The Difference Between Range and Positioning

Perhaps one of the biggest objections people have with the current state of defensive metrics is that the stats don’t account for the starting position of the defender. Shift plays are excluded from the calculations, but when a center fielder plays in 20 feet, the system doesn’t know that he’s starting from a different spot than the average center fielder, which could obviously lead to some imprecise accounting.

This is true for every position except pitchers and catchers, as the starting location of the fielder influences the probability they will make a play, independent of anything they do from the moment the ball is pitched. If you start out of position, even if you run at top speed and take a perfect route, you might not be able to offset the initial disadvantage of not being in the right spot to begin with. This creates problems, but there’s a lot of nuance to these problems that are worth discussing, even as we get closer to having StatCast and rendering the discussing irrelevant (we hope!).

Let’s start by addressing the point of defensive metrics. They are either used to account for how well a defender prevented runs or assessing their defensive ability. One is about accounting for past success and one is about inferring true talent. Positioning impacts these two functions differently, so let’s discuss them in turn.

When talking about measuring prevented runs, being out of position should matter. If you are a shortstop who starts three steps to the right of the default position and a ball is hit hard to the left of the default position, the fact that you started out of position had a negative impact on your team’s ability to prevent runs. It’s a real thing that happened, just like starting in a good position helps your team prevent runs.

The question in this part of the discussion is whether it is fair to credit/debit the fielder for positioning when the coaching staff plays a large role in setting the fielders. Jhonny Peralta is generally considered to be someone who gets to a lot of balls based on solid positioning, but is that Peralta’s doing or his coaches? The positioning needs to be accounted for at some level to properly account for the impact defense had on the team, so our question is whether we should reward Peralta or someone else, but it should be rewarded.

When talking about assessing talent, the conversation turns. In this instance, we want to know about a player’s skills and ability to execute. We care how well Peralta can get to a ball hit six feet to his left at 95 mph. We don’t really care where he happens to be standing when assessing his range, per se. Positioning only matters here if it is the player’s skill rather than the coach’s skill. If you get to a ball because someone put you in front of it, that’s a good thing, but the problem is that the defensive metrics don’t know where you were standing and can’t tell if they should give you points for range.

Put another way, if you got to the baseball because you have great reaction time, UZR gives you the exact same score as if you got to the ball because you were literally standing in the perfect spot to start. When we were talking about accounting for past runs saved, the difference doesn’t matter. When you’re trying to figure out how good a player is, it matters a lot. We care about “why” in this context. Does the fielder have good range or are they well-positioned?

To some extent, range and positioning are very different. Andrelton Simmons has better range than I do. It’s a slam dunk, obvious fact. But while Simmons is also better at positioning than I am, it’s plausible that I could be positioned as well as Simmons. I would never get to a ball in the hole, but you could shade me in that direction just like you shade him in that direction.

The important thing to remember is that the stats can’t tell if you made (or missed) a play because of where you started. It essentially thinks you’re rangy if you’re smart enough to guess where the ball is going. So the run is saved all the same, but when it comes time to evaluate a player’s skill based off that data, it might be misleading if the positioning is coming from the dugout. We can’t isolate quickness, route running, and positioning with our current stats, so they’re all basically grouped together.

This is a problem for a lot of people who want “range” to mean ability to cover ground, when it practice it means the ability to convert outs as a function of distance from the average starting position. It’s a small distinction, but a necessary one. It’s really a language problem, more than anything because we’re taught that range means running speed when the stats think of it as the distance from the origin that you can cover.

Let’s use a crazy example. Imagine you’re a center fielder and for some reason, you’ve decided to play on the warning track. Now let’s imagine a routine fly ball is hit to shallow center field. You react to it immediately and get to top speed very quickly. You do everything right and make a sensational diving catch. In this example, you only get a small fraction of a run saved because based on average positioning, that’s a very easy play to make, when in reality, it took superhuman effort to execute the play.

Now imagine the exact same scenario, but you miss the catch by an inch. In this situation, you get assessed a rather large negative defensive score because it would have been an easy play on average, if you started in the right spot. If you fail to convert this ball into an out, the problem is where you started on the field. Was that your fault? Maybe. If it was, then the system worked perfectly. If it was a coaching error, then what we’d ideally want to do is give you lots of credit for almost making the play and somehow dock the team for poor coaching. We don’t have the granularity of data to do that, so it all gets thrown into the player’s score. This where people have trouble because if you’re watching the play, the skills on display were excellent and you can’t see why the fielder shouldn’t get credit. But it reality, the hit was allowed because of a defensive flaw.

So in reality, defensive metrics, even in non-trivial samples aren’t just about the player’s skills or how they played on a given set of days. The way I like to think about them is as an accounting of the Player-Team coverage of that position. Peralta at shortstop in 2014 was a measurement of how well the Cardinals prevented runs at shortstop while Peralta was in the game. Most of that is about Peralta, but some of it is about how well he was coached, how well the catcher communicated the pitch that was coming, and then all of the other normal noise involved in defense.

That might not be super satisfying for everyone, but it’s the best we’ve been able to do with the data we have so far. That might all change when StatCast comes online and we can give credit based on actual starting position rather than assumed starting position. Right now, we can’t locate the starting point of every fielder because we don’t have complete information (read: camera angles). Soon, we should be able to do that and some of these issues can be resolved, at least as far as true talent is concerned.

A player’s true range is a function of how well they move, but their effective range is a function of range+positioning. Right now, we measure the latter and that leaves us with some issues. This is why we don’t ever take the metrics as gospel, but they do a decent job painting a general picture. This week, we should start to see more and more StatCast data, so it’s going to be very interesting to learn just how good or bad the metrics are at estimating player defense.