+EV Game In San Diego

[ninehigh99]

Barona Casino in San Diego is dealing Mississippi Stud (a 4.91% house edge game) effectively face up, which yields a +1.5% player edge in a full game. Everyone is aware that playing face up helps the players, but they've been doing it for a year, and management isn't too worried about it.
I wrote up an advanced strategy card for this 6-way collusion, that's very simple to play:
http://discountgambling.net/mississippi-stud-ev-barona/
http://discountgambling.net/2009/08/16/mississippi-stud-barona-casino-ca/

There's usually a full game on Friday and Saturday nights. It's a shame there isn't a full game going on all the time, but that's the thing about critical mass ...

 

[Pro21]

Your numbers do not agree with the numbers in Exhibit CAA. If I had to bet on whose number were right, well....

 

[ninehigh99]

Which numbers? The base -4.91% house edge that I and the Wizard of Odds agree on, or did someone else publish a collusion strategy for seeing all 6 player hands in a game? I stand by my work, you can have my source code if you want it.

 

[Pro21]

Sorry, let me amend. The -4.91% is the same, and he gives no number for 6 player collusion, the book just says, "Yes, the game can be beaten, marginally, by sharing the information on the 12 cards from a full table of six players,..."

I was looking at some of your other posts.

 

[ninehigh99]

Yeah, it's pretty marginal (+1.5% on a $5 ante is 7.5 cents per hand). It's just that the Barona dealing procedure for this game (players are asked to turn up their hands after its dealt) makes "collusion" really easy. I bring my collusion strategy card to the table, and read off it out loud in front of the floor supervisor. No one cares, especially since I just bet $5.

Just mentioning this fact to others, if they care.

 

[SleightOfHand]

Can anyone confirm the strategy that the OP gave? Also, what kind of standard deviation does it have?

 

[ninehigh99]

Here's the C++ source code for the tool I used:
(Dead link: http://files.me.com/stephenhow/9qxv0o)

When you compile it, you should see:

Macintosh:Debug show$ ./miss_stud 
usage:   miss_stud -n<street> [-b<bet>] p0 p1 [c0] [c1] [k0 k1 k2 ...]
  p0 p1 is the player hand (required)
  -n2: 2nd street, no community cards
  -n3: 3rd street, c0 is only community card
  -n4: 4th street, c0 c1 are community cards
  [k0 k1 k2 ...] are known cards
  -b<bets>: amount bet so far
usage:   miss_stud [-r] -s<hands>
  -r: randomize

You can compare my results vs the Wizard of Odds for no confederate info. Then you can add known cards, and see the EV effect. Finally, you can run with the -a (monte carlo analysis) option to randomize the confederate cards, etc.