clock menu more-arrow no yes mobile

Filed under:

Forget the Gold Glove — Wong’s Web is Platinum

Kolten Wong is putting together a historic season on the field.

Cincinnati Reds v St Louis Cardinals Photo by Dilip Vishwanat/Getty Images

Prior to 2018, Kolten Wong had always shown potential with the glove. He was capable of pulling out some really excellent plays but seemed to struggle with consistently converting the routine.

This season has been another story.

Wong is having one of those years where the eye test confirms the stats, and vice versa. His glove has been smooth and crisp, his arm sharp and precise. Even when he’s scrambling to make a play with what looks like an errant throw — like Tuesday night’s web gem — everything seems to turn to gold.

And so should his glove, when awards season rolls around.

The Gold Glove now officially includes statistics as a portion of the decision (though maybe not as much as it should, but that’s a story for another day). In 2013, SABR launched a partnership with Rawlings which made a new metric — the SABR Defensive Index (SDI) — account for 25% of the Gold Glove decision, alongside the usual votes from sportswriters, players and coaches. What is SDI? Ultimately, it’s a conglomeration of all the prominent defensive metrics in the game, both today and when the metrics were just gaining traction.

SDI is composed of five systems used to judge a player’s fielding prowess, all developed independently, and divided into two categories: batted-ball and play-by-play. The batted-ball category is the newer of the two and comprises the defensive metrics most commonly thrown around by both professional sportswriters and your local bloggers (hello) alike. Defensive Runs Saved (DRS), from Baseball Info Solutions data, and sabermetrician Mitchel Lichtman’s Ultimate Zone Rating (UZR) are the most common, with SABR Defensive Committee member Chris Dial’s Runs Effectively Defended also featured. The other category, play-by-play, is pretty much how it sounds. The systems are built on paper data — what you’d expect to find on an official scorecard. That includes who fielded the outs, the categories of hits, the location where hits dropped on the field, etc. Those two are Defensive Regression Analysis, created by committee member Michael Humphreys, and Total Zone Rating. Consider this article a rudimentary defensive primer of sorts, because it’s easy to forget — or never learn, in many cases — how theses stats are calculated and what represents a “good” value.

We’ll skip the exclusive second basemen comparison because a look at the full National League picture highlights it pretty well. After taking a look at the numbers, it’s not only clear that Kolten Wong deserves the 2018 NL Gold Glove at second base, but the Platinum Glove awarded to the league’s best fielder as well. By all of the most commonly accepted defensive metrics at our disposal, no NL fielder has done a better job this season than Wong. In fact, as we’ll look at later, his 2018 campaign is already ranking as one of the National League’s best this millennium.

Batted-Ball

DRS

Defensive Runs Saved is similar to most of the prominent offensive statistics used today in that it’s measured in runs above or below average, as the name would describe. Fielding Bible has a detailed explanation of the calculation if you want to get into the weeds, but ultimately it takes a combination of stolen base prevention (pitchers/catchers), bunt runs saved (corner infielders), double play opportunities (middle infielders), throwing arm ability and home runs prevented (outfielders), and overall range and ability to convert a batted ball into an out. DRS is position-specific, meaning that it doesn’t account for the difference in difficulty of shortstop compared to first base, but it essentially allows us to say someone is so many runs better or worse than the average fielder at their position. FanGraphs rates any fielder with 15 runs or more as Gold Glove caliber. Here are the NL leaders through the end of play Tuesday:

DRS - 2018

Name Position Innings DRS
Name Position Innings DRS
Kolten Wong 2B 746.1 18
Lorenzo Cain CF 909 17
Nick Ahmed SS 959.1 17
Adam Duvall LF 725.1 16
Addison Russell SS 899.2 14

The initial reaction may be that those numbers are pretty close. Look a little closer. Wong has at 200 fewer innings (at the least) than all but one of his counterparts on that list. This isn’t a metric averaged to accommodate for playtime fluctuations, as we’ll see in another system further down the line. Wong has saved 18 runs above average in his 746.1 innings, which beats the next-highest total of 17 accumulated in 909 and 951.1 innings by the second and third place fielders.

UZR

UZR is similar to DRS in that it uses “zones” for rating defensive success, but they differ in a few areas. According to the Fielding Bible, “Defensive Runs Saved uses a rolling one-year basis for the Plus/Minus system, while UZR uses several years of data to determine each play’s difficulty level.” FanGraphs offers a high-level look at the calculation of UZR:

At it’s most basic level, it’s a measure of the average amount of damage that batted ball would do and how often it is converted into an out, relative to average at the position. So if the average left fielder makes a player 40% of the time on the ball in question and that batted ball (based on location, speed, etc) is worth 0.8 runs on average, fielding it cleanly earns you 0.48 runs toward your UZR (0.8*0.6).

These differences, along with the fact that UZR is park-adjusted, are probably why it’s the most commonly used ranking for defensive ability, eventually being factored into the defensive component of fWAR, which IS positionally adjusted. The only caveat is that UZR doesn’t offer analysis of pitchers and catchers.

UZR - 2018

Name Position Innings UZR
Name Position Innings UZR
Kolten Wong 2B 746.1 11.3
Lorenzo Cain CF 909 9.7
Kyle Schwarber LF 792.2 6.3
Carlos Gonzalez RF 772 6.2
Freddie Freeman 1B 1114.2 6.1

A different cast of characters here, and that’s not an anomaly. Perhaps this will begin to shift as defensive metrics begin to share components and Statcast data makes batted-ball measurements more precise, but the leader in DRS hasn’t consistently shared the same honor in UZR over the years due to the differences in calculation. The disparity has existed four times in the 10 seasons prior to 2018, with the two prior seasons included in those four. Wong, however, lays claim to the top spot in both. There’s Cain again, who really has been a tremendous center fielder for Milwaukee. He still can’t compete with Wong, nearly two full runs below him in the category.

