Baseball Between the Numbers: Why Everything You Know About the Game Is Wrong (Baseball Prospectus Team of Experts)
Among baseball analysts, sabermetricians are defined by their skepticism and demand for hard evidence: “You think pitching is more important than hitting? Why? By how much? How do you know? Why are you looking at me like that?”* They particularly focus on trying to prove or disprove the value of conventional wisdom—asking, for instance, whether traditional offensive statistics like batting average and RBIs, and traditional pitching statistics like ERA, are really useful ways of deciding how good a player is, and how big a role factors like luck and environment play in those statistics. It’s an indication of their success that stats like on-base percentage and slugging percentage are now shown regularly on broadcasts and at games, and that it’s now general knowledge even among casual fans that a player’s home park can have a significant effect on his superficial numbers (with Exhibit A being Coors Field in Denver).
Baseball Between the Numbers: Why Everything You Know About the Game Is Wrong takes a series of questions about the game and tries to answer them with hard data and statistical analysis. Some of these are old sabermetric canards (whether the RBI is a useful statistic, whether there’s such a thing as clutch hitting, when teams should really use their closer), but others are just interesting questions that get to the heart of evaluating players; managerial strategy; the relative value of pitching, hitting, and defense; and even questions of payroll and stadium financing. Is Barry Bonds better than Babe Ruth? Does batting order matter? Is Alex Rodriguez overpaid? Do catchers really have an impact on pitching performance? And my favorite: Why doesn’t Billy Beane’s shit work in the playoffs? (For those who haven’t read Moneyball, this was Billy Beane’s famous answer for why the A’s did consistently well over the course of a regular season and consistently failed in the postseason: “My shit doesn’t work in the playoffs.”)
The answers to these questions are invariably interesting, but just as interesting is the discussion of what the questions even mean. Most people, if asked whether Barry Bonds is better than Babe Ruth, would go straight for their career numbers, and compare their batting average, home runs, RBIs, and so on. A more serious fan might also look at stats like their OPS (on-base percentage plus slugging percentage). But the discussion here starts asking the really hard questions: Wasn’t Ruth facing much easier pitching than Bonds does? Doesn’t Bonds benefit from modern nutrition and training methods? What about the different ballparks they played in? And how do their various core stats translate into actually helping their teams win—which is, after all, their purpose in the first place? This is how you end up with measures like EqA (Equivalent Average, a composite measure of total offensive performance) and adjustments like the Time Machine Effect and Timeline Adjustment, which are complex, but necessary to get at the truth of the question.
The book did have a few minor annoyances. The most substantial of these was that the discussion of VORP (Value Over Replacement Player), one of their fundamental concepts, doesn’t occur until more than halfway through the book, meaning every time it came up prior to then (which was often), you always being referred to a future chapter—an idea as important as that really should have been thoroughly explained much earlier. Less substantial but still bothersome was that the 27 chapters are organized into an innings-outs scheme, from Chapter 1-1 to Chapter 9-3. This is clever, but is also always producing things like “Table 1-2.7”—which, in an already numbers-heavy book, is just too many numbers to simply identify a table, and became a little tedious after 100 pages or so.**
Balancing these somewhat, though, is that this is the first book I’ve ever seen to cite the Onion in an endnote, which you’ve just got to respect.
If you love baseball and are at least willing to put up with the math, you’ll like the book—I liked it a lot, but Maria passed it over to me right around Table BP.7 in the introduction (“Babe Ruth’s EqA, Adjusted for Time Machine Effect”). It also helps, I think, if you’ve encountered at least some of these ideas before, either in Moneyball or from reading a sabermetrically inclined sportswriter or two, but there is a helpful glossary if you get lost in the maze of VORP and WARP and PECOTA and BABIP and WinEx and SNLVAR. Despite the complex subject, though, you don't have to be a mathematician to understand it and enjoy it. And even if you don’t follow all the details of every last regression analysis or take the time to examine every last line graph, when you’ve finished, you’ll undoubtedly look at the game differently than when you started.
* Rejected opening #1: “If baseball analysts were states, sabermetricians would be Missouri—the Show-Me State.” What—I didn’t use it!
** And this may bug only me, but one of my pet peeves is the use of RBI as both the singular and plural form: 1 RBI, 120 RBI. I realize there are reasons for doing this (namely that RBIs appears to read literally as run batted ins), but I don’t happen to agree with them, and I frankly doubt that those who say “He had 120 RBI last year” would also say, for example, “During the war he and his unit were captured and held as POW.”