On Michael Brantley’s surprisingly bad defensive metrics

Michael Brantley

 

“Shortstops make most plays. One remembers good plays made longer than one remembers plays not made.”

Let’s dub this a “Derek Jeter.” By now, better defensive stats have become somewhat mainstream in baseball analysis. In general, we should now know that errors and assists aren’t the only statistics available. On the highest and most nerdtastic level, there’s Ultimate Zone Rating (UZR) and Defensive Runs Saved (DRS), two of the increasingly popular and far more complicated defensive metrics. Each takes into account a whole ton of factors, including range and standardization to a baseline player at that position.

That above quote was from a recent article by FanGraphs’ Jeff Sullivan. The article was about Adeiny Hechavarria of the Florida Marlins, a moderately obscure flashy young shortstop who is not viewed too favorably by said advanced metrics. Similarly, Jeter, ye’ of five career Gold Gloves, was perennially one of baseball’s worst defensive shortstops per these statistics, but he certainly wowed with the occasionally highlight (or six).

While reading Sullivan’s article, I couldn’t help but think of a certain Cleveland Indian: Michael Brantley. Yes, you know Brantley. The one who is having an MVP-like offensive season, as Sullivan himself recently wrote at FOX Sports. But on defense, despite his impressive 11 outfield assists, it appears he’s actually not as elite as we might think. Consider:

Updating through yesterday’s games, Brantley now has a -6.9 UZR in 1,058.1 combined outfield innings. That’s impressive improvement in just one week … but it’s still not quite good. The stats aren’t pretty.

Sporting News’ Ryan Fagan recently awarded him an Anti-Gold Glove for left field, where his UZR is actually a horrendous -12.5 in 685.1 innings (counter-intuitively compared to 1.3 in 373 innings in center). On the DRS side of things, the 27-year-old isn’t performing too poorly. He’s at -5 for this season, down from the mostly average production of +2 in 2013 and -1 in 2012.

Either way, the eye test doesn’t seem to match the math. These defensive statistics are not necessarily perfect, of course. Here’s a fantastic example, via FanGraphs’ Dave Cameron:

Small sample size? Well, not anymore. We’re now on year four of [Jhonny] Peralta being rated as an excellent defensive shortstop by UZR. Since the start of the 2010 season, he’s played over 4,600 innings at the position, and he has a UZR/150 of +10 runs per season over that stretch.

Yes, that’s the same Jhonny Peralta once booted from Cleveland’s shortstop position for Asdrubal Cabrera, who subsequently ranked as one of baseball’s worst defenders. It’s funny, isn’t it? Peralta was excellent in Detroit at short. He also got booted to left field upon his suspension and the acquisition of Jose Iglesias. Then, the Tigers didn’t keep him in free agency. Peralta’s now in St. Louis and having a career defensive season. Go figure.

How can this all be? Let’s go to another example, this time in the form of Kansas City’s Alex Gordon, a surprising contender for best WAR in baseball. That’s mostly due to a one-win uptick in his defensive value. Via Cameron again at FanGraphs:

We’re really talking about evaluating a player based on his performance on something like 50 marginal plays throughout the course of the season. It would be ludicrous to expect performance over an N of 50 to be the same every single trial, especially when the result of the play made or not made has such a large swing in run value.

How has Brantley performed when breaking down his Inside Edge data? He’s grabbed 100% of seven likely (60-90%) opportunities. He missed on his only even (40-60%) opportunity. He is at a good 50% on four unlikely (10-40%) opportunities. And he missed on all seven remote (1-10%) chances. That’s a total of 19 plays making the difference of his season. And he hasn’t done too bad.

The biggest knock on Brantley’s defensive resume, however, is his range. It’s at -5.9 this season, thus costing the Indians nearly half a run himself on just his inability to get to certain balls a typical outfielder might. So it’d perhaps make sense he’d only even have 19 marginal plays; he’s just not getting to the usual 50. Last year, this actually was even worse at -7.0. But he had an even better season in 2013 with his arm and at avoiding errors.

So when we talk about the Indians defense, we have to consider that Brantley is regarded as solidly below average by the advanced metrics. And yes, this Indians defense is near historically bad. From two weeks ago by Sullivan at FanGraphs:

The Indians are on track to have the second-worst UZR since 2002, and the third-worst DRS since 2003. Where is this coming from? According to both metrics, the Indians have had a lousy defensive infield. According to both metrics, the Indians have also had a lousy defensive outfield. So, again, you’re getting an idea of their consistency.

Alas, Brantley is a Jeter is a Hechavarria, per the metrics. They al have the ability to be great defenders, but while showcasing for the highlight reel, they too often miss the routine plays. And it’s one of the many issues contributing to a terrible, terrible defense holding back a frustratingly .500 Indians squad.

  • The_Real_Shamrock

    Trade him!

  • mgbode

    if you are going to be a below average defender, then you need to be a very good hitter. Brantley certainly qualifies though, yes, it’d be nice to have him get better.

