card dealing

[RobcPettit]

Hi, I hoping somebody may have worked, or may have an idea on how to
work on this idea. Not expecting an answer but though Id throw the
question out here.
Ok. Im dealind cards. The games baccarat. All Im doing is if the banker
has a 3 (could be any card though) and te player has a 4 (again
anycard really), is with excel dealing all possibilities for the player
to go on and win. In a1 I put players card, b1 = bankers card. So a2
starts 4A, B2 = 3A. Then can a2 beat b2 yes/no. Then b2 = 32, question,
33 question etc al the way to 3kk (max 3 cards). then a2 changes to 42
and it all starts again. As you can imagine this can take a good 20
secs to run on low cards. I realise its a difficult question to ask but
can this be speeeded up. Im thinking that rather than adding the cards
to a range before asking the question Id put them into an array. At the
momment I deal cards one at a time because it alternates player/dealer
upto 3 cards. Im using select case to make the questioning easier, and
ive got screen updating set to false. Any ideas appreciated.
Regards Robert

 

[Guest]

If you are trying to work out the odds, someone beat you to it:

http://www.gambling-baccarat.com/baccarat_odds.htm

 

[RobcPettit]

Hi, thanks for your reply. Im am calculating odds, but for the game as
it unfolds. Cards are dealt and not replaced, therefore the odds change
as the game progresses.
Regards Robert

 

[Guest]

If you are interested in learning about casino games get John Scarne’s
"Complete Guide to Gambling". He is considered the world’s foremost gambling
authority. He explains all the games, which ones are the best to play, and
which bets give you the best advantage over the house (i.e. the least likely
to lose your money!)

As for a Mini-Baccarat-dealing spreadsheet, what I did was simply randomize
a large number of cards (as many decks as you like) and put them down column
A, keeping in mind that 10 through King have a value of zero. Then I went
down the column dealing the first four cards to Player-Banker-Player-Banker.
Card five is either given to Player or Banker (or kept for the next deal)
depending on the specific rules of the game. Likewise for card six. Only
four, five, or six cards will ever be dealt. My columns come out looking
like this:

(Sorry, this font does not align well)

Deck
Player (tot) Banker (tot)
-- -------------- ----------------
7 7
0 0
0 0 (7)
3 3 (3)
8 8 (1) P
4 4
9 9
3 3 (7)
9 9 (8) B
4 4
9 9
9 9 (3)
9 9 (8) B
9 9
0 0
9 9 (8)
0 0 (0) P
6 6
5 5
2 2 (8)
2 2 (7) P
2 2
1 1
1 1 (3)
6 6 (7)
3 3 (6) B
8 8
9 9
0 0 (8)
5 5 (4) P
9 9
6 6
3 3 (2)
0 0 (6)
7 7 (9)
7 7 (3) P

Since the Hit/Stand rules are cut in stone there’s not much thinking
involved in the game. Just pick one or the other and pray like heck.

I don’t recall the odds of winning betting on Banker or Player but I do
recall that Banker wins more than 50% of the time (52.5% or 55%?) and Player
wins less than 50% (47.5% or 45%). The key is that the house charges a 5%
commission whenever you win on a Banker bet, thereby offsetting the fact that
Banker wins more often.

Essentially, the house doesn’t care which bet you place - the money wagered
simply moves back and forth between the bettors while the house keeps the
commissions...MOST of the time. There is a phenomenon though, that the house
hates: it’s called a "run".

Now and then there will be a long string of Banker or Player wins, 8, 9, 12,
15 in a row. When the players detect those runs they really come alive with
some huge bets, hollerin’ and high-fivin’ all the way! Now the house really
does lose money, and if it’s one thing casinos are not in the business of
doing, it’s losing money. If there was a way to break up those runs, believe
me, they’d find it.

Well, a few years back they introduced automatic shuffling machines to the
Mini-Baccarat games. About that time I began noticing a decrease in the long
Player/Banker runs of days past. Maybe it was my imagination, or maybe the
machine was too consistent in it’s "riffling" mechanism, or maybe...Nah,
casinos wouldn’t actually cheat would they?...maybe the shuffling machines
actually have high-speed LASER image readers that can detect the card face
and pre-arrange (stack) the deck!

"But," I thought, "they always offer one of the players the option to cut
the deck. Wouldn’t that disrupt a pre-arranged shuffle?"

So now we come to the purpose of my BacDealTest.xls: I wanted to see if
cutting a stacked deck would make a difference to the outcome. I added a few
macros to allow me to select a row within the deck, cut and reprocess the
deals, and compare the sequence of Player/Banker/Ties between the two runs.
Surprise! Regardless of where the deck is cut the original sequence
eventually falls back in place, usually within about two dozen deals, often
less.

So, a deck COULD conceivably be arranged to prevent any disastrous long runs
of either Player or Banker, and cutting the deck does not undo the
arrangement.

The comparison of wins looks something like like this:

Original order vs. after cut
..
..
..
P
P
P
B
B
P
B P
B B <--
T P
B P
P P <--
B T
B P
T B
P B
P P <--
P P <--
B P
P P <-- (From here to
B B <-- the end of
P P <-- the overlap,
P P <-- i.e. at least
P P <-- half of the
B B <-- deck, they
T T <-- matched)
B B <--
B B <--
B B <--

 

[RobcPettit]

Thanks for the reply Charlie. Very interesting. The way youve layed
your cards out has given me a couple of Ideas as well. Not quite the
sane as what I was looking at, but I can see how to improve mine. One
point thats I havnt made clear in my post is that Im trying to
anticipate what the odds will be for whatever cards pops out of the
deck. Eg if player had an ace then gets a 2, the odds would be
different (if only slight) if player gets a 3, or 4 etc. What Im trying
to predict is if the odds will rise or drop given the turn of a
particular card. Or at least the percentage of risers/fallers. I can
sort of do this, but Im looking at ways to speed up the macro. I bought
a couple of books on calculating the odds, thinking there was a way of
doing this with combin in excel, but I havent found it yet. Again
thanks for the time, has given me an idea.
Regards Robert