal sub-pots. (It doesn't matter if they are exactly equal.) Then deal out a different board for each sub-pot, awarding each one separately. This actually turns out to be a slight advantage for the hand that is ahead, unless you gather up the cards and reshuffle for each separate board you deal out. But the difference compared to true odds is slight.



[/ QUOTE ]

Can someone please find out a defentive awnser to this question. Donbuttons/Apathy myself had an 2 hour debate about this and never firuged it out.

[Big_Jim]

Here is the simplest example possible:

Say you have set over setted someone on the turn and the pot has $44. So the underset is drawing to one out.

If you run it once:
He'll win 1/44 times, so his EV is $1.

If you run it twice and don't reshuffle the cards:
He'll get half the pot 1/22 times (he can never win the whole pot) so his EV is again $1.

If you run it twice and do reshuffle the cards:
He'll win the whole pot 1/44^2 = 1/1936
Get half 1/44 * 43/44 * 2 = 86/1936
And lose the whole pot 43/44^2 = 1849/1936

so his EV is:
44 * 1/1936 + 22 * 86/1936 = $1

Therefore:
It doesn't matter at all, EV wise.

QED

[Zag]

I agree it doesn't matter with a single out. That's why I used two outs in my example. Obviously, with only a single out, there is no way that you might waste one of your outs in the way I described. I'm not saying that you are wrong, only that your example doesn't do anything to convince me.

On the other hand, I did a little more monkeying and I am inclined to think I was, in fact, wrong all along. But I'm still not sure.