Four teams asked five players to do things this season that they’d never done prior to 2013. This article is about how well those things have worked for the first six weeks, and whether they can continue.

1. Shin-Soo Choo: start in center field
It’s not that Choo has turned into a superb center fielder. That was never the plan. Starting Choo in center, a position he hadn’t played at all since 2009 and hadn’t played regularly since 2002 (as a 19-year-old in A-ball), was always going to be an exercise in extreme double-entry bookkeeping: Would the runs his bat added outnumber the runs his glove gave up? So far, the answer is an easy “yes.” Choo’s .347 TAv ranks 10th among players with at least 100 plate appearances, and he’s second only to Miguel Cabrera in VORP.

Naturally, there’s been some luck involved, or at least some unsustainable performance. Choo has always been a high-BABIP guy, but he’s running about 40 points above his career average despite (for him) a below-average line-drive rate. And while he’s also always been a guy who gets hit by plenty of pitches, he’s never been plunked at this rate. (Choo has been hit 11 times, nearly twice as often as any other player. His career high is 17, and he doesn’t appear to have changed his stance or approach at the plate in a way that would promote hit by pitches.) Both the BABIP and the HBP have helped drive up Choo’s major-league leading .467 on-base percentage.

The stats tell a somewhat confusing story about Choo’s performance in center. (A lot of defensive stats tell a confusing story after six weeks).* FRAA says he’s been a couple runs below average. That figure prorated over a full season would be an acceptable loss, in light of how well Choo hits. On the other hand(s), DRS has him at -7 and UZR has him at -5.6. Those figures prorated over a full season would seriously eat into his offensive value, once his offensive value regresses somewhat

*According to our batted-ball statistics from MLBAM, Reds pitchers have allowed the highest line-drive rate in baseball. According to BIS’ batted-ball stats at FanGraphs, they’ve allowed the fourth-lowest. The two sources come close to agreeing on the Reds’ roughly average groundball rate, but they can’t agree on whether the non-grounders were flies or line drives.

Over the winter, no one would have scoffed at the suggestion that Choo would be the worst center fielder in baseball, but it’s sort of hard to square the raw stats with what the more pessimistic advanced stats are saying. Choo’s 70 putouts rank 10th among big-league center fielders. He’s caught 8.75 percent of the Reds’ balls in play allowed; last season, Cincinnati’s Drew Stubbs, who’s generally acknowledged to be an excellent center fielder, caught 8.59 percent. With Stubbs in center, Reds pitchers allowed a .379 SLG on balls in play and allowed opposing batters to reach base at a .302 clip. With Choo in center, Reds pitchers have allowed a .365 SLGBIP and a .291 reach-base rate. And the Reds’ defensive efficiency has improved from .712 to .720.

I’m just throwing circumstantial stats at the wall. None of this is to suggest that Choo has been better than Stubbs, or even that Choo has been average. It’s not at all valid to compare the current Reds to last season’s Reds and attribute the difference to Choo. The team has different pitchers, and other different defenders, and has evidently started shifting much more often. But what those stats do suggest is that Choo’s defensive inadequacies, such as they are, haven’t been crippling. He hasn’t routinely tripped over his own feet or let balls roll to the wall. He’s taken indirect routes, and he hasn’t looked like a natural, but the Reds have seen him in center for six weeks, and they’re still starting him there. He’s not so bad that Cincinnati can’t win with him where he is.

There was this,

and this,

and this,

but those plays took place in the first week of the season, and on most plays, you wouldn’t necessarily know that Choo is a novice at the position. Chris Heisey isn’t hitting, and the Reds would be better off with a good defensive center fielder, Choo in a corner, and Heisey on the bench, but it still seems feasible—maybe more feasible than it seemed six weeks ago—that they can make do with what they have. Remember, the Reds had a .254 OBP out of the leadoff slot last season. The .471 they’ve enjoyed thus far this year out of the same slot forgives a few misplayed flies.

2. J.P. Arencibia: catch R.A. Dickey’s knuckleball
Arencibia isn’t great at catching non-knuckleballs, so the prospect of having him catch R.A. Dickey was worrisome enough to make Alex Anthopolous ask for multiple Met catchers in the trade that sent Dickey to Toronto—and then, once he got them, sign Henry Blanco to boot. Determined not to be benched every fifth day, Arencibia began meeting up with Dickey in January to get familiar with the knuckler, then continued to catch him in spring training and in the WBC.

