The Top Pickers vs. the Pack

So far, sports and stocks present fertile ground for these online experiments at identifying experts, much as they were for the wisdom of the crowd a decade ago, when the Web was still young. And like the wisdom of the crowd, the wisdom of the few is likely headed beyond sports and stocks, as the Web holds the potential to identify savants in many fields.

Exactly where this approach might turn next is unclear, but PicksPal is preparing an entertainment site designed to find people skilled at predicting movie blockbusters and busts, celebrity magazine covers, Academy Award winners -- even Jessica Simpson's future love interests.

Tom Jessiman, the founder of PicksPal, said he resolved to develop a way of finding individual expertise after he soured on conventional sports pundits while working at CBS Sportsline.com.

"I thought there's got to be people out there driving cabs in D.C. and lobstermen in Maine and on an oil rig somewhere that are better than these guys. I wanted to find out where these guys really are," he recalled.

Jessiman launched his site a year ago hoping to attract large numbers of contestants passionate about betting on sports, not for money but for points that translate into prizes and bragging rights. His plan all along was to use this pool of players as a way to fish out experts and then sell their predictions. He initially thought the experts would vary week to week as their fortunes shifted.

"But a very small set of people keep winning," Jessiman said. "It kind of blew me away."

Last month, PicksPal began identifying the 30 most successful players over the previous five weeks. Whenever these experts together place enough of their accumulated points on a specific bet, it qualifies as a "genius pick" and is offered to the public for $19.95. Jessiman said the genius picks, which average five a day, are hitting 70 percent of the time.

Acevedo began competing on PicksPal in March and established his genius credentials by finishing as a weekly winner on the site last month. He admitted he was surprised he had outperformed so many others. His fellow teachers, mostly young women, have little serious interest in sports, he said, and no idea of his swelling Internet reputation.

A longtime sports fan whose tastes run from English soccer to NASCAR, Acevedo said he brings the same intuitive approach to sports picking as he did to his recent studies at Ramapo College in New Jersey, where he majored in philosophy and literature. "I go with my gut," he said. "It doesn't feel like I'm a genius."

The notion of separating the best from the rest is heresy for those who advocate the wisdom of crowds. According to its proponents, a large number of diverse, independent individuals will typically outdo experts because even experts lack perfect information and make mistakes. But with a crowd, the many small pieces of information and perspectives held by individuals come together to form a more complete picture while the mistakes can cancel each other out.

In a classic example cited by James Surowiecki in his recent book, "The Wisdom of Crowds," a professor asked 56 members of his class to estimate the number of jellybeans in a jar. The average of the guesses came to 850, only 21 beans off, and was better than the individual estimates of all but one of the students.