Any thoughts of querying Paul the psychic German sporting octopus about who will win the forthcoming World Series have been cruelly nipped in the bud. The eight-legged wonder, who gained worldwide fame by predicting eight consecutive World Cup matches, died overnight in his tank at just age two-and-a-half.
Obviously, no one predicted this.
So Paul will not get the opportunity to choose a sacrificial shellfish out of a tank marked "Giants" or "Rangers." He enjoyed his last three months of life as a retiree after going 12-for-14 in World Cup and Euro Cup soccer picks. Paul is the rare creature that truly went out at the top of his game. In retrospect, it'd be silly to ask a German psychic soccer octopus to predict World Series baseball games. What does he know of baseball?
Just after the World Cup -- which Paul, naturally, correctly predicted would go to Spain -- we asked San Francisco State mathematics professor Mariel Vazquez what the odds were of Paul's amazing run. She had an easy answer for us -- and a hard one:
For each World Cup contest, Paul was presented with two buckets ofThat one "experiment is all we're gonna get.
mussels. One bucket was designated with the flag of one World Cup
soccer team, and the other with the opposing squad. In each instance,
Paul chose the eventual winning side. Assuming each run was independent
of the others, the odds of this occurring are simply 1 over 2 to the
eighth power -- or one in 256.
however, isn't quite ready to sign off on the easy answer. Was each of
Paul's predictions truly independent? Did he go after a particular
bucket? Did he prefer a type of mussel in one bucket as opposed to the
other? Did he simply choose the left one over the right one? Since this
wasn't a scientific experiment, the level of scrutiny necessary to
answer these questions doesn't appear to have been present.
since the World Cup is limited to 32 teams and run every four years,
Paul hasn't really been put to the test. "The sample size for the
octopus was quite small," notes Vazquez. "If you were tossing a coin
eight times, you won't get four heads and four tails. You could have
all tails in one experiment of eight [flips]. But if you repeat the
experiment a million times, on average, you'll get a 50-50 response."
to be a killjoy, Vazquez admitted that Paul's psychic extravaganza has
proven "amusing," and she'd been pondering the odds of his feat even
before SF Weekly cold-called her. "But you can't draw a conclusion from one experiment."