Alex Gordon is working his way into the American League MVP race, thanks in part to the success of his Kansas City Royals, who have used a 25–8 stretch since July 22nd to move into first place in the American League Central. On Tuesday night, it was Gordon’s 2-run homer in the bottom of the ninth that pushed the Royals past Minnesota and kept them 1.5 games ahead of Detroit.
For the season, Gordon is hitting .283/.356/.454 with 16 home runs, and his 122 OPS+ is the only above-average mark in the Royals’ regular starting lineup, according to Baseball-Reference. He is the leader and face of what may be the first Royals team to make the playoffs since the colorful days of George Brett, a cool 29 years ago.
But if Gordon has anything resembling an MVP case, it does not rest on his bat, which pales in comparison to stars like Mike Trout (.288/.373/.559, 29 home runs), Jose Abreu (.308/.366/.598, 33 home runs), and Robinson Cano (.326/.394/.465, 11 home runs). Rather, it rests on his ability as a left fielder.
Gordon has been 22.4 runs better than the average left fielder, according to Ultimate Zone Rating — best of any major league fielder relative to position. Defensive Runs Saved, another defensive metric, rates him 22 runs better than the average left fielder — again, the best major league defender relative to position. Both metrics are available at FanGraphs.
The idea that Gordon is a brilliant left fielder is hardly controversial. Gordon moved to the position in 2011 and has proceeded to win three consecutive Gold Gloves. Chances are he’ll win a fourth this year, thanks to plays like this:
And throws like this:
But the idea that his fielding is valuable enough to bridge the vast gap between his bat and the likes of Trout is met with suspicion. Defensive metrics are plagued by data quality issues, and the “garbage in, garbage out” idea applies just as much to baseball statistics as it does to computers. Tim Marchman detailed these issues expertly in Sports Illustrated a few years ago. As such, when Alex Gordon’s year-by-year UZR fluctuates from plus-12 in 2012 to plus-7 in 2013 and jumps all the way to plus-22 in 2014, skepticism takes over. Gordon has been 15 runs above average this year with his bat alone. Has his glove really gotten that much better in a year?
Probably not, but looking simply at skill level ignores some important points of how defense in baseball works. One of the major issues with assigning defensive value is the almost complete lack of control a defensive player has over the chances he receives. A left fielder can see five balls in an inning if his pitcher is getting torn up by the opposition, or he can see nothing in a full game if the opposing team beats the ball into the ground or to right field the whole game. According to Baseball-Reference, Gordon has had eight games this year without a single defensive chance and seen eight games in which he’s had at least five. A fielder’s workload is highly variable game-by-game and season-by-season.
And even more variable can be the quality of those chances. A vast majority of the balls hit in play are either obvious hits or obvious outs — only a small percentage of plays on the margins of positional ranges tend to be decided by actual fielding skill. Baseball Info Solutions has attempted to measure this with a dataset called Inside Edge (available at FanGraphs), which rates plays along the following spectrum:
Routine (turned into an out 90–100 percent of the time)
Likely (60–90 percent)
Even (40–60 percent)
Unlikely (10–40 percent)
Remote (1–10 percent)
Impossible (0 percent)
Gordon’s Inside Edge data shows just how few of the balls hit in an outfielder’s direction are decided by the differences in major league fielding ability:
Gordon is 836-for-841 on “Routine” opportunities (in light blue) since 2012 and 0-for-296 on “Impossible” opportunities (in black). Remove those, and we see the marginal opportunities, the ones that separate the good major league left fielders from the bad.
Despite having played 300 fewer innings so far this season than in either of the past two years, Gordon already has seen 24 more marginal chances than 2012 and 14 more than 2013. Gordon may not have influenced these chances at all, but he has made the most of them, with above-average success rates in all four marginal Inside Edge categories.
In the outfield, the difference between a catch and a misplay is often extra bases. A single chance can easily swing multiple runs in a situation with runners on base — the average double is worth almost exactly one run more than an out to the offensive team. UZR, DRS, and other advanced defensive metrics generally work by measuring the likelihood the average fielder would make the play and the difference in expected runs on a catch and a drop. So if Gordon catches a ball only half the league’s outfielders catch and it would have gone for a double, Gordon’s hypothetical UZR is credited with half a run above average. With so many more of these marginal chances to flash his leather on this year, Gordon has been able to pick a number of extra runs off the outfield grass.
Unfortunately, this is a spectacularly difficult thing to measure, and inputs are often as crude as a stringer in a press box recording approximate plot points. Ideally, with a system like MLB’s new StatCast, the measurement can be more precise. Until then, defensive metrics will be met with an understandable skepticism.
But that doesn’t mean there isn’t something real to the large increase in UZR Gordon is experiencing this season. Gordon is, as he has always been, a stellar left fielder. He may not have willed extra chances his way this season, but he’s made the most of them, which is critical for the Royals.
Enough for an MVP vote, though? That’s personal taste. I would still lean toward Mike Trout’s brilliant offensive season, but Gordon’s defensive value and performance is worth noting despite the haziness of defensive metrics. In an American League lacking other truly great seasons, Gordon’s deserves some time in the limelight.