Is Center Field Cody Bellinger’s Best Position?
Cody Bellinger is in the midst of an MVP-caliber season. His 7.3 WAR is second in the majors, his 166 wRC+ ranks third, and he’s among the league leaders in almost every offensive category. Barring a major surprise (from, say, Anthony Rendon), the NL MVP should come down to him and Christian Yelich. In addition to his incredible production at the plate, Bellinger has improved by leaps and bounds in the field. He’s putting up elite defensive numbers in right field for the Dodgers, and now they’re planning on moving him to center field full-time for the rest of the season.
Roberts said Cody Bellinger could play mostly CF down the stretch b/c of his range, with Pollock in LF.
— Ken Gurnick (@kengurnick) September 2, 2019
The inconsistent play of A.J. Pollock in center provoked this move. An elite center fielder in his time with the Diamondbacks, Pollock’s defense has slipped terribly this year (-8 DRS, -5.8 UZR); meanwhile, Bellinger’s defense has improved dramatically this year (19 DRS, 9.8 UZR). The Dodgers are hoping this shuffle will give them the optimal alignment in the outfield in their quest for a championship.
At this point in his career, it’s more accurate to call Bellinger an outfielder than a first baseman. A few weeks ago, his major league time spent on the grass surpassed his time spent on the dirt. When he was drafted out of high school, Bellinger’s defensive scouting reports often mentioned that he had the athleticism to play in the outfield, but only in a corner. Here’s how Bernie Pleskoff described his defensive potential in his MLB.com scouting profile: “If needed, he could be a successful outfielder. I don’t think he has the speed to play center, but I think he could succeed in right.”
The Dodgers began giving Bellinger time in the outfield in 2015, two years after he was drafted. He’s seen time at all three spots and that versatility has definitely benefited his team. After Corey Seager was injured for the year, Bellinger was thrown into center for a little less than 500 innings last year. He played very good defense at the hardest position in the outfield, including making this extremely difficult catch in the World Series:
Here’s how the advanced defensive metrics grade his abilities at each outfield position:
| Position | Innings | DRS | UZR | OAA |
|---|---|---|---|---|
| Left | 315.1 | 2 | -1.1 | 4 |
| Center | 584.1 | 8 | 2.1 | 6 |
| Right | 947.1 | 21 | 10.0 | 7 |
Both DRS and UZR absolutely love his abilities in right. His 19 runs saved per DRS is tops in the majors this year, and his 9.8 runs saved per UZR is second after Mookie Betts. Baseball Savant’s Outs Above Average (OAA) is a little more even in its appraisal of his defensive abilities.
While DRS and UZR are both useful tools, I prefer OAA’s granularity a little more. Statcast uses a ball’s hang time and the fielder’s distance from the ball’s landing spot to calculate a catch probability. From there, we can calculate how many balls a fielder caught and compare that to the average catch rate for similar balls, giving us Outs Above Average. Using directional OAA, we can begin to build a profile for Bellinger’s strengths and weaknesses in the outfield. Directional OAA splits his fielding skill into six segments based on the direction he’s moving to get to the ball’s projected landing point. Here’s how he fared across the six directional slices in his three years in the majors:

His biggest improvement has been on his ability to come in and make plays in front of him. When the ball is hit over his head, he’s been strong at tracking to the left but has struggled going straight back and to the right. Here’s a scatter chart of the balls he’s had to go back on that have fallen in for hits (ignoring directionality):

There’s a smattering of balls in play with extremely difficult or impossible catch probabilities towards the bottom of the chart. The easiest balls to catch that have fallen in for hits anyway were complicated by the outfield wall (the dots outlined in green). Those hits at the wall have an extra amount of difficulty beyond simply getting to the right spot at the right time.
Outfielder Jump is a new Statcast fielding metric, which was introduced this year. It takes three components — reaction, burst, and route — to calculate how good of a jump an outfielder gets on a batted ball hit to him. Unsurprisingly, Bellinger’s ability to get a good jump on the ball gives him a huge advantage in the field. His burst is particularly good, which makes up for his less than ideal initial routes. His sprint speed sits in the 89th percentile so even if his first step is in the wrong direction, he has the ability to make up the ground with his long legs.
Traditionally, center field has been one of the highest positions on the defensive spectrum with first base and the outfield corners some of the lowest. Based on those assumptions, what the Dodgers are asking Bellinger to do is rather risky. He’s shown that he’s an elite defender in right but moving to center could erase all of that defensive value simply because of the positional adjustment and greater difficulty of the position.
Before he left Baseball Prospectus for a job with the Mets, Russell Carleton revisited the assumptions of the traditional defensive spectrum. He concluded that outfielders, no matter what their regular position is, can usually switch between the three outfield positions without serious consequence.
“Emergency center fielders have—and this is not a misprint—performed at roughly the same level defensively as regular center fielders over the years. That could be an artifact of our sample. These might be fourth outfielders or corner guys who are fully capable of playing center, but never get to do it because there’s someone better on the depth chart.”
Carleton goes on to show that infielders who make the switch to outfield have a much harder time adjusting to the new position than an outfielder sliding over to another outfield position. And while the data show that center is the hardest position in the outfield, the drop off from center to the corners might not be as steep as we thought.
