Random number troubles

nobody at shell.portal.com nobody at shell.portal.com
Wed Apr 20 10:57:11 PDT 1994


_The Toronto Star_
Wednesday, April 20, 1994

`Computer nerd' outsmarts casino

Wins $200,000 pot -- twice in a row

MONTREAL (CP) -- Ask Daniel Corriveau how he beat
staggering odds to win $400,000 at the Montreal
casino and he'll talk about a butterfly flapping
its wings in Bejing.
    After the computer consultant hit a $200,000
jackpot twice in a row playing electronic Keno 10
days ago, the casino shut down the popular
lottery-type game and started an investigation.
He has yet to collect.
    "I'm confident I will get the money," Corriveau
said.  "It's a normal process for the casino to be
investigating."
    Celebrated by Quebecers as a mild-mannered genius
who beat the system, the province's latest hero
is a computer nerd who claims to have used "chaos
theory" to defy mind-numbing odds at the casino.
    The arcane mathematical concept, which the 40-year
old Corriveau found himself expounding on television,
is based on the notion that random-looking data aren't
so random.
    One of the theory's axioms is that if a butterfly
flaps its wings in Bejing, it will have an effect on the
weather system in New York City.
    The rules of Keno are less esoteric.  Placing bets
of between $2 and $5, gamblers try to pick some of the
20 numbers that are drawn from an 80-number pool in
the computerized game.
    On April 10, Corriveau managed to pick 19 of 20
numbers twice in a row, a feat not accomplished even
once since the casino opened last October.
    Corriveau said he discovered "a bug in the system"
that made the Keno odds more player-friendly.
    Corriveau visited the casino about a dozen times
over four months, writing down the winning sequences
of numbers.  The brainy bettor plugged the data in
to his home computer and put on his thinking cap.
    "I found the same 19-number sequence twice in 240
draws," he explained, "That proved the weakness in
the system."






More information about the cypherpunks-legacy mailing list