UZR/150

UZR/150 is the same metric as above, but scaled to 150 games. It isn’t explicitly used in SDI calculations but it’s a variation of traditional UZR that adjusts for playing time. Here’s that table:

UZR/150 - 2018

Name Position UZR/150
Name Position UZR/150
Kolten Wong 2B 20.5
Lorenzo Cain CF 18.3
Kyle Schwarber LF 12.5
Corey Dickerson LF 11.8
Caros Gonzalez RF 11.7

Oh, yeah. Wong has had fewer chances, remember? His 20.5 runs above average, if it maintains, would be the highest posted by any qualified NL player since 2013, when it would’ve ranked second to Gerardo Parra. It would’ve been first in 2011 and 2012, too.

We’re through two of the three batted-ball metrics and that’s where we have to stop. Runs Effectively Defended is a system which doesn’t feature data publicly and is ultimately pretty hard to find information about online. It’s based on STATS Zone Rating, which was initially housed by Sports Illustrated.

Play-by-Play

We run into the same issue with Defensive Regression Analysis (DRA); the numbers aren’t publicly available and, quite frankly, it’s almost impossible to find anything on the web about it. What we can look at, though, is Total Zone Rating.

Total Zone Rating

Baseball Reference houses the Total Zone Rating data, developed by Sean Smith, where they also feature his description of the metric’s function. The benefit of a play-by-play metric is apparent when viewed through a historic lens — we don’t have batted-ball data on Willie Mays’s or Ozzie Smith’s great plays. It offers a baseline for valuing players who have since gone by in ages without the technology we have today. The issue, to me, lies in the calculation. Here’s an excerpt from Smith’s explanation:

“The responsibility split between infielders was originally 50/50, but has been refined based on more detailed analysis. Singles to left are charged 60/40 to third and short, to center it is 52/48 between short and second, and to right it is 55/45 first base/second base.”

Sharing responsibility of singles to right field with José Martínez based on play-by-play data isn’t too fair to Wong. The results have Nick Ahmed leading the table with 19 runs above average and Wong falls in all the way down at 15th with 10. There are some peculiarities in the list, however. Marcell Ozuna ranks 12th, three spots higher than Wong. Looking at batted-ball data, Ozuna should by no means be in the top 15 defenders in the NL. What makes it even more interesting is that Smith names outfield arms as a large component of the metric, which we’ve all been able to tell has been a crux of Ozuna’s defense in 2018.

Nonetheless, play-by-play data is included in the SDI metric, and it doesn’t seem to be hurting Wong’s chances at all.

SDI

The SDI numbers aren’t on a continually-updated database for public consumption like other metrics. Given that it takes five major defensive evaluation systems and consolidates them into one concise number, it takes a little time. The first-half numbers were released through games on July 15, however, and they highlight what we’ve already seen:

SDI - 2018

Name Position SDI
Name Position SDI
Kolten Wong 2B 10.7
Adam Duvall LF 8.1
Lorenzo Cain CF 7.6
Corey Dickerson LF 7.5
Brandon Belt 1B 6.6

These numbers ran through all of Matheny’s tenure, where his playing time wasn’t always ideal. He has a pretty considerable deficit in terms of innings logged compared to his counterparts. Yet here he is, leading the (currently) definitive metric in deciding the statistical portion of the Gold Glove process, making up 25% of the final decision. He was second only to the Oakland Athletics’ phenom Matt Chapman in all of Major League Baseball in SDI.

It isn’t included in SDI, but let’s throw in that Wong is leading the NL in the defensive component of fWAR as well. By a positionally-adjusted UZR, he leads the league.

Wong’s season hasn’t just been good in the context of 2018, though. He’s been putting up one of the best defensive seasons for a National League player since the turn of the century.

Historic Implications

UZR has only existed in some capacity since 2001. DRS is even younger, starting in 2002. Looking at the batted-ball metrics in that time, here’s where Wong’s 2018 — if the season ended today, with his 746.1 innings — would rank on average in each year’s voting.

DRS - 5th

UZR - 7th

UZR/150 - 2nd

Def - 10th

Keep in mind, those numbers are across all positions, for the entire season. He’s lumped in with late-2000s Brandon Phillips and Chase Utley, mid-2010s Andrelton Simmons, and 2015 Kevin Kiermaier. Moreover, his numbers are compared against their full-season stats, ranging from 1000 to 1400 innings. In some cases, Wong has half the playing time as his counterparts and still finishes in the top 10 in every category. Since UZR/150 adjusts for playing time, you can really see how phenomenal he’s been.

There are still 34 games on the Cardinals’ schedule in 2018. Wong still has plenty of time to play and those extra chances could throw off his pace, but it could also make his numbers even that more impressive, and he could continue to climb that 21st century leaderboard. What we know to this point is that Wong has looked incredible to the naked eye, that he leads in all batted-ball categories of defensive evaluation, and that he holds the highest spot in the SDI rankings officially used to assign the Gold and Platinum Gloves. That’s only 25% of the system, though. The remaining 75% of the vote will call upon sportswriters, players, and coaches to rank the fielders in contention, easily giving the human side of the evaluation the most weight. Let’s hope the majority are up on analytics because, to this point, Wong’s glove has been nothing short of platinum.