The Statcast defensive metrics seem to confirm these revised assumptions about the defensive spectrum (at least when it comes to outfield defense). The implications are especially interesting for Bellinger. His raw abilities give him a great jump on balls hit to the outfield and he’s developed a solid fielding profile based on where the ball is hit. Those skills should transfer no matter which outfield position he’s playing. Reading the ball off the bat and the spin of the ball in flight will take some minor adjustments, but the strong foundation is there. If sliding over to center isn’t going to significantly affect his defensive abilities, the Dodgers may have stashed one of the best center fielders in the game in right field for most of this season.
Jake Mailhot is a contributor to FanGraphs. A long-suffering Mariners fan, he also writes about them for Lookout Landing. Follow him on BlueSky @jakemailhot.
That Carleton analysis is interesting, considering how much the Mets have prioritized putting Conforto and Nimmo in center when given the opportunity.
I guess one other way to think about this is that it’s not just about Bellinger being good in center, it’s about being better than Pollock. Pollock anecdotally and by the numbers, appears to have taken several steps back in CF defense, and if he can be an average or above average LF and Bellinger plays even average defense in CF, that’s a net upgrade for your team. The best defensive configuration would most likely be Verdugo in CF, Pollock in LF and Bellinger in RF, but if Verdugo’s not going to be back, this probably makes the most sense if Pederson or Beaty is in the OF too.
The problem with statcast’s OAA is that it’s not on the same scale as the other metrics. It does not assign run values to the outs, and so it is hard to compare them directly to the other two metrics. Bellinger is 14th on the OAA leaderboard for this season, and the 13 players ahead of him all play primarily center field (and Jason Heyward, whose innings are split 60/40); OAA does not distinguish between OF corners and CF, and center fielders consistently score higher (whether because their plays are inherently harder or whether they have more opportunities, or some combination thereof).
OAA actually supports the analysis that Bellinger has been one of the very best right fielders in baseball this year (he is tied with Betts).
The lack of like-for-like comparison in both positions and run values make using OAA tricky.
In your own chart, you can see how Bellinger has been 1 OAA per 100 innings in CF, and 1 DRS per 70 or so innings in CF; his 500 innings are a fairly small sample, but that is outstanding performance! The Dodgers’ decision to put Bellinger back in CF makes more sense in the context of the fact that he did it last year and was quite good at it.
By this time the question isn’t CF or RF, it’s why the Dodgers left at 1B in the minors. The fact the he successfully converted on the fly at the MLB level (to CF) is just amazing.
For what it’s worth, and I could be wrong about this, I feel like I remember scouting reports saying at the time that he was called up that scouts thought he was capable of playing an above-average outfield (in a corner spot, although they thought he could cover CF in a pinch). However, they felt that he was such a superlative defensive first baseman that it was worth keeping him at 1B despite the positional adjustment.
With the benefit of a few years of advanced metrics (and hindsight), the Dodgers and the scouting industry seemed to have definitely been too low on his outfield abilities, and also potentially too high on his 1B abilities. Someone should fact-check me on this info, though, since this is all based on my (potentially faulty) recollections.
This is correct, however, I don’t think they were too high on his 1B abilities. He is a pretty great 1B too. He is just a way better outfielder than they thought partly because it seems the scouting community underestimated his speed.
Ok, thanks for the feedback. I agree he’s amazing as a first basemen. But if you get a bat like that at CF you keep 1b open for the next David Ortiz, should one just happen by. Or, you know, the next…..Yordan Alvarez….
Or a Max Muncy.. which then leaves 2b open for Gavin Lux next season.
That first highlight is a good example of a lot of things.
What kind of MLB CF would not make that catch? I am not buying that as an 8% play. That ball is caught 90% of the time for sure. It would be a gross misplay if it wasn’t caught. That is a good example of how bad Statcast is at measuring defense. I am not 100% certain what could contribute to that but that was pretty routine. There have to be significant errors in those measurements.
That is also a good example of bad instincts. Belly takes a bad route filled with indecision. That is what I have seen a lot out of Belly in the OF.
Lets talk metrics. It sounds to me like all that you are actually measuring is how poorly a player is positioned. That is what distance covered is from one perspective. The general way we attempt to quantify defense is just really far removed from reality. Players do not get a normal distribution of attempts. What you catch is very much dependent on what is hit. Binning of balls is overly crude. x-angle is an afterthought and a bunch of realities are non-existent in the metrics.
The more I pay attention the more I wonder if bad defense isn’t valued by machines. Not saying that errors are rewarded, but clumsy play seems valued by machines. Isn’t it fun that humans and their inability to differentiate clumsy from difficult serve as the impetus for defensive metrics in the first place. In a corner OF spot you really don’t have to do much other than catch what is hit at you but in CF you kill your team by taking bad routes. To mistake corner OF defense as being valuable one way or the other is just disconnected from the on field outcomes. See Kyle Schwarber or any team that has a great corner OF. Its not a big influence on the game.
I think you want to refrain from saying things like “the drop off from center to the corners might not be as steep as we thought”. That just isn’t true and there are 100+ years of anecdotal data that don’t support it. Everyone with on-field experience is going to disagree. When you have an entire world full of experts and then some new theory that goes against it, you might want to consider that seriously. This is the same “analysis”that promotes the idea that SS defense isn’t all that valuable and that several of the best defensive season of all-time are by catchers over a five-year window. Its very much the blind leading the blind as the same people using the same bad tools to create the proposals use them to measure the efficacy. Never mind the conflict of interest. There are so many reasons to question defensive metrics. It is likely that the methodology being used to measure defense is wrong – the only question is to what degree and how.
The fact he made that catch look as easy as you think it is only supports the idea he is a good CF.