Follow the Batted Ball

By Scott Barzilla
Updated: June 9, 2011

One of the things those of us in stats all want to do is to go from step A to step F without explaining the steps in between. The main goal in this section is to avoid skipping steps. In the introductory article I mentioned batted ball statistics. Before I move onto those statistics I want to talk about what those statistics are and why they are important. First, let’s take a walk down memory lane.

Voros McCracken published an article on DIPS (Defense Independent Pitching Statistics) a scant ten years ago for Baseball Prospectus. Interestingly enough, the initial theory has been dwarfed in the meantime, but the proposition was simple enough. Pitchers had limited control over balls that are hit in play. McCracken found that strikeouts, walks, and home run rates remained fairly stable over time. Every other statistic had more volatility over time. Naturally, the same would be true for hitters.

In the intervening ten years, a couple of things have happened. First, people have noticed that some pitchers consistently have lower than normal batting averages on balls in play or consistently higher BABIPs. The same is true for hitters as well. So, like good scientists do, other sabermetricians continued to study it. At the same time, more and more technology and information became available. Sure enough, they discovered that McCracken was mostly correct, but that they could categorize balls in play in different categories.

Those categories include line drives, groundballs, and fly balls. Publications like The Hardball Times have found that batting averages on certain events vary. Naturally, we see a higher average on line drives than the other two batted ball types. Groundballs produce the second highest average. On the other hand, they also noticed that flyballs (while producing a lower average on balls in play) produced the second highest runs per event rate.

Further studies showed that those batted ball statistics were as stable as the so-called DIPS statistics. So, while we cannot predict what will happen when someone hits a line drive, groundball, or flyball, we can predict in general a level of success at the plate based on those numbers. We will look at those numbers from 2010 at all of the positions and on the mound to see how well the 2011 results fit.

From those statistics, other publications develop their own metrics to summarize the ability of a hitter or pitcher. For instance, most of the publications I mentioned last time publish a metric called Fielding Independent Pitching (or something similar in the case of Baseball Prospectus). Those numbers vary because the assumptions of those calculating the metric change. There is nothing wrong with that, but we want to do everything we can to filter out other people’s bias or assumptions no matter how intelligent they may be.

Below we see the 2010 major league averages for different batted ball events (save strikeouts and walks). We take these outcomes and use them when we look at batted ball rates for individual pitchers and hitters. Additionally, we will look at fielders at each position because this will play a big role in the success or failure of pitchers on those staffs. Similarly, teams can use those numbers to better judge who should be a part of those staffs. Naturally, teams can take a look at individual positions defensively and affect change that way. So, without any further ado, here are your league wide batted ball statistics for 2010.

Runs Per Event-ML

LD= 0.38

FB= 0.18

GB= 0.04

Other-ML

K%= 18

BB%= 9

HR/OF= .10

GB= 74% Outs

OF= 83% Outs