Baseball: Next Season’s Highlights

The crystal ball of baseball has hit the stands, and Steve Goldman, editor and coauthor of Baseball Prospectus 2008: The Essential Guide to the 2008 Baseball Season, has gazed into the murky depths of the coming season.

Goldman, with Joe Sheehan, executive editor of Baseball Prospectus online, will be at Barnes and Noble at BU at 6 p.m. tonight, March 12, to answer the big question: will the Sox take the series again? Goldman talked with BU Today recently about some of the book’s predictions, the methods used, and unpredictable intangibles.

BU Today: Let’s cut to the chase: will the Sox take the series?
Goldman: They are the team to beat.

Does past performance predict future results?
It doesn’t completely, but it certainly influences the results. From our point of view, we look at who’s on the team, and we project how those players can do. There are a couple of things about the Red Sox that come into play. One, they have a management that has made intelligent decisions. Thanks to Theo Epstein and company, the Sox will come into the next season with virtually the same team, so in this case, past performance has a lot to do with how the team will do.

How much irate mail does Baseball Prospectus get when the predictions come out?
Not as much as you would think. We have a high-tone class of reader. The thing I find ironic, not just with the predictions, but with any article we run, is we’ll get two pieces of e-mail, and one will say that we’re obviously biased toward the Red Sox. The other will say we’re against the Red Sox. We publish about 1,500 articles per year on the Web site, without editorial bias or directive from on high, just whatever the writer thinks. There are always two opposing camps. The great thing about Baseball Prospectus, the thing I enjoy, is that we’re people who argue based on the merits, and we’re open to persuasion. If you can make a good case, I’m open to revising my opinion.

Can you explain how you come up with the Baseball Prospectus predictions? What’s the algorithm?
We study runs scored and runs allowed. We look at individual players, using our system, called PECOTA, which stands for player empirical comparison and optimization test algorithm. It’s one of the best projectors of player performance available. It takes the aggregate of the runs scored and the runs allowed and comes up with the predictions.

Is PECOTA always right?
It’s correct very often. We run comparisons at the end of the year for overall team performance and player performance and see how well it predicted. I disagree with it all the time; in certain places I feel that it’s off. For instance, I think it erred on the high side with the Mets this year, because I thought they were a brittle team. But that’s not something that PECOTA knows. It takes the runs scored and runs allowed, and it can factor in a player’s age and if he’s prone to breaking down, but it doesn’t have the instinct that a human has. The Mets are probably going to break even. I thought with the Yankees it might be overly generous also. With the positive projections on the pitchers, I thought it was too optimistic.

So you’re saying that the predictions would be accurate only if the season were played in a vacuum. Yet people are champing at the bit to buy it. Why?
Because it’s fun, and because they can argue about it. People buy it for their fantasy draft. But if they’re looking for some kind of certainty, our joke is that it’s ‘deadly accurate.’ I mean, PECOTA can’t predict if someone will get run over by a car. Stuff happens, and you have to include those possibilities.

And finally, the eternal baseball question: if you could see one game from any time, who would you want to see play?
Well, if it’s any place in time, and if we’re restricting the time machine to baseball, I’d love to see Ted Williams when he was swinging. He retired 10 years before I was born. I’ve seen films of that swing. Someone recently said they’d like to see Satchel Paige pitch in the prime of his career, and I thought that was a great answer. Satchel Paige was prone to doing dramatic things, like waving off his infield so he could strike out hitters without help. Ironically, I think Ted Williams would have enjoyed that. Maybe seeing them play each other — that’s what I’d like.

Kimberly Cornuelle can be reached at kcornuel@bu.edu.