Saber-Slant #6: The Value of Batting Average

Baseball is a simple game, right? There’s a guy throwing a ball and another guy trying to hit that ball with stick (colloquially called a bat in the US). The battle is between these two. Simple, right?

Well, not really. Were baseball to be that simple, batting average (BA) would be a great measure of a baseball player’s talent. Fortunately for most fans, there’s a little more to it than that. There are things hitters do that cannot be captured by batting average. Why, then, is BA still being used as a catch-all statistic of a player’s value? What good is BA if it doesn’t tell you how good a player is?

Well, like most stats, batting average is not designed to tell you that.

If we could talk to batting average…

If batting average could speak, it would probably lament its misuse among mainstream media and fans. See, batting average is a very good stat at telling you what it purports to tell you. It (or at least a logical version of it that includes sacrifice flies) tells you how many hits a player got out of his total opportunities that ended in a ball in play or strikeout. It’s not the most natural definition, but it makes sense to some degree.

The problem is not with what batting average tells you, but how people use and interpret it. Batting average never told you to use it as a gauge of player talent. Why? Let’s use an example similar to one I used a few weeks back. Here are three players from 2009 who hit .300.

Player A: .300/?/?
Player B: .300/?/?
Player C: .300/?/?

Not much to differentiate between them. What else can we measure offensively that could help us with that goal? Well, batting average tells us about hits over AB, and that gives us some idea about how often a player avoids outs. Outs are very bad offensively, since they are the only limiting factor in a baseball game. But walks and hit-by-pitches also are non-out events that can occur, and they are not taken into account by BA. That’s why we have the bane of Moneyball-haters everywhere, OBP.

Player A: .300/.366/?
Player B: .300/.394/?
Player C: .300/.331/?

But each event that avoids an out is not created equal. Walks are not worth as much as singles, and singles aren’t worth as much as home runs. The ability to move runners further is also a valuable part of offense in baseball, and that can be estimated by looking at SLG.

Player A: .300/.366/.532
Player B: .300/.394/.489
Player C: .300/.331/.425

Even though each player had the same batting average, each had a drastically hitting profile. Player A, Justin Upton, had a lot of power and decent ability to avoid outs. Player B, Shin-Soo Choo, drew more walks and thus made less outs and had decent power. Player C, A.J. Pierzynski, did not belong in the same conversation as those two, as he did not excel in anything but collecting hits.

Why use batting average then, when OBP and SLG are better? Well, I wouldn’t say that they are “better.” Each stat is descriptive; that is, it is designed to tell you one aspect of a player’s game. You would not make a value analysis of a player based on just BA, just like you wouldn’t do one using just OBP or just SLG. You would be throwing away information that is important to a player’s offensive contribution. Taken as a whole, we can make better evaluations, but using any descriptive statistic like this on its own will naturally leave out other details.

So what’s the value of batting average? It tells you something about how often a player collected hits. It tells you nothing more and nothing less, and we should use it as just that. If you want to know how well Ichiro Suzuki gathers hits, BA is your friend. If you want to tell me Ichiro is a great hitter because of his BA, then you are treading on shaky ground.

Tags: A.J. Pierzynski, Justin Upton, Shin-Soo Choo, Triple Crown