Moneyball - Michael Lewis [34]
It was James’s first sustained attack on baseball’s conventional wisdom. He concluded it with a question:
So if we can’t tell who the good fielders are accurately from the record books, and we can’t tell accurately from watching, how can we tell?
“By counting things,” he replied. Then he went on to propose a new statistic—the “range factor,” he called it. A player’s range factor was simply the number of successful plays he made in the field per game. There were obvious problems with range factors, too—an outfielder on a team staffed by fly ball pitchers, for instance, had more opportunities to make successful plays than an outfielder on a team staffed by sinker ball pitchers—but the details of the thing didn’t matter. What mattered was James’s ability to light a torch in a dark chamber and throw a new light on a dusty problem. He made you think. There was something bracing about the way he did it—his passion, his humor, his intolerance of stupidity, his preference for leaving an honest mess for others to clean up rather than a tidy lie for them to admire—that inspired others to join his cause. The cause was bigger than fielding statistics. The cause was the systematic search for new baseball knowledge.
The cause wasn’t original. James was hardly the first person to notice that there was still stuff to be figured out about baseball, and that the game’s underlying rationalities might be discerned through statistical analysis. Going right back to the invention of the box score in 1845, and its subsequent improvement in 1859 by a British-born journalist named Henry Chadwick, there had been numerate analysts who saw that baseball, more than other sports, gave you meaningful things to count, and that by counting them you could determine the value of the people who played the game. But what got counted was often simply what was easiest to count, or what Henry Chadwick, whose reference point was cricket, had decided was important to count.
Chadwick was the critical figure in this history. To anyone who asked, “How could baseball statistics be so screwed up?,” Henry Chadwick was usually the beginning, and occasionally the end, of the answer. Chadwick’s stated goal in counting the events that occurred on a ball field was reform: he wanted players to be judged by their precise contributions to victory and defeat. He was as upset about the immorality he witnessed on the baseball diamond as he was about the drinking and gambling he found on the city streets—about which he also never tired of complaining. He longed to affix blame and credit for baseball plays, and to do it, he grossly oversimplified matters. Fielding errors were just one example of Chadwick’s moralistic mind at work. Another was his interpretation of the base on balls. In cricket there was no such thing as a walk: Chadwick had to get his mind around a new idea. The tool was ill-designed for the task: Chadwick was better at popularizing baseball statistics than he was at thinking through their meaning. He decided that walks were caused entirely by the pitcher—that the hitter had nothing to do with them. In his initial box score Chadwick recorded a walk as an error; even in the later box scores, after he had listened to, or at least heard, the obvious objections from others, Chadwick never credited the hitter. He simply removed the walk altogether from the record books. “There is but one true criterion of skill at the bat,” he wrote, “and that is the number of times bases are made on clean hits.” Enter the batting average, ever since the chief measure of a player’s offensive value.*
The more you examined these old measurement devices, the less apt they seemed. Chadwick, with help