Baseball

Fixing Baseball’s Gold Gloves

|
Image for Fixing Baseball’s Gold Gloves

Welcome to a 119-year trek through Major League Baseball history. Over time, all MLB awards will be fixed, but the first order of business is the Gold Glove.

Gold Gloves have been given out since 1957. In that year, awards were given to one player at each position in baseball. This practice was abandoned after just one cycle, and subsequent Gold Gloves have been distributed to the best American League and National League fielders at a particular position.

From 1961 to 2010, outfielders were lumped into one category rather than three distinct positions. Before 1961 and since 2011, players at each defensive outfield spot have been judged against their peers.

Notably, there have been four ties in voting. In 1985, Dave Winfield and Gary Pettis tied for the final American League outfield slot. In 2007, Aaron Rowand and Jeff Francoeur tied for the final National League outfield slot. 2012 saw Jeremy Hellickson and Jake Peavy split the American League pitcher spot. Finally, in 2018, Anthony Rizzo and Freddie Freeman tied for the National League first baseman spot.

These three points are notable because they lay the groundwork for this 119-year project. Years before 1957 do not need fixing; they need a Gold Glove winner in the first place. Adjustments are needed for 1957 through 2021. For each season, 1903 to 2021, there will be 18 Gold Glovers. Each outfield spot will be judged differently, and there will be no ties. Here is a breakdown of the stats used for each era of baseball.

Stay tuned for a follow-up disclaimer piece outlining potential shortcomings with various stats.

1903-1952: The Wild West

For hitters and pitchers, there are surprisingly detailed accounts of just about every aspect of the game. There are a handful of missing runs batted in from the dead-ball era, and caught stealing was not an official stat for base-runners for a while, but the backbone of the modern statistical system is there. One can calculate on-base-plus-slugging, walk rate, or even advanced stats like OPS+, wRC+, and others. Pitchers, similarly, can be judged through a modern lens. FIP is available as is ERA+.

Defense, on the other hand, has little to work with. Even concrete defensive positioning can be difficult to nail down. The three most basic defensive stats are putouts, assists, and errors. Pitchers and catchers also have stolen bases allowed and caught stealing. Double plays turned is also available, but it generally only matters for middle infielders.

Putouts

For the most part, players receive a putout for doing one of four things. If a player catches a flyball, they record a putout. Similarly, if they are standing on a base and get a force-out, they will get a putout. Even tagging a runner and recording an out will turn into a putout. Catchers get credit for completed strikeouts as well.

Think of these as basketball points. These are outs recorded by a defensive player. Over a large enough sample size, the best fielders will rack up many putouts.

Assists

Assists are accumulated by facilitating in the out (or attempted out) of the other team. For example, if a batter hits a ground ball to the second baseman, the second baseman will get an assist for throwing the ball to first base to record the out. This also applies to relay throws and double plays. If a player touches the ball before an out (or attempted out) is recorded, they will receive one assist.

Errors

Not all putouts and assists go as planned. Occasionally, players will drop the baseball, miss the ball under their glove, or pull a Fernando Tatis Jr. and throw the ball to no one in particular. In these cases, the official scorekeeper may rule the play as an error

Fielding Percentage

Fielding percentage is the conversion rate of fielders. The numerator for the fraction is total putouts plus total assists. The denominator is putouts, assists, and errors put together. For example, In 2021, Tatis had 52 total chances in his time as an outfielder. He made one error. This leads to a fraction of 51 successes over 52 chances: a .981 fielding percentage.

Verdict: Rfield

These stats reveal little context about the fielding situations. Errors are subjective, and bad positioning is not punished through errors. Instead of using these rudimentary stats, the go-to statistic will be Baseball-Reference’s fielding runs. Rfield is based around zero with positive figures earned for above-average defense and negative figures for below-average defense. This stat is available for every season on Baseball-Reference.

For reference, in pre-1953 MLB, the worst season was Billy Urbanski’s -26 in 1935. The best season was Frankie Frisch’s +37 in 1927.

The goal of Rfield is to quantify defense in terms of runs compared to the average. For the 1903 to 1952 selections, Rfield will be the primary defensive measuring stick for the eight position players. Pitchers will be judged with range factor. While cross-era comparisons with range factor are generally useless because of increasing strikeout numbers, in-season comparisons have a solid sample size.

1953-2002: Total Zone

Total Zone

Rfield quantifies how well a fielder fields compared to the average. Total Zone (TZ) accomplishes the same feat by using batted-ball data to determine zones on the field that each player is responsible for. While this batted-ball data is not quite Statcast data, it gives a general idea of who should receive credit/blame for batted balls. For example, a long fly ball to centerfield is the centerfielder’s responsibility.

Like Rfield, TZ is represented on a scale with zero at the center. Negative numbers represent below-average fielding. Positive numbers represent above-average fielding.

