Most of the great mysteries of modern baseball thought come from trying to understand the more—and I hesitate to use this word, because I think it lends itself to the wrong impression—intangible facets of the game. Many sabermetricians tend to just ignore them because they aren't easily quantified. This is rather sophomoric; smart, because you don't want the ambiguous clouding what is crystal clear, foolish, because these less defined qualities certainly have some function in performance and you don't want to ignore anything vital. Whenever something comes along that promises to clarify and quantify some aspect of the game that had previously been out of reach, my ears perk up.
That's just what has happened with The Hardball Times' Chris Jaffe's new book, Evaluating Baseball's Managers: A History and Analysis of Performance in the Major Leagues, 1876-2008, currently available. Chris was kind enough to forward the content discussing the guys who've managed the Mets: Casey Stengel, Gil Hodges, Joe Torre, Davey Johnson, Bobby Valentine, Art Howe.
Managerial evaluation has been one of the most difficult things for sabermetricians to accomplish, leading most to almost write off the role of managers, often imagining that one is as good as another. This probably isn't accurate; fans certainly witness the tactical roles managers play, but they also serve as motivators, communicators, and personality handlers, and it certainly stands to reason that some guys are better at these things—or at least better suited to particular situations—than others.
The problem has been that the non-tactical aspect of the job is so difficult to get your arms around (and the tactical parts aren't that much easier). I have some background in psychology, and researchers there often face similar problems. Psychological concepts are often easy to envision but difficult to measure. However, a clever scientist will figure out a way to design a condition under which he'll be able to indirectly measure the effect—it just takes some careful planning. Jaffe comes up with an intriguing method: he looks at five ways managers could make an impact on the run and win totals of their teams. First, he looks at how many runs individual hitters outperformed their expected output based on the player's two preceding years and two succeeding years in terms of runs created. Next, he does the same with pitchers using component ERA. Third and fourth, he looks at how the team, as a whole, outperformed its expected total number of runs scored and allowed based on runs created and component ERA. And finally, he looks at how the team outperformed its pythagenpat projections. While there's certainly going to be some noise in the data, I do think it's a wonderful approach to something previously thought beyond the grasp researchers.
This isn't all Jaffe looks at either. While the run totals above or below expectation may tell you something of a manager's efficacy, it doesn't say anything about how he managed. To that end, Jaffe has built a "tendency database" to highlight any philosophies a particular manager may have had. How often did he bunt? Did he make lots of in-game substitutions? For older managers, did he like to save his best pitchers for the most difficult opponents? Jaffe neatly summarizes these in a short paragraph along with the numerical data. For instance, here's what Jaffe says about Valentine's tendencies:
Team Characteristics: Bobby Valentine loved to make in-game changes. His bench players had a good chance to be called on as a pinch hitter or defensive replacement. Even if they did not receive many starts, Valentine ensured they would not become rusty. His pitchers struck out plenty of opponents.
After that, Jaffe concludes with a lengthy comment about the manager's career, often focusing on something in particular that it made it unique or especially arresting. This is where the content really shines. He talks about Casey's ability to leverage the Yankees' Big Three, Hodges's reliance on his bench, Johnson's aversion to the intentional walk. Jaffe was especially surprised by how well Art Howe repeatedly performed with young teams, often exceeding reasonable expectations. He also notes that it made him spectacularly ill-suited to manage those veteran Met teams. It all makes for a very interesting read, especially considering how tragically ignored managers are in most baseball literature and analysis.
If the Mets chapters are any indication of what we can expect from the rest of the book, Evaluating Baseball's Managers figures to be treasure. I certainly recommend it to any baseball fan with a fondness for history or any affinity for thinking about the game in new ways. Other than Bill James's book on managers—difficult to find these days—there's really nothing else like it out there.