In last year’s smash hit Moneyball, author Michael Lewis wove a compelling narrative about how a statistically-oriented revolution that took hold in the early ’80s among baseball followers had finally penetrated the front offices of a major-league team, the Oakland A’s. In doing so, he brought a stereotypically nerdy take on the game, called sabermetrics (defined by Bill James as “the search for objective truth about baseball”) to a mass audience. The book polarized the world inside baseball, but fans of all stripes ate it up.
In a new book, The Numbers Game: Baseball’s Lifelong Fascination with Statistics, author Alan Schwarz has capitalized on Lewis’ success, not only revealing the roots of that statistical revolution (which is really more of an evolution), but also illuminating how the game’s growth in popularity over the past 150 years has been thoroughly intertwined with the development of numbers to describe the action. Schwarz profiles dozens of obscure but important roleplayers within the movement, a few mentioned in passing within Moneyball, but many unfamiliar to even the most diehard fans. In doing so, he’s created an essential book for any baseball library, one that simultaneously makes for breezy summer reading and holds up as an essential piece of research.
After a brief introduction to the primitive box score as it existed in the 1850s (listing “hands lost” or outs, and runs for each player on both sides), Schwarz begins his exploration with Henry Chadwick, the game’s statistical forefather. A Britsh-born cricket reporter for the New York Clipper, Chadwick was drawn to the way box scores (which had developed a bit by the late 1850s) could illustrate the skills of the players and offer a means for analyzing their play. Chadwick, who desired a complete record of what happened on the field for every contest, invented the nine-by-nine grid within which observers could note the results for each batter using a system of letters and numbers. Thus recorded, the action could be distilled into an elaborate box score which provided positives (fielding plays) and negatives (batting outs), a perfect form of double-entry bookkeeping that survives to this day.
Schwarz traces the evolution of baseball’s rules and the manner in which Chadwick, who had a moralizing streak a mile wide, would devise new statistics to reward the manner of play which he deemed worthy. Chadwick calculated stats such as batting average and fielding percentage, added up total bases and fielding chances per game (which James reintroduced as Range Factor nearly a century later), and figured whole hosts of other numbers. From there the author goes on to detail the introduction of other crucial stats such as Runs Batted In and Earned Run Average and the men who championed them. He also reveals how tangled the game’s record-keeping was back in those days, with sources constantly disagreeing about season totals, and numerical chicanery obscuring batting titles and other records. A new niche arose to keep track of year-by-year records of individual players and teams, along with lists of noteworthy accomplishments. As World War I came along, Al Munro Elias and his brother Walter became the game’s first official third-party record keepers, creating an organization whose relationship with baseball survives to this day.
In his rapidly-moving tale, Schwarz surveys the various personalities who strove to expand understanding of the game. He introduces F.C. Lane, a proto-sabermetrician who attempted to quantify the run-value contributions of hitters by calling attention to their extra-base hits, revealing batting average as merely window-dressing compared to power hitting. He brings us Allan Roth, who presented himself to Brooklyn Dodger GM Branch Rickey as a whiz who could give the Mahatma an edge with his detailed breakdowns of situational hitting splits — how a hitter performed at home or on the road, in day games or night games, against lefties or righties, in all manner of count, baserunner, and out situations — and became the first full-time statistician employed by a big-league team. Working as Rickey’s right-hand man, Roth not only helped call attention to the importance of on-base percentage (see this seminal 1954 Life magazine article, “Goodby to Some Old Baseball Ideas”), he also armed neophyte broadcaster Vin Scully with such tidbits as Gil Hodges’ 0-for-20 performance against Johnny Antonelli and prepared statistical notes for the press prior to each ballgame. I have my own fondness for this one-man bureau; when I became fascinated by stats as kid, my father would tell me that I should aspire to be “the next Allan Roth.”
In the early 1950s, a new way of delivering stats to fans began when they were included on the back of Topps baseball cards. Schwarz tells that tale as well as the one of the kid, Hal Richman, who invented the tabletop Strat-O-Matic baseball game. From there he shifts gears to focus on a handful of scientists who turned their attention to baseball in their spare time, bringing with them a rigorous approach and a zeal for large-scale studies. One was George Lindsay, an officer in the Canadian army who, when he wasn’t trying to figure out Cold War-era aerial defense maneuvers explored the truly important questions. Scoring over 1,000 games with the aid of his retired father and tabulating the totals in his spare time, Lindsey explored platoon advantages and created the first base-out matrix, which said, for example, that an average of 2.3 runs could be expected to score in a situation with no outs and the bases loaded, and allowed analysts to quantify the advantages (or disadvantages) of the stolen base and the sacrifice bunt (the Lindsay portion of the book is excerpted at Baseball America, where Schwarz is a regular columnist). Another such scientist was Earnshaw Cook, an aristocratic Baltimore metallurgist who wrote a dense statistical tome called Percentage Baseball that prefigures Bill James’ Baseball Abstract work — complete with a formula to estimate the probability of scoring a run that washes out to something similar to James’ Runs Created approach (the Cook portion is excerpted at ESPN.com, where Schwarz contributes weekly articles). Yet another pair were brothers Harlan and Eldon Mills, former World War II combat pilots who rented time on an IBM mainframe and, based on data from the Elias Bureau for which they paid dearly, ran thousands of simulations for each possible inning-base-out situation to quantify the values of each event and its impact on winning.