Arencibia’s hard work earned him an Opening Day start, and he caught Dickey for six innings. This was the first pitch of the season:

Uh oh. Many more like it followed:

Of the 104 pitches Dickey threw, nine went to the backstop (as in the wall behind home plate, not Arencibia). Arencibia hasn’t caught a single pitch thrown by Dickey since. The 41-year-old Blanco, who caught Dickey in 2010, has become his personal catcher in Toronto, and while the ancient battery has worked pretty well, yielding only three passed balls—as many as Arencibia was charged with in one start—in the six starts since, Blanco has hit .174/.211/.256 in 90 PA over the past two seasons. Josh Thole, who had 18 passed balls last season, is hitting .360/.442/.547 in Triple-A, so Blanco might not be long for the majors.

This is another reason why there aren’t more knuckleballers in the majors. It can be a pain to make sure you have someone capable of catching them.

3. Edward Mujica: become a closer
The Cardinals entered spring training with no shortage of prototypical power closer candidates. Incumbent closer Jason Motte led the NL in saves last season with a fastball that averaged almost 98 miles per hour. Trevor Rosenthal’s fastball averaged 98 during a dominant September and 99 in seven scoreless postseason appearances. Effective setup man Mitchell Boggs was more of a groundball guy than a strikeout artist, but his four-seamer averaged almost 97. So it seemed unlikely that Mujica—a 28-year-old with four career saves in seven seasons who we called “a reasonably solid seventh-inning arm” in Baseball Prospectus 2013—would be the one to get the bulk of St. Louis’ saves in the first six weeks of the season.

But that’s how it’s happened. Motte was sidelined by elbow issues that will force him to have Tommy John surgery next week. Boggs got the first shot at replacing him but blew two of his first four save opportunities and walked 12 batters in 10 2/3 innings before being mercifully sent down. Rosenthal blew a couple early-season leads himself and allowed a two-run homer to Ryan Braun in the same game as Boggs’ second blown save, giving him a superficially inflated ERA. So Mike Matheny turned to Mujica to solidify St. Louis’ pen, and so far he has.

Despite his lack of closing experience and his low-90s fastball, Mujica is a perfect 8-for-8 in save situations. He’s shown his usual superb control and his usual tendency to allow solo home runs, which hasn’t yet come back to bite him. He’s also shown an ability to miss more bats, which probably isn’t just a small-sample fluke. Since his trade to the Cardinals, Mujica has relied more heavily on his splitter (which he calls a split-change) at the urging of Yadier Molina and pitching coach Derek Lilliquist. With the Marlins last April-July, Mujica threw the pitch 39 percent of the time. With the Cardinals in August and September, he threw it 57 percent of the time. This season, it’s up to 64 percent, and he’s retired his slider and cutter. Since 2008, the pitch has been extremely effective, combining the low ball rate of his fastball with a much higher whiff rate and lower home run rate.













Obviously, Mujica won’t sustain a sub-.200 BABIP all season, and he’ll never be more than a few bad outings away from being replaced by Rosenthal, who has better stuff and a richer future ahead of him. But he’s another nail in the coffin of the notion that an effective setup guy needs ninth-inning experience or overpowering stuff to succeed as a closer for a first-place team.

4. Matt Carpenter: start at second base
Even now that they’re not as averse to going up in the zone as they once were, Cardinals pitchers lead the majors in groundball rate by three percentage points—the same margin that separates the second-place staff from the 13th. So it’s of some importance that their infielders be able to get their gloves on the ball. It wasn’t always clear that Carpenter—who played his first five games at second as a professional at the major-league level last season—could do that, but the work he put in at second over the winter appears to have paid off. The various defensive metrics approve of Carpenter’s small sample at second so far, and while the Cardinals’ batting average allowed on groundballs has risen slightly, that’s not necessarily attributable to Carpenter, or significant at this point in this season.*

*Incidentally, the average batting average on grounders in the NL is down from .238 in 2012 to .230 in 2013, while the average in the AL is up from .238 to .240. Overall defensive efficiency shows the same sort of pattern. I don’t know whether that’s noise or something pertaining to shifts.

With Rafael Furcal out for the season and Pete Kozma locked in at short, Carpenter solidifying second and pushing Daniel Descalso into a utility role has saved St. Louis from the fate of starting two light-hitting middle infielders. Carpenter has been one of the NL’s top five offensive second basemen, and his minor-league stats suggest that he can keep it up. Plugging that hole with a hitter has served St. Louis well.

5. Luke Hochevar: pitch out of the bullpen
Only one other pitcher has pitched in at least 100 games, started at least 95 percent of them, and had a lower ERA+ than Luke Hochevar’s 79. That pitcher is Kyle Davies, another recent Royal, which suggests that Hochevar really picked the right team to be bad for. Whether because they still saw some talent in his 6’5” frame, because they weren’t winning anyway, or because they were loath to concede that Evan Longoria, Clayton Kershaw, Tim Lincecum, or Max Scherzer might have been a better use of their 2006 draft pick, the Royals kept running out Hochevar long after most teams would have tossed him aside. Hochevar, for his part, kept failing to reward them for their faith.

