Colletti Benefits From Writers' Bias
by Glenn Dickey
Apr 04, 2006

THERE’S NOTHING more distasteful to most sportswriters than a new idea. The latest one is teams using computer models as part of their talent evaluation, and never mind that NFL teams were doing it more than 40 years ago and the A’s have been doing it for a quarter-century. It still counts as a new idea for writers. I mean, Connie Mack never used computers, did he?

The latest beneficiary of this attitude is Ned Colletti, now the general manager of the Dodgers. Writers often refer to him as a “baseball man,” though his background was in public relations, not in any kind of talent evaluation. He replaces Paul DePodesta who writers thought of as an intellectual because he graduated from Harvard with an economics degree. In fact, though, DePodesta had gone to work with the Cleveland Indians after graduation and worked up to being an advance scout for two years. Not that any of this really matters. Sandy Alderson came to the A’s from a law firm and seems to have survived the terrible handicap of not having a baseball background. Intelligence trumps background.

Computer evaluation actually started in the NFL in the early ‘60s when the Dallas Cowboys, San Francisco 49ers and Los Angeles Rams formed a company to pool information. Each team set up their own computer model with the shared information. The models were different because each team assigned different values to some of the components, such as competitiveness. The teams also set up physical models for players at the different positions.

The A’s started their program when the Haas family bought the team before the 1981 season. There were so many stories about it at the time that the A’s finally stopped talking about it because the notion arose that they were using computers in the dugout, which was absurd.

But the Giants, on a much quieter basis, were also using computers for statistical information breakdown in the ‘80s. Marty Lurie, who now does pre- and post-game shows for the A’s, was part of the Giants’ program and he remembers that Roger Craig didn’t want any of the information he was using put up on the message boards.

One example of the way a computer could be used to make baseball evaluations is to program it with information on ballparks and the effect on a hitter of being in a strong lineup. That type of information would probably show that Colletti has put far too much stock in the numbers Bill Mueller compiled in Boston, playing in a hitter-friendly park and in a lineup so strong that he batted eighth at times in a year in which he won the American League batting title. Playing in a weaker lineup and in a pitcher-friendly park, Mueller will probably revert to the good fielder with little power that he was in his early career with the Giants.

You don’t even need a computer to track Nomar Garciaparra’s offensive stats in recent years and come to the conclusion that he’s not likely to have a great offensive year. Giants fans will remember that Kenny Lofton was a defensive liability as a center fielder here four years ago. Hard to believe he’s gotten better with age, and he’s now on the disabled list. Colletti watched Brett Tomko with the Giants – but signed him to a Dodgers contract, a real head-scratcher.

Yet, baseball writers are falling all over themselves to predict great things for the Dodgers because of Colletti. “He sees the field,” wrote one San Francisco columnist. Down, computer!

THE OAKLAND A’S are the prime example of computer analysis in baseball, and many of the journalistic attacks on De Podesta probably came because he was a surrogate for Billy Beane, for whom he was an assistant before going to the Dodgers.

The A’s use computers for both economic and baseball analysis. Because Beane has worked with a limited budget, he’s had to figure a way of getting the most bang for his buck, so he uses computer models to determine player values.

One example: For years, the A’s concentrated on collegiate players in the draft because they were a greater value. College players are further along in their development than high school players – Huston Street was only one year removed from college when he won the American League Rookie of the Year award last season – and if they can’t help the A’s, they can be used in trades.

Last year, though, Beane determined that high school pitchers were undervalued, and he invested heavily in them in the draft.

That’s where the “Moneyball” approach is. Though writers often refer to Beane’s baseball philosophy as “Moneyball”, that’s really a different thing.

Beane believes in the importance of on-base percentage, as Alderson did before him – and as Branch Rickey did when he was building the St. Louis Cardinals in the ‘30s. Other baseball people have depended on that, too, though the statistic has only recently been part of the media record. You can check this yourself with league statistics. If you compare a team’s batting average and on-base percentage with their runs scored, you’ll see that on-base percentage is a more accurate way of determining a team’s potential for scoring runs.

Every team uses computers to analyze statistics. The basic question – as it was in the ‘60s with the 49ers, Cowboys and Rams – is how much value you assign to different statistics in setting up the model you want. Beane’s model gives a high value to power hitting, a low value to base stealing – and a high value to on-base percentage. With the model established, he can use the model to search for players who fit his profile.

That’s only part of it, though. Despite some silly stories about this, the A’s do not “pick players out of the computer.” Players who fit their model are thoroughly scouted before the A’s draft or trade for them. Beane knows the value of scouting; his first job after he retired as a player was as a scout for the A’s.

NONE OF THIS is secret knowledge. Beane is as candid with and accessible to the media as any sports executive I’ve ever known.

Still, many writers resist this because computer analysis makes baseball seem a much different game than the one they’ve known since they were 12. Now you know why Norm Van Brocklin said that, if he ever needed a brain transplant, he wanted one from a sportswriter -–"because I’d know that it’s never been used.”


E-MAILS: I try to answer most of my e-mails but occasionally, I get those annoying messages that I have to fill out a form to be able to answer. I got one today on a message about Mike Montgomery. I do not fill out these forms, ever, so if you want a direct answer, change your system.

What do YOU think? Let me know!

© Copyright 2017 Glenn Dickey. All Rights Reserved. The material on this site is protected by US Copyright Laws and cannot be used without the express written consent of the owner. Site design and maintenance by 5 happiness webmaster.