  • https://twitter.com/jimkanicki jimkanicki

    on the other hand brantley is third among MLB LFs in assists this year. not a word about that in this ‘analysis.’

    analytics strikes again.

  • Achilles

    I have to say, that I think these measurements are bunk. As Rosen says, it doesn’t pass the eye test. I mean, are there a lot of balls that are hit in front of, to either side of, or behind Brantley, where I am saying to myself: “an average fielder would have got to that.” Sorry, but no.

    I get that the stat geeks are devout about their ability to change baseball into a algorithmic math problem, but it’s more chicanery than science in a lot of instances.

    That said, the Indians have been horrible on defense. Terrible. I don’t need some computer nerd to invent mathematical ways to explain it, I’ve been watching it unfold all season. But Brantley in left field is the least of our problems.

  • mgbode

    But on defense, despite his impressive 11 outfield assists, it appears he’s actually not as elite as we might think.

    Jacob mentioned the assists. It’s why he brought up Jeter. A few impressive plays possibly masking an overall deficient defense.

  • Achilles

    The problem in general with analytics is that it undervalues the more creative and intuitive parts of baseball, the grey areas where a lot of games are won.

    Brantley is one of the best I’ve ever seen at adjudging where a ball is going to bounce, how it will carom off the wall and how to position himself to both field it cleanly, pivot and use a strong arm to transform a would-be double into an out.

    Where is my statistical analysis of this? There is none, which is why an overreliance of so-called “advanced metrics” is no replacement for watching a guy play. They might be interesting and help supplement analysis, but they are not the end-all, be-all.

  • https://twitter.com/jimkanicki jimkanicki

    lookit. if you want to see how poor corner outfielders hold a team back all you had to was watch the whitesox game last night.

    if top 3 in outfield assists qualifies as a defensive liability, perhaps it’s the analytics and the anaylst that needs review, not brantley’s defense.

  • mgbode

    defensive analytics definitely need more review. I dislike them more than I trust them.

    however, how many of those assists are from him not catching balls that other LFers would have caught in the first place? not an easy question to answer, but an important one to level-set on.

  • The_Real_Shamrock

    Exactly stats can be used/interpreted in many different ways.

  • https://twitter.com/jimkanicki jimkanicki

    all the likes. and you havent even touched on how rare outfield assists are or also that they are game changers — either gunning down a runner in scoring position or preventing a run.

  • Steve

    There actually is. Fangraphs measures this.

    ARM (outfield arm runs): Outfielder’s get credit (plus or minus) depending on what the runners do on a hit or a fly ball out. A runner can stay put, advance, or get thrown out. A fielder will get credit not only if he throws out more than his share of runners, but also if he keeps more than his share of runners from advancing extra bases.

  • Steve

    Are you able to see where Brantley is positioned before the play, the angle and speed of the ball off the bat, and how much he had to move to make the play? No? Then you don’t have anywhere near as much information as those you needlessly degrade.

  • Steve

    “perhaps it’s the analytics and the anaylst that needs review”

    Of course, the “analyst” who can’t do anything but say “count da outfield assists!” need not review his work.

  • Achilles

    Sounds good in theory, but I think the data collection on this is likely to be significantly flawed, as runners will do different things situationally (depending on innings and outs and score), having nothing to do with the individual fielding and/or throwing prowess of outfielders.

    Furthermore, the statisticians would have to account for the speed of the runner, his decision making ability (whether he tried for second because it was there and the outfielder made a great play or whether with was a plain stupid decision), and countless other variables that are too subject to manipulation on the information gathering side of the equation.

    There is a place for statistics in baseball, but they have severe limitations in terms of providing a full picture, particularly in a game that is so contextually bound to different situations.

  • Steve

    You think, but you don’t know, and certainly can’t be bothered to figure out. You’ll have to pardon that I’m unmoved by your argument. If you want to critique the system, first understand and find its flaws. So far, you’ve only demonstrated that you weren’t even aware of what they are attempting to capture. I have no idea why you even think that puts you in a valid place to criticize.

    As far as the “but they have severe limitations in terms of providing a full picture” argument – no one understands better than the guys collecting the stats how limited they actually are. That’s why they continue to study instead of doing what you’re doing – complaining.

  • Scott

    I believe this same logic applied to Choo when he was here. His cannon from right gunned a handful of runners down. Made one forget about all the times he wildly threw the ball into row B.

  • http://www.wahoosonfirst.com Ed Carroll

    Choo was regarded as a decent right fielder and terrible center fielder.

    And, this isn’t quite logic here, it’s what the statistics say. Though it is certainly logical to use statistics as proof of an argument.

  • http://www.wahoosonfirst.com Ed Carroll

    Choo was regarded as a decent right fielder and terrible center fielder.

    And, this isn’t quite logic here, it’s what the statistics say. Though it is certainly logical to use statistics as proof of an argument.