The sabermetric revolution in baseball has changed the way we look at the numbers compiled on the field. When it comes to player evaluation, we can look at scouting reports and statistical analyses.
But many of the most important people on a baseball team never set foot on the field.
So how do you analyze managers?
Until now, few comprehensive studies have been attempted to analyze managers or coaches throughout history. The economist J.C. Bradbury did a series of studies of former Atlanta Braves pitching coach Leo Mazzone in 2005. And in in 1997, Bill James wrote his Guide to Baseball Managers, one of the first serious attempts to subject baseball's managers to sabermetric analysis.
Chris Jaffe goes even further as his new comprehensive book functions both as a history of baseball's managers and as an analysis of the 90 or so most noteworthy men in the profession — the 77 managers who lasted a decade on the job and the 12 most noteworthy of the rest. Jaffe is a historian by trade, and his grasp of baseball history is both impressive and infectious. His book is the definitive reference on the history of baseball managers and it's an invaluable book for anyone interested in baseball history or baseball analysis.
It's also great fun to read. Jaffe's prose is engaging, not pedantic. He explains himself clearly, and where necessary admits to possible flaws in his arguments. His methodology is complex, but he's able to explain it relatively simply. His analysis of manager quality rests chiefly on what he calls the Birnbaum Database, a dataset developed by baseball researcher Phil Birnbaum — who intended it to measure luck, not managerial skill. The Birnbaum Database attempts to predict how far any given player season is from the norm that might have been expected from his previous two seasons and following seasons. Jaffe simply assigns credit to the manager for the deviation from the norm of the players under his care.
For example, Bobby Cox has a tremendously high score with individual pitchers, because for a decade and a half pitchers have come to Atlanta and had career years. In Jaffe's accounting, Ron Gardenhire is the most successful younger manager, in large part due to his uncanny ability to fashion shutdown bullpens from the scrap heap and win division title after division title with occasionally middling talent.
Jaffe's other two analytical innovations deal with each manager's tendencies by lending a particular eye toward starting pitcher leveraging, which used to be a major part of the game but disappeared in the 1960s. He logs leveraging by measuring the quality of opponent that each pitcher faced, and the rest of the manager's proclivities with what he calls the Tendencies Database, by which he assigns a score to each of the 77 longest-tenured managers based on their teams' ranking in various statistics, relative to the rest of their leagues. For example, Bobby Valentine's teams used the most pinch hitters in history; Billy Martin's teams had the most pitcher appearances on no rest; Bruce Bochy's teams (most of which featured Trevor Hoffman(notes)) had the greatest percentage of wins saved. Jaffe uses these rankings to paint a picture of each manager's signature style in relation to his peers and in context of history.
My favorite parts of the book are his analyses of early managers, some of whom literally developed elements of the modern game. Cal McVey — whose managerial career was basically finished by 1879 — was apparently the very first manager ever to platoon his pitchers, so every left-handed specialist making the major league minimum owes him a debt of gratitude. Gus Schmelz not only was one of the pioneers of the bunt, he may have effectively invented spring training. Charles Comiskey, who went from pitcher to manager to owner, all with fabulous success, was the first man to build a team entirely around pitching and defense.
The book isn't flawless. Despite a delayed publication that made the book's stats only current through 2008, it has its fair number of typos, missing words and punctuation. The typos generally don't affect readability, but they're noticeable (and Jaffe noticed many of them too, correcting them on his blog). His tendency-based analysis of many managers is fine, but his capsule bios spend more time on what that manager was most likely to do, and less time on exactly how good that manager was at what he did. He never quite offers up his own managerial Mount Rushmore — according to the scores he crunches from the "Birnbaum Database," the top four are Joe McCarthy, Tony La Russa, Bill McKechnie and Walter Alston, with Bobby Cox a distant 11th, but he refrains from indicating whether he agrees. Jaffe's book is the last word on managers for now, but it by no means should be the last word on the subject — it deserves and demands to be picked apart, critiqued and followed up.
This is one of the best baseball books I've read in a long time, a serious effort by a good writer with a love of history and stats and a fascinating subject that hasn't been studied much. With just a month left until the season starts, there's never been a better time to read it.