THE NEW MONEYBALL
One floor below street level on Yawkey Way, in his windowless corner office which looks out upon a bank of cubicles that could pass for a telemarketing firm, Red Sox general manager Theo Epstein is deep in consultation with a trusted adviser he identifies only as Carmine. Adrian Gonzalez may be the team's most valuable player and the object of an 11-year Ahab-like pursuit by Epstein, but the G.M. never makes a move without consulting Carmine, a five-year-old proprietary computer program that is the virtual brains of the Boston operation.
Epstein fell for Gonzalez long before he had Carmine. He first saw the All-Star first baseman in 2000, smashing what appeared to be long home runs far over the unmarked wall in left centerfield at Eastlake High, 30 minutes from where Epstein worked as director of baseball operations for the San Diego Padres. Epstein fell in love with that stroke the way baseball men, connoisseurs of kinetics, swoon over the arc of a pitcher's arm or a shortstop's sleight of hand. "I fooled you," Gonzalez revealed to Epstein this spring, a few months after Epstein at last fulfilled his quest by trading three of his top six prospects to the Padres for Gonzalez. "It was only 270 feet to left center."
What Epstein did know back then was that an information revolution was coming to the hidebound world of baseball. A Yale grad and law student in his late 20s, he was too young and too curious to be content with the status quo. Two years earlier, for instance, the Padres did not have any college statistics to prepare for the draft: The club didn't know something as basic as whether a player it was considering hit .250 or .350. Days before the draft, Epstein asked an intern to call college sports information directors across the country to get team statistics faxed to the office. Armed with at least some rudimentary data, Epstein ventured that a college pitcher with more walks than strikeouts, for instance, might not be the best use of a first-round pick.
In 2002, with the Red Sox under the new ownership team of John Henry, Tom Werner and Larry Lucchino, his former boss in San Diego, Epstein was hired as an assistant to Boston general manager Mike Port. The club's operations were so archaic that you could still hear the clackety-clack of Port tapping out memos and reports on his typewriter and, upon completion, the thwack of his stapler. In another part of the office, however, the future of the game was incubating among bright young minds not long out of elite colleges, including Epstein, Jed Hoyer (Wesleyan), Ben Cherington (Amherst) and, upon joining them in December of that year, Josh Byrnes (Haverford). They understood that statistics represented a reservoir of information—and a potential competitive advantage.
The young men had heaps of fun, working crazy hours and, to blow off steam, knocking golf balls around the office, playing football among the cubicles and celebrating big wins with postgame refreshments at Boston watering holes. Then one day in 2002, a best-selling writer by the name of Michael Lewis walked into the Red Sox offices and knocked the smile right off Epstein's face. Lewis was working on a book about baseball's nascent information age, but Epstein wanted nothing to do with him.
"I can't believe Billy is letting him write this book," he told his colleagues.
Billy Beane, Oakland's general manager, had granted Lewis access to his front-office operations, which meant revealing how the A's were mining information from statistical analysis, a tool used extensively at the time by only the Athletics, Indians, Blue Jays and Red Sox. "He's handing out the blueprint," Epstein told Hoyer.
Explains Epstein now, "It just seemed that [the book] would take a nuanced idea—while not a great secret, because Branch Rickey was using a lot of it a half century ago and Bill James had been writing about it for decades—and make it mainstream pretty quickly.
"The book hit The New York Times best-seller list. People who own baseball teams read The New York Times best-seller list. So they started asking questions about the processes their front offices were using, and it changed things really quickly."