MLB

Who Should Have Won the American League Gold Gloves?

Perhaps more than any other MLB award, the Gold Gloves are doing right by statheads.

The Gold Gloves are getting it right.

MVP voters continue to emphasize RBI and batting average while marginalizing on-base percentage and defense, and the Cy Young still revolves around ERA despite its faults, but the Gold Gloves might be the gold standard for statheads on the awards circuit.

Part of this is by design, of course, as the SABR Defensive Index counts towards at least a quarter of the voting, with managers and coaches making up the other 75%.

SDI is an aggregate of the top existing batted-ball and play-by-play defensive metrics, including Ultimate Zone Rating and Defensive Runs Saved, as well as Runs Effectively Defended, Defensive Regression Analysis, and Total Zone Rating.

FiveThirtyEight found that there was only a six-run difference between the ideal team based on SDI and the actual Gold Glove Winners.

Let’s go around the horn and look at what the numbers say about each Gold Glove winner in the junior circuit, and who the top players were in terms of SDI, UZR and DRS.

Pitcher: Dallas Keuchel

SDI: Keuchel
UZR: N/A
DRS: Keuchel

Keuchel was the top defensive pitcher in the Majors in terms of SDI and DRS (UZR is not measured for pitchers).

He did not stand out for his ability to control the running game (0 stolen base runs saved), but his ability to simply field balls in play was the difference here. He led all pitchers in plus/minus runs saved (the measure of a defender’s “range and ability to convert a batted ball to an out”) by four runs.

In terms of overall DRS, Keuchel saved 13 runs, and the gap between he and Mark Buehrle (who finished in second in the AL) is eight runs; this is equivalent to the difference between Buehrle and 29th-place Scott Kazmir.

Catcher: Salvador Perez

SDI: Caleb Joseph
UZR: N/A
DRS: Joseph

Catcher is the first of three positions in the American League where the voters and SDI disagree.

Perez finished seventh in SDI and second in DRS.

The Royals backstop was second in the league in runners caught stealing but also allowed the second most stolen bases. In terms caught stealing percent, Perez was ninth among the 11 American League catchers who caught at least 800 innings; runners were successful against Perez 69.5% of the time.

Joseph was only seventh though, allowing a success rate of 67.3%, but makes up the difference in terms of DRS via blocking.

Also, Joseph was one of the better catchers in the league at “stealing strikes.”

Joseph was sixth in the league in framing runs above average according to StatCorner and 14th in the Majors per Baseball Prospectus.

Perez, meanwhile, ranked was second to last in the American League at StatCorner.

First Base: Eric Hosmer

SDI: Mike Napoli
UZR: Travis Shaw
DRS: Mark Teixeira and Albert Pujols

Everybody disagrees!

We can start off by saying the voters probably erred by selecting Hosmer.

By UZR and DSR, he was not much more than an average fielder (coming in at 1.0 run above average in both stats, ranking 14th and 15th in the league, respectively).

Hosmer only made four errors and tied for the league lead in fielding percentage, but these are not ideal metrics for judging a player.

By making the plays hit to him, Hosmer was a league-high 3.0 runs above average in terms of Error Runs but gave almost all this value with a lack of range. He was 2.0 runs below average in terms of the range component of UZR.

Napoli led the league in range (6.2 runs above average), followed by Shaw and Pujols, who were tied for second with 4.0. Napoli made 7 errors though, which cost his team 1.3 runs above average, so he slides to second in overall UZR.

Shaw edges out Pujols and Napoli thanks to being above average in terms of both error runs and double play runs above average.

In terms of making great plays, seemingly no one was better than Teixeira.

Per the Inside Edge data via FanGraphs, Teixeira made a league-high 40.0% of plays that had a 10%-40% chance of success and was third in terms of plays with a 40%-60% success probability.

Second Base: Jose Altuve

SDI: Ian Kinsler
UZR: Kinsler
DRS: Kinsler

This is the final disagreement of the day between the humans and SDI, but it seems like the voters got this one wrong.

Kinsler led the SDI with a score of 10.7, five points more than second-placed Eric Sogard (Altuve came in third with a score of 4.6).

