GSN High Stakes Poker Thoughts

[shipitfish]

I saw two interesting things on GSN's High Stakes Poker the past two weeks.

First, on last week's episode, Jerry Buss went broke and asked for a loan from Daniel Negreanu. This itself wasn't all that surprising; that's typical of what I know about low, middle and high stakes poker. What was strange was that Daniel took it right from his stack. I have been taught many times that one can't take money off the table, especially to loan to another player. I assume no one complained here for a few reasons:

• Daniel had the whole table covered anyway, and still did after loaning money to Jerry.
• They want Jerry to stay in the game because his money is tight-weak dead money.

Still, it seems strange that the impact on cross-table loans in poker and NL HE in particular wasn't discussed at all.

Second, while Sammy Farha is not that popular among most, I saw him do a pretty honorable poker act in this week's episode. In the final hand of the night, which was prophetically AA vs. KK, Sammy holding KK asked Barry Grenstein if he wanted to run it twice. Of course, Barry said no. Then, when Sammy flopped a K and became the favorite again, he offered to run it twice from that point too. That's pretty nice of him, given Barry didn't give him the same chance when he was a favorite.

Of course, Sammy knew that Barry was likely to refuse, since it would look just awful for Barry to say no preflop, but yes once he was a huge underdog. So, it's basically a psychological free-roll for Sammy to make the offer. But it seemed really nice, anyway, on the surface.

Comments

Re: dealing it twice

[dkwad.livejournal.com]

Ok, thanks. I didn't mean to suggest the pros were doing it weird; I really have no experience with this. I just went off on unshuffled vs. shuffled because it's easier to think about than unshuffled vs. running it once, and shuffled is definitely the same as running it once.

Re: dealing it twice

[dkwad.livejournal.com]

Whoa, I guess that intuitive comparison in the K77 example isn't so intuitive after all. When I do the math on that, it says KK wins twice as much if he runs it twice instead of once! But in situations where the underdog has more outs, running it multiple times seems to hurt rather than help him (because he burns up outs when he hits twice in the same deal?).

I'll report back when I've worked out more examples and double-checked the math. Maybe I'm just stoned and unable to multiply correctly.

"Maybe I'm just stoned and unable to multiply correctly."

[dkwad.livejournal.com]

Ok, I have looked at several situations and triple-checked all math. I was flat wrong; dealing it twice without reshuffling is the same as dealing once.

Sorry for the noise!

Re: "Maybe I'm just stoned and unable to multiply correctly."

[shipitfish.livejournal.com]

No problem at all. Could you post the detailed math if you get a chance?

Re: "Maybe I'm just stoned and unable to multiply correctly."

[(Anonymous)]

Here is the EV of dealing it twice, that I think everyone wants to know but is too lazy to work out. When you deal twice, you do not change the expected value! Let's say that two players are all-in on the turn, and only the river card remains. Let's say that player A wins with 13 of the remaining 43 cards, and player B wins with 30 of them. So the EV of player A dealing it once is (+13-30)/43 = -17/43*1. The EV with dealing it twice is (the chance of winning it twice)*payoff + (chance of winning once, losing once)*payoff, and (chance of losing twice)*payoff.
winning twice:
13/43*12/42 * 1 +
one lose, one win
(13/43*30/42 + 30/43*13/42)*0 +
two losses
30/43*29/42 * -1
= (156 - 870)/(43*42) = -714/(43*42) = -17/43
So the "scoop the pot" theory is BS, since although you decrease your odds of scooping, you also decrease your odds of losing both by the same amount. So definitively, only the variance is changed, not the EV. There's the math to prove it! But then again maybe I'm stoned and made a mistake :)
Jojo Mcbean
www.wikipollo.com