The Colorado Rockies and defensive efficiency

The Colorado Rockies defense was on full display on Tuesday night. There were leaping stabs, diving catches, and slick pick ups all over the field. A representative example:

We’ve been hearing a lot about defense from the ROOT broadcast this season because the Rockies are currently first in the National League in fielding percentage; however, this doesn’t mean the Rockies have the “best defense in the league,” as the broadcasters are so fond of saying after every spectacular play. This is because fielding percentage doesn’t tell the whole story.

One of the foremost challenges in baseball statistics has always been quantifying defense. Errors have been tracked since the dawn of the game, and likewise fielding percentage, which is simply the number of outs made minus errors committed divided by number of outs made. There are obvious limitations to this: If Nolan Arenado goes to his left to get a ball that Kris Bryant would never have reached, but then makes a throwing error, are we really willing to say that Bryant is the better defender, as his fielding percentage would now be higher? By no means!

There have been many attempts to quantify this difference (metrics like Defensive Runs Saved and Ultimate Zone Rating, for example) but as long as we’re talking about teams and not individual players, there is a far simpler approach.

The goal of a defense is to convert balls in play into outs. Obviously players want to avoid errors, but they want to avoid them so they can record outs and keep hitters from turning into baserunners. Whereas fielding percentage tells us how well a team avoids errors, defensive efficiency tells us how well a team converts balls in play into outs.

Defensive efficiency can be calculated as (1 - BABIP), if BABIP all you have, but it misses a lot. Baseball-Reference’s Defensive Efficiency, however, gets deeper than that by considering every out made in the field in order to give a defense appropriate credit. That means accounting for double plays, outfield assists, and even how many errors result in an extra baserunner (71%, on average). The result is what they abbreviate as DefEff.

If a team never allowed a baserunner or only allowed home runs, DefEff would be 1.000 and you would have a very good defense. League average over the past several years has been .688, though last year the Chicago Cubs had one of the best defenses of all time and finished with a .728 DefEff. In contrast, the Rockies finished 28th in baseball in 2016 with .667 DefEff.

This year is a different story. Obviously the fact that the Rockies have kept their errors to a minimum is indicative of a more proficient defense, but we can also see that they have a more efficient defense as well. Their .702 DefEff entering Wednesday was tied for the third best in baseball, 14 points better than league average. The 35 point difference from last year is the second largest turnaround in baseball (oh, by the way, the Diamondbacks have the third largest).

What’s behind this defensive turnaround? Obviously a reduction in errors helps: The team was 12th in the NL in fielding percentage in 2016. That can be largely attributed to Nolan Arenado, who has just one error this year after 13 last year. Nolan is on pace to have his best defensive season since his rookie year, but that doesn’t seem to be significant enough to pull up the entire team’s efficiency this dramatically (though, considering Nolan’s year so far, I wouldn’t put it past him).

Meanwhile, the pitching staff doesn’t provide very many clues for why the defensive efficiency is so much better. The team’s GB/FB ratio is about the same (1.60 in 2017, 1.66 in 2016) while the ground ball rate is actually lower this year (48.8% to 49.1%). They are also giving up less hard contact (30.6% compared to 32.4% last year), but on its own that doesn’t tell us as much as we’d like.

But there’s also the matter of “efficiency.” If you want to be truly efficient at turning balls in play into outs, ideally you’d turn one ball in play into multiple outs. Last year, the Rockies were 11th in baseball with 149 double plays; the Rangers’ 191 was the most in baseball. So far the Rockies have turned 85 double plays, which is the most in baseball in 2017, and a pace of 186 over 162 games. This includes five double plays started in the outfield, compared to just two all last year.

Wednesday’s blowout against the Diamondbacks dropped the team from third to ninth in defensive efficiency: still good, but not as significantly good. With a ground ball heavy pitching staff (they are third in baseball in ground ball rate) and a defense that is proving adept at turning those ground balls into multiple outs, however, we can expect that to rise.

There is a lot to the old adage “defense wins championships.” But next time someone invokes that about the Rockies, remember that it’s not just about avoiding errors, but converting more balls put in play into outs that makes this Rockies team different from their predecessors.