clock menu more-arrow no yes mobile

Filed under:

Measuring Offense with Process, not Outcome

Are the hitting statistics that we use flawed in their methods?


A double comes in many forms. A blooper, a blast in the gap, a slow roller down the line that turns into a "hustle double," and even a ground-rule double. All of these are considered equal when evaluating a hitter, but should they be?

When a batter steps up to the plate, his ultimate goal is to make hard contact. The successes and failures of a hitter should be based on the kind of contact they made, not whether or not the fielders were able to reach it or not. My argument is as follows: offensive statistics, as they are commonly used, are flawed because they do not paint an honest picture of the contribution of the hitter.

Take, for example, the following two plays. In the first video, we see Jedd Gyorko hit a bloop double that ends up crediting him with an extra-base hit, while in the following play we see Mike Napoli robbed of an extra-base hit by Colby Rasmus and ultimately get penalized for the at-bat.

Classic offensive statistics describe the outcome of the plays, but that still doesn't help evaluate the batting performances.

I believe that a hitter should be judged on a select group of things:

  1. Velocity with which the ball is struck by the bat.
  2. The angle at which the ball is hit upon contact with the bat.
  3. The timing and ability of a hitter to keep the ball in fair territory as well as make contact.
  4. The discipline of a hitter, whether they are able to swing at good pitches and lay off bad ones.
For the purpose of this article, we will ignore the plate discipline and contact aspects of evaluating hitters, and instead focus on the batted ball aspect. In the two examples used above, Mike Napoli hit the ball not only with great force, but he also did so at an angle which propelled the ball to the warning track. Although this ball hung up long enough to be caught by Colby Rasmus, a safe generalization to be made would be that balls hit with similar force and at this angle tend to be extra-base hits.

This is obviously difficult to measure in a more precise fashion due to the lack of public HITf/x data, and because we lack the ability to actually calculate that data we have to stick to making these assumptions.

Looking at the ball hit by Gyorko, we can see that he did not hit the ball with great force, but that he also hit the ball at a very high angle. This type of a hit generally is caught by a fielder, but because the right fielder not only failed to catch the ball, but also allowed it to skip past him, Gyorko is credited with a double. In a community such as the online baseball research one that preaches the evaluation of process over outcome, why is it that we continue to judge batters on their outcomes?

As I alluded to before, it will be difficult for the public to make great strides in this area without access to HITf/x information, but I feel that it is important to note the flaws in these numbers. This year's SABR Analytics Conference featured a presentation by Ben Jedlovec or Baseball Info Solutions, who presented research similar to what I seek about building batted ball profiles for hitters.

Essentially, this article was a long-winded way of posing the following question: Are the current offensive statistics we use adequate? Should we be crediting batters based more on the quality of contact they make rather than the outcomes of their at-bats?