Before this season, the Royals lined up some capable starters to make a push for the playoffs, so Hochevar lost what had looked like a lifetime rotation spot and moved to the bullpen, where R.J. Anderson predicted he’d have success with a simplified approach. R.J.'s prediction looks prescient. Although it hasn’t yielded more whiffs—most of his added swings and misses have come on the curveball and the sinker, which has been getting fewer grounders—Hochevar’s fastball velo has ticked up from 93.4 last season to 95.1 now, typical of a pitcher who’s transitioned to throwing in short bursts. But he hasn’t let go of any of his wide array of complementary pitches. Last September, Royals pitching coach Dave Eiland said, “He’s got fix or six different pitches. There are not many starting pitchers in this game, ever in this game, who have been successful for a long time who tried to throw that many pitches.” There are even fewer relievers who have had that kind of arsenal, but Hochevar hasn’t altered his approach as much as one would think. He’s throwing more four-seamers, sinkers, and curves, but he’s still mixing in cutters, changeups, and sliders a combined 16 percent of the time.

One area in which Hochevar has simplified things is his delivery. Doug Thorburn wrote last week that “the modern windup is borderline useless,” concluding that “a pitcher who focused all of his efforts on a single motion could reap huge rewards with respect to pitch repetition and command.” After alternating between the windup and the stretch last season,

Hochevar, like many relievers, has pitched exclusively out of the stretch in 2013.

He’s historically had a large bases empty/runners on split, which may have been mechanical in origin. In a tiny sample this season, Hochevar has had better results with runners on.

It’s too soon to say whether Hochevar will be setup man material—he still has a miniscule BABIP and a spotless strand rate—but it looks like he’ll be useful, which is more than he’s been able to say in most seasons. If Hochevar can go from being one of the worst starters to being a useful arm out of the pen, many relievers’ self-esteem will suffer.

Thanks to Ryan Lind for research assistance.

Thank you for reading

This is a free article. If you enjoyed it, consider subscribing to Baseball Prospectus. Subscriptions support ongoing public baseball research and analysis in an increasingly proprietary environment.

Subscribe now
You need to be logged in to comment. Login or Subscribe
the first and third SSC gifs are the same.

Thanks for this, cool article.
Thanks, fixed the GIF.
Does the line drive discrepancy in fielding data also exist in batting data (perhaps part of the reason for Choo's career low line drive rate)?
Am I the only one who thinks the ubiquitous rope-necklace thing looks silly?
"...the average batting average on grounders in the NL is down from .238 in 2012 to .230 in 2013, while the average in the AL is up from .238 to .240."

Harder ground / shorter/thinner grass in the spring?
"A lot of defensive stats tell a confusing story after six weeks"

Or six months.. or six years..
The 3 videos of Choo tell more than any fielding metric ever could.
The plural of anecdote is not data, regardless of whether the anecdotes are in animated gif form and hilarious.
You're wrong, of course. Observations (what's going on here rather than your condescending "anecdote") most certainly are data. They are simply not conclusive data, at least as far as fielding is concerned. They allow hypotheses to be framed -- in this case, the hypothesis that Choo is really, really uncomfortable in center field -- that can be supported or rejected by additional tests, which can take the form of more observations (hard to do in baseball because it's hard to make enough observations) or statistical analysis (getting better for fielding, although still imperfect) or better, a combination of the two. That's Scientific Method 101. Deal with it.
Except the hypothesis already WAS that Choo is uncomfortable in center field, people were saying that since 2 seconds after the trade was announced, so finding three instances of him goofing up is just feeding confirmation bias. These ARE anecdotes -- I saw him drop a catch! I saw the ball land behind him! -- because they were not gathered in any systematic way. Observational data would involve watching all of his plays, or at least some significant fraction thereof, and tracking how frequently he makes mistakes. And the conclusion drawn from that could easily be exactly the opposite, that he's actually incredible outside of the three mistakes. Maybe not, probably not, but we don't know that.

I would go so far as to say that conclusions drawn the 3 gifs are worse than meaningless, they're downright misleading, because they were all taken from the first week of the season. They absolutely do NOT support the hypothesis that "Choo *IS* really, really uncomfortable in center field" in any way, shape, or form, since we know he was essentially new to the position at the time and has much more experience now.

Derek Jeter is a great defensive shortstop because millions of Yankees fans say so.