A reasoned analysis of Kyle Schwarber’s left field play

For all the members of Cubs Nation that have already wiped pre-2016 Cubs history from their memories, I will remind you about some very recent history: in 2015, the New York Mets unceremoniously swept the upstart Cubs 4-0, ending their season. At this point, I doubt that many Cubs players dwell on that sweep. One Cubs player who may wake up in a cold sweat thinking about that series, however, is Kyle Schwarber.

At least two, and maybe all three, of these plays probably take center stage in Schwarber’s nightmares:

The funny thing is, those might be the three worst plays Schwarber made in left field over the entire 2015 season. Yet the narrative that emerged from his 2015 season is that Schwarber is a defensive liability in left field.

Is that reputation deserved? I’m not so sure based on three sources of defensive data: fielding runs above average (FRAA), defensive runs above average (DEF), and catch probability. Each of these metrics tries to provide an objective measure of defensive performance and, when it comes to Schwarber, they do not agree. Let’s look at the first two together, and catch probability separately.


Both FRAA and DEF start with the number and type of individual plays made, and adjust for various factors such as batter handedness, expected plays per position, and ballpark, among others. In both statistics, positive numbers are good (i.e. the player reduced the expected number of runs scored by this amount), and negative numbers are bad.

In 2015, FRAA gave Schwarber a Blutarsky in left field. Zero point zero. Unlike a Blutarsky quarterback rating, that means Schwarber was perfectly average as compared to other MLB left fielders. DEF, on the other hand, gave Schwarber a negative rating of -3.4, placing him in the bottom 25% of all left fielders who played at least 250 innings (Schwarber played 295.2 innings in left field in 2015).

Still, -3.4 DEF can be overlooked for a guy who put up a TAv of .307 and an ISO of .241 in his first extended time playing MLB outfield. In fact, some other offensively gifted left fielders who performed far worse by these defensive metrics rarely get the “defensive liability” moniker:

Player 2015 Def 2015 TAv 2015 ISO
Kyle Schwarber -3.4 .307 .241
Stephen Piscotty -6.0 .314 .189
Christian Yelich -7.3 .288 .116
Michael Brantley -9.1 .294 .170
Ben Zobrist -11.5 .295 .173

I included Ben Zobrist because obervers often float him as a Cubs’ regulars who could replace Schwarber in left field for defensive purposes. While Zobrist is a fine offensive player, he played left field for over 300 innings for the A’s and Royals in 2015 and racked up a bad DEF of -11.5. Moreover, in 2016, Zobrist wracked up a sub-par -5.5 DEF in left field for the Cubs. Yet no one suggests that he is a defensive liability in the outfield. Based on FRAA and DEF, Schwarber may be average to below-average, but he is still better than our “super-utility” player Zobrist. It’s hard to argue that Schwarber deserves the “defensive liability” moniker based on these measures.

Catch Probability

Recently introduced at Baseball Savant, “Catch Probability” provides new insight into outfield defensive performance with Statcast data. Using launch angle, exit velocity, and batted ball direction, Statcast calculates where every batted ball will land and how long the ball is in the air. Statcast also tracks the location of all fielders, including the closest outfielder. Combining these data points along with the outcome of thousands of fly balls over the 2015 and 2016 seasons, catch probability delivers exactly what it suggests: the frequency with which similar batted ball events are caught by outfielders. To provide context, catch probability divides fly balls into five different categories based on likelihood of being caught, based on (a) the distance the outfielder had to cover to catch the ball and (b) the amount of time the outfield had to get there:

  • 5 star plays: 0 – 25%
  • 4 star plays: 26 – 50%
  • 3 star plays: 51 – 75%
  • 2 star plays: 76 – 90%
  • 1 star plays: 91 – 95%

In large part, catch probability reflects similar leaders as FRAA and DEF. Adam Eaton, Kevin Kiermaier, Billy Hamilton, and Jason Heyward are all near the top of the 4- and 5-star plays, as well as converting at or near 100% of the 2- and 1-star plays.

How does Schwarber stack up? Here is a look at the number of opportunities Schwarber had in 2015 to make a play at each level of difficulty, the percent of the plays he made, and the average percent of the plays made by left fielders in 2015:

***** **** *** ** *
Opportunities 9 4 4 1 10
Catch % 0 50% 25% 100% 100%
LF Average 8.2% 42.4% 67.5% 84.3% 93.3%

I see two pieces of good news here. First, Schwarber is perfect on 1- and 2-star fly balls, so he made the easy plays. Second, Scharber made two of four 4-star plays, showing that he has the physical tools and instincts to make some of the tougher plays. Looking back through MLB highlights, my guess is those two plays are among these three videos:

The bad news: He made only 25% of 3-star plays, well below average for left fielders. He only had four opportunities, so with more chances that will normalize, especially given his ability to make several more-difficult plays as noted above. I also do not put much weight on his failure to make any of the most difficult plays. Those plays are relatively rare and Schwarber does not need to play gold glove-caliber defense—he just needs to hold his own.

Looking back at Schwarber’s 2015 NLCS blunders, those batted balls share a common thread: they are driven directly at Schwarber. Outfielders frequently cite balls hit directly at them as the most difficult to judge, even Willie Mays, and an Ivy League cognitive sciences professor confirmed that observation in a 2010 study. Thus, none Schwarber’s gaffes was a particularly easy play. They remain, however, melding recency and outcome bias, to create the narrative that Schwarber is a defensive liability in left field.

By contrast, based on FRAA, DEF, and catch probability, it’s more likely that Schwarber plays an average left field, and he may improve with more experience. With all of the Cubs other defensive standouts, and given Schwarber’s offensive potential, I’m sure the Cubs front office and Cub fans are more than happy to let him muddle around left field for the foreseeable future.

Lead photo courtesy Dennis Wierzbicki—USA Today Sports

Related Articles

4 comments on “A reasoned analysis of Kyle Schwarber’s left field play”

Very encouraging. I don’t think any astute observer of Schwarbs would believe he wouldn’t work on improving in LF. I’m more than content if he is just average and rakes offensively.

Stunning about Zobrist though…


Those sample sizes make me itchy.


Finally, some brains among small sample idiocy. This is the best and most honest piece I’ve seen on Schwarber’s LF play.

For those not believing/understanding advanced metrics, we could always point to the fact that Schwarbs made all of 1 E in the ’15 playoffs. Better yet, that matched his total E all regular season out there.

But in a nod to even greater truth, the advanced metrics show where he placed among his peers. In ’15, 59 LF played at least 250 innings. Here’s where Schwarber ranked among them:

DRS: 39th (despite playing about 1/3 a season)
UZR: 34th (same qualifier)
DEF: 28th (better than avg)

Shocked about Zobrist? We’ve only just begun…Here’s a list of whom Schwarbs beat in DEF (per Fangraphs):

Piscotty 30th
C Crawford 34th
Brantley 35th (MLB Network chose him as the #1 LF IN ALL MLB going into ’16)
Holliday 37th
Uptown 39th
Gardner 41st
Zobrist 43rd
Parra 44th (former GGer)
Revere 50th
Byrd 51st
Werth 55th
MCabrera 58th
HRamirez 59th

When we remember that Schwarber was 22 and was playing a foreign posn, his performance wasn’t just average–it was ASTOUNDING.

Look for an even better defensive season out of him this year–then laugh at the know-nothings.

Thanks for getting the truth out there.


One slight correction: it was UZR/150, not the cumulative UZR, where Schwarbs finished 34th outta 69 LFers.

Leave a comment

Use your Baseball Prospectus username