The gap was even wide in terms of DRS and UZR. Kinsler was worth 19 DRS, a full 11 runs better than second-place Logan Forsythe, and the gap between the two equal to the gap between Forsythe and 10th-place Dustin Pedroia (among second basemen who played at least 500 innings).

Altuve tied for sixth with 3 DRS.

In terms of UZR, Kinsler’s 6.3 was 0.9 points better than Hanser Alberto, while Altuve tied for sixth with 3.2.

Third Base: Manny Machado

SDI: Machado
UZR: Adrian Beltre
DRS: Beltre

At the hot corner, we have the voters and play-by-play data versus ball-in-play data.

It’s hard to get too up in arms over this one, even if you are a big proponent of DRS and UZR; both guys were excellent this season.

Beltre led the league with 18 defensive runs saved and an 11.8 UZR. Machado was second in DRS (14) and third in UZR (8.4, behind Josh Donaldson’s 9.2), respectively.

Machado led all qualified third basemen with a .777 revised zone rating, so that helps explained how the other components of SDI outweighed DRS and UZR.

Per UZR, though, Beltre’s range gave him a two-run edge over Machado, and Beltre also edged Machado in DRS thanks to a three-run advantage fielding bunts.

Shortstop: Alcides Escobar

SDI: Escobar
UZR: Francisco Lindor
DRS: Lindor

Play-by-play versus batted ball data: Part Two!

By the batted ball data, Lindor is the clear winner here, leading the position in UZR (10.5) and DRS (10).

Escobar had a league-best RZR, and also tied for fourth at the position in UZR, so that explains the high SDI ranking.

Interestingly though, DRS hated Escobar, rating him at a run below average.

Escobar’s troubles on double plays hurt him here, as while it cost him an eighth of a run in terms of UZR, he loses two full runs worth of DRS.

Left Field: Yoenis Cespedes

SDI: Yoenis Cespedes
UZR: Cespedes
DRS: Cespedes

Consensus!

Despite getting traded to the National League at the trade deadline, Cespedes logged 865 innings in left with the Tigers, enough to dominate his peers at the position.

Cespedes beat out Eddie Rosario (who came in second in the three metrics) by 3.2 points of SDI, 7.6 runs worth of UZR, and 1 DRS.

His well-documented strong arm was second to Rosario in terms of the throwing component of UZR, while his range contributed 8.1 runs above average to the Tigers (nearly four runs better than Jackie Bradley, who was next highest in the UZR range component).

Center Field: Kevin Kiermaier

SDI: Kiermaier
UZR: Kiermaier
DRS: Kiermaier

Kiermaier was not only the top center fielder in the American League, but by SDI, UZR, and DRS, he was also the best defender in baseball.

His 29.2 SDI score nearly doubled NL-leader Brandon Crawford's score, while his 30.0 UZR was almost 10 runs better than Jason Heyward’s, and his 42 DRS was 17 runs better than that of Andrelton Simmons.

If you combined the DRS scores of Crawford and Heyward, you would equal Kiermaier.

Kiermaier’s DRS is the highest on record (since 2003), while only three players have had a higher UZR in a single season (since 2002).

Among AL centerfielders, Lorenzo Cain had the next highest UZR; Cain was worth 14.3 runs above average, almost 10 runs fewer than Kiermaier range score alone!

Right Field: Kole Calhoun

SDI: Calhoun
UZR: Calhoun
DRS: Lonnie Chisenhall

Calhoun takes the prize here, thanks to the combination of his range and arm.

He led the league in UZR with 13.8, with 7.8 runs coming from range and 5.6 coming from his arm (the other 0.4 came from error runs above average).

His range led the league at the position, while J.D. Martinez edge him by 3.2 runs with his arm (Martinez, though, was below average in terms of range).

Chisenhall was second in terms of UZR (9.3), coming in second in range runs by 0.3 runs, with much of the remaining gap coming from the difference in arms (Chisenhall produced 1.6 runs above average on his throws).

There was a smaller gap in terms of arm value in DRS, while a 10-run disparity in plus/minus puts Chisenhall over the top here (Chisenhall led the way with 11 DRS, followed by Rusney Castillo’s 8, and Calhoun’s 6).