February 3, 2002 Story Search

Malwitz: Is winning on football wagers an impossible goal?

##### Published in the Home News Tribune 2/03/02
Steve Skiena, a professor of computer science and author of "Calculated Bets" (Cambridge University Press), puts his equivalent of a cigarette warning label on page 187 of his book: "Those who gamble to make money -- you are likely a sick individual and need help."

Today is Super Bowl day, when Americans will wager an estimated \$4 billion on the game. Some will be in the form of friendly office bets. Some will be with break-your-kneecap bookies.

But Skiena, a 40-year-old East Brunswick native who is spending the year teaching in Israel, is too smart to bet on football. He may also be too tuckered to watch. The Super Bowl kicks off at 1:30 Monday morning Jerusalem time.

Skiena's speciality is jai alai, a sport confined to Florida now that Connecticut has given it up. The scoring system -- too complicated to explain here -- is designed for wagering. Like the lottery or roulette wheel, jai alai follows certain laws of probability trends, and mathematicians have always been fascinated by probability.

In "Calculated Bets," Skiena, who teaches at the State University of New York in Stony Brook, wrote about his computer program for jai-alai betting that turned \$250 into \$1,800" -- before "The system started breaking down."

Mathematicians have long been fascinated by games of chance. Seventeenth-century French mathematician Pierre Fermat, author of the legendary Fermat's Last Theorem, developed his probability theory, said Skiena, "When one of his friends was losing at dice and wanted to know why. There are few areas in the world where the mathematical structure is as clear as gambling games."

The lottery, roulette wheel and jai alai are gambling games. Football is not: The ball is oblong and bounces funny.

Pierre Fermat would likely not touch football, and neither should anyone apart from just-for-fun office poolers.

Skiena tried football. In 1977 he was attending East Brunswick High School and wrote a computer program to pick winners. To its credit, said Skiena, East Brunswick was ahead of the computer curve, with better computers than he had the next year as a freshman at the University of Virginia.

The high school computer was about the size of a refrigerator, with discs the size of large pizzas. But it was marvelous, said Skiena. "You told the machine to do something, and it did it."

The Home News was interested enough in the computer kid from East Brunswick that it let him pick National Football league games that fall. He had a record of 135-70 simply by picking the winners of games.

The next year, his picks appeared in the Philadelphia Inquirer, but he had to pick against the spread. He was right only 46 percent of the time.

"The bookies are real good at selecting point spreads. They know half the people will be wrong, and they keep a fraction of all the bets," said Skiena.

"The house always has the advantage. If the house didn't have the advantage, it wouldn't be there."

Today? Skiena doesn't have a clue. Neither do I.

St. Louis, 30-22. But don't bet on it.

Rick Malwitz's column appears Tuesday, Thursday and Sunday. (732) 565-7327; e-mail Rmalwitz@thnt.com

from the Home News Tribune

Published: February 3, 2002