Where do advanced statistics belong in the game of baseball? This is the question that has, for the most part, consumed baseball discussions for the past 10+ years. By and large, the Sabermetrics community (the term coined by the pioneer of advanced statistics, Bill James) has gained ground over the traditional scouting community during these discussions, causing it be more and more tolerated, and even adopted by major league front offices. Billy Beane (General Manager for the Oakland Athletics organization) is the poster-boy for Sabermetrics among baseball executives, but he is hardly alone in his adoption of advanced stats. The Rays, similar to the A’s in terms of market size, have used advanced stats very effectively over the past few season, appearing in one World Series and winning multiple AL East titles. Theo Epstein was fresh out of college when Moneyball caused a stir in 2002, and he has since led the Red Sox to two World Series titles, while openly admitting his use of advanced stats. In addition, many organizations in baseball even now have an entire staff of statisticians crunching numbers, analyzing players, and looking for any deficiency in the market to exploit. Terms like on-base percentage, slugging percentage, batting average on balls in play (BABIP), and walks + hits per 9 innings pitched (WHIP) are now known by the average fan and quickly referenced in baseball discussions. Most people get it, some still don’t.
Movies have been filmed on this topic: “Moneyball” chronicles the unique ideals that turned the 2002 Oakland A’s into a 90 win team, a year in which they lost Johnny Damon, Jason Isringhausen, and Jason Giambi to free agency, simply because they couldn’t afford to pay them. On the other side, “Trouble with the Curve” is about an old scout in the Braves organization who is caught in the transitional period from old scouting techniques to the recent statistical analysis. In the end, the statistician basically wastes the 1st pick of the draft, while the old scout turns out to be correct in his estimations. Undoubtedly there will be more movies to come on this subject.
Very outspoken in his disdain for Sabermetrics (and Moneyball in particular) is Joe Morgan, Hall of Fame 2nd baseman. Morgan’s main argument is this: how can a bunch of scientists know anything about a game they’ve never played? Morgan would argue that he knows more about the game of baseball than any stathead could possibly know, simply by having played the game. Morgan has some legitimate points; statisticians don’t know the odds and ends of how to play baseball, or even how to manage baseball, despite all our equations for certain situations. However, that fact doesn’t negate the validity to sabermetrics at all. Sabermetrics is, most simply, a tool. It can be used to predict future performance and to evaluate and compare past performance. That is why advanced stats and old-school scouting can coexist. Statisticians have definitively proved that teams should draft college players who are more seasoned than high school players, but how did Albert Pujols get drafted? By a midwest region Cardinals scout, surveying the area for an outstanding talent. No computer needed to tell that scout that Pujols was special; the ball sounded different coming off his bat and went a whole lot farther. There are countless other examples of scouts who found can’t miss prospects in some unlikely places, prospects who went on to change the future of their organization.
Back to statistics. One of the longest debates inside the grand debate is the value of wins. Adam Wainwright voiced his thoughts in one interview only two weeks ago. In This piece by Jesse Spector of the Sporting News, Wainwright says,
“I think there is, let me say, some validity to Sabermetrics, but I do think there needs to be more open-mindedness from those sabermetricians as well, as well, the Brian Kennys out there that will disregard all old-school baseball stats. Wins do mean something. There is an art to winning baseball games.”
The value of the win can be debated all day, but Wainwright’s overall point is absolutely correct. There needs to be more open-mindedness from the sabermetricians out there, many of whom would see the win completely tossed from the back of a baseball card, as well as losses, RBI and errors. These stats may not be as effective in measuring player performance as many other advanced stats, but does that mean they should be altogether disregarded? I like sabermetrics, enjoy learning about new ways of analyzing players, but I’m willing to admit that there are some aspects of the game that can never be fully captured in a number. That is why I believe that the sabermetricians are often at fault in this discussion; though their points are valid.
There probably won’t be complete harmony between two sides of this discussion, but when there is, success is sure to follow. Perhaps Theo Epstein has put the relationship between stats and scouts most acutely:
“Performance analysis is one lense, and scouting observation is the other lens, and the GM needs both lenses to see.”