If you’re a fan familiar with the sabermetric smorgasbord available on the web, some of this should ring a bell. Contemporary statheads have explored exactly the same turf — base-out matrices, win expectancy, linear run estimators — many of them without any awareness of these protosabermetricians who slaved away for hundreds of hours to do the same work. In the ’50s and ’60s, not only was such work considered frivolous by the masses, but the means to collect and process the data was beyond the realm of all but a select few. Today anybody with a spreadsheet can run numbers on an entire league of players, and anybody facile enough with a database can mine the game’s history from his or her desktop.
My favorite portion of The Numbers Game concerns the production of MacMillan Publishing Company’s first edition of The Baseball Encyclopedia, published in 1969. It’s fitting the book debuted that year, because for the way Schwarz portrays it, the achievement was no less significant than putting a man on the moon. Led by David Neft, a devoted staff of researchers scoured microfilm archives across the country, reconstructing thousands upon thousands of run-scoring innings from long-forgotten ballgames not only to iron out discrepancies in record-keeping, but also to tally stats such as RBI and ERA which didn’t become official until the ’20s. The hefty tome was the first book to be entirely typeset on computer, and tested how well the machines could manipulate and store large amounts of data. A tongue-in-cheek ad, sadly suppressed, presented The Baseball Encyclopedia next to the King James Bible, declaring the massive, 2,338-page behemoth “VOLUME TWO.”
Inevitably, a considerable chunk of the book revolves around Bill James, who brought a new mode of inquisitiveness to the game and created new tools to examine a broad variety of questions. James pops up not only as the creator of the Abstract series (Schwarz runs down the man’s ten most significant statistical discoveries) but in several other places — climbing the bestseller list, fighting the Elias Bureau for the right to obtain the detailed proprietary data it kept, creating a network (called Project Scoresheet) as an alternative means to collect that data, arbitrating a dispute between the volunteer-based Project Scoresheet and the proprietary STATS, Inc data house, and eventually ascending to the front office of the Boston Red Sox. Schwarz casts James as the hero in confrontations with some of the book’s less savory characters, such as Seymour Siwoff, the modern-day head of Elias, and John Dewan, whose conflict of interest fueled the STATS-Scoresheet split and who later cut STATS co-founder Dick Cramer out of the company.
Schwarz goes on to sketch out the myriad ways in which the statistical landscape changed post-James. A fledgeling national newspaper called USA Today capitalized on the zeal for more statistics by revamping the traditional box score. A writer named Daniel Okrent, who had introduced Bill James to the masses in a 1981 Sports Illustrated profile, created a game called Rotisserie Baseball in which participants “bought” players in an auction and placed based on the actual stats accumulated by players in various categories. STATS Inc. created a piece of stat-crunching software, programmed by Cramer, called Edge 1.000, which the Oakland A’s and several other teams bought in the mid-Eighties. STATS programmer Pete Palmer joined forces with several members of the Society for Baseball Research (SABR) to create a more accurate baseball encyclopedia after Major League Baseball refused to recognize discrepancies in Ty Cobb’s hit total and other anomalies; that book became Total Baseball. A national Public Radio correspondant named Eric Walker wrote a small book which Oakland A’s legal counsel Sandy Alderson happened across; the book extolled the virutes of on-base percentage and challenged much of baseball’s conventional wisdom. When Alderson became GM, he hired Walker as a consultant and began implementing his ideas. This, as much as Bill James, was the root of the Moneyball movement, a fact that Walker himself, who felt slighted by the brevity with which Lewis dealt with his role in A’s evolution, has sought to illuminate. And then of course, there’s the Internet, with its capacity not only for up-to-the-minute stats (often via STATS) but also the goal of preserving every box score and game event from baseball’s lush history via Retrosheet, and soon even more unique data coming from Major League Baseball’s website.
Schwarz has done an admirable job in tracing the evolution of our statistical obsession with baseball and in saluting the noble efforts of oft-forgotten men who helped shape the way we view the game today. He reveals that our fascination with the game’s numbers is itself nothing new, and that it’s our means of producing, digesting, and storing ever-more-granular data where the real evolution lies. Though the author keeps a brisk pace, occasionally some of his transitions between topics seem forced, and at times the books chapters feel a bit too serialized — as though a reader might encounter them weeks or months apart and need reminding of who’s who all over again. But really, those are minor quibbles compared to the book’s overwhelming strengths. If you eat box scores for breakfast, The Numbers Game is for you.