In this timespan, the best two seasons by TZ are Barry Bonds’ +37 in 1989 and Andruw Jones’ +36 in 1999. In most seasons, the league leader in TZ will be somewhere between +20 and +30.

On the other end of the spectrum, Dante Bichette had a legendarily poor -34 TZ in 1999. Chris Gomez had a -33 in 1997.

Range Factor

Range factor was created in this era by Bill James, but it applies from 1903 to 1952 because it only concerns putouts, assists, and time played. It boils down to putouts plus assists over the time aspect (usually games or nine-inning chunks). Errors are not included.

Range factor is useful (for some positions) especially when it is compared to the average player at a position. For example, Brooks Robinson had a range factor per nine innings of 3.20 at third base for his career. The MLB average was 3.09. Multiplying that rate out over Robinson’s 25,000 innings at third base means Robinson made approximately 300 more plays than the average third baseman.

Gold Gloves

From 1957 to 2002, Gold Gloves were distributed. While many were deserving, the criteria to win a Gold Glove was never fully fleshed out. There is a bevy of poor selections, but Rafael Palmeiro’s third Gold Glove stands out as one of the worst. Palmeiro was a good defensive player (at first base), but in 1999, he won the Gold Glove after playing just 246 innings at first base over 28 games

Like errors, Gold Gloves are human-centric awards. Selection is subjective. The other awards generally are decided by pure numbers. Defensive awards, particularly in this era, were not objective. For every easy selection of Ozzie Smith or Roberto Clemente, there was a head-scratcher like Palmeiro.

Verdict: Total Zone … Mostly

TZ is useful for all eight fielders. It will be the baseline for selections in this era. Range factor per nine innings will help with pitchers, and it may be used in the event of a tie. 

2003-2021: Fielding Bible

Defensive Runs Saved

Over the last 19 seasons, the Fielding Bible has tracked defensive metrics. In its current iteration, it tracks eight stats to pump out one figure: total runs saved (also known as defensive runs saved: DRS). There is positional variability such as the battery partner having a “stolen bases” category. A major segment of the calculation is “PART” which includes positioning, balls hit in the air, range, and throwing.

In Fielding Bible’s existence, the leaders in DRS are Adrian Beltre (+200) and Andrelton Simmons (+197). The best season by defensive runs saved is Simmons’ +41 in 2017. The worst season is Adam Dunn’s -43 in 2009.

Pitchers can rejoice as they are finally welcomed into the statistical party. Including the aforementioned PART and stolen bases categories, pitchers are also measured on bunts and “GFP/DME.” GFP/DME stands for good fielding plays and defensive misplays and errors. Since 2003, the top two defensive pitchers are Mark Buehrle (+87) and Zack Greinke (+84). Dallas Keuchel ranks fifth, and two-time Gold Glover Max Fried is second to Keuchel over the last two seasons.

Ultimate Zone Rating

Ultimate Zone Rating, UZR, is Fangraphs’ answer to the defensive question. UZR, like TZ and DRS, is on a scale with zero as average. UZR was first used in 2002, and it is more conservative than DRS. For example, Simmons’ ridiculous 41 DRS in 2017 was given a more modest +19.7 UZR. Across MLB in 2017, 29 players earned 10 DRS. Just seven players earned 10 UZR.

Verdict: DRS Gets the Nod

Both stats are valuable, but DRS will be the primary determiner in Gold Gloves from 2003 to 2021. As mentioned, 1903 through 1952 will lean on Rfield. 1953 through 2002 will lean on TZ. In a modern context, all three can be referred to as Rfield, but they will be clarified when necessary.

Baseball-Reference measures all three stats as equals. For example, Willie Mays earned +8 Rfield in 1951. There is no TZ (or DRS) equivalent. In 1954, he earned +21 Rfield, equivalent to his NL-leading 21 TZ.

In 1999, Omar Vizquel racked up +14 Rfield, the same as his TZ mark. In 2007, his Rfield of +16 was based on his DRS rather than his +20 TZ.

Is it perfect? No, but it gives a strong baseline. Also, players will be compared to each other within the same season, so using the same stat will keep a level playing field. The goal is to compare 1991 Cal Ripken Jr. to 1991 Alan Trammell, not 1991 Ripken to 2017 Simmons.

Stay tuned for the disclaimer featured in part two.


Follow me on Twitter at @MrSplashMan19 for more of my content! Don’t forget to join our OT Heroics MLB Facebook group, and feel free to join our new Instagram –  @overtimeheroics_MLB, and listen to our baseball podcast, Cheap Seat Chatter! We’ll see ya there!

Come join the discussion made by the fans at the Overtime Heroics forums! A place for all sports!

Main image credit Embed from Getty Images

Share this article

Ryan Potts is an avid football and baseball fan. He covers the NFL and Major League Baseball, focusing on the Baltimore Ravens and Atlanta Braves.