Back in February, the state of Baltimore's outfield defense became a topic of discussion after longtime center fielder Adam Jones and executive VP of baseball operations Dan Duquette each made their thoughts on the situation known. Jones rightly pointed out that while Mark Trumbo and Seth Smith would help the offense, neither were considered strong outfield defenders. Duquette replied that there were perhaps some things the current outfielders could do to improve their defensive efficiency.
We tried to take all of that together and find some ways the Orioles could improve their outfield defense, ultimately deciding that since it was too late in the offseason to move Jones to a corner or find a starting-caliber defense-first outfielder, the best course of action might be to push the notoriously shallow playing Jones back a little further. Given that Jones is entering his 10th season as the starting center fielder in Baltimore, it's a suggestion that's been floating around for years, but it's never led to on-field changes.
Until this year, that is. While it's still early, Jones is one of 34 center fielders who have been on the field for at least 500 pitches in each of the past two seasons -- and no one who has been on the same team for both seasons has taken a bigger leap backwards than he has.
Biggest changes in center-field depth in 2016-17
+17 feet -- Jones, Orioles
+17 feet -- Carlos Gomez, Rangers
+15 feet -- Rajai Davis, A's
+11 feet -- Denard Span, Giants
+11 feet -- Tyler Naquin, Indians
+11 feet -- Lorenzo Cain, Royals
(minimum 500 pitches on field in both years, all data through Monday's games)
That's saying that Jones averaged 309 feet last year, and has now moved back to 324 feet, the 11th deepest of 38 center fielders with 500 tracked pitches in 2017. Here's what that looks like in visual form, with the dots representing the starting points for Jones on each pitch:
Gif: See Gif
Gomez and Davis each spent at least part of 2016 with a different club than they're on now, and that's important to note, because we're looking just at raw depth from home plate without accounting for the depth of home ballpark, though that's on the roadmap for this year. Still, this looks like something of a small league-wide trend. In 2015, all Major League outfielders averaged 312 feet from home, which went to 316 feet in 2016 and 317 feet so far this year.
"The number guys are smarter than the players," Jones told MLB.com when asked about the change this week. "It's weird playing a little deeper, but that's the way our front office wants me to play. I'm not insubordinate. I will do what they ask and sacrifice in other areas. That's what they see in the data."
We talked a lot last year about how much that helped William Fowler, who moved back 20 feet on the recommendation of the Cubs and suddenly, in his age-30 season, was viewed more favorably by advanced defensive metrics than he had been in years. It's not that he was necessarily a better defender, it's that he was put in position to be a better defender. Meanwhile, the Pirates movedAndrew McCutchenin by 10 feet, it backfired terribly, and he was moved off the position entirely before Starling Marte was suspended. Now, it's not as simple as "deeper is better," of course, because that's only two examples, and a team like Arizona played very deep as well, yet had some of the weakest outfield defense in the game.
That said, we had an inkling this might happen for Jones, didn't we? If you remember back to the spring, his amazing catch to rob Baltimore teammate Manny Machado in the World Baseball Classic wasn't just the enduring moment of the tournament, it's something we're going to see replayed on highlight reels for years to come. But as we noted at the time, Jones was playing 14 feet deeper than his 2016 average for that play -- deep enough that it might have been the difference between getting there and not. We have 664 tracked balls since 2015 where Jones was the first to touch it, and his positioning on that play was deeper than 97 percent of them.
As you can see here, the 2017 starting points on balls Jones is getting to (both outs, and hits where he was the first to touch it) are well deeper than they were in the previous two seasons.

And this is mostly a Jones-specific change. While the Orioles' left fielders have moved back by one foot (from 298 to 299) and the right fielders have gone back three feet (295 to 298), neither corner has moved back nearly as far as their long-time center fielder has.
Now for all that, the question really comes down to this: Is it working? Does it make the O's outfield defense better? The answer to that, to be perfectly honest, is that it's too soon to say. This is more of an "early trend to watch" than it is a concrete statement of success or failure.
Still, we can at least point to some interesting early signs. Let's show you two videos of very similar batted balls, one from this year and one from last year, with extremely different results.
In one particular game last September, Jones was positioned 303 feet away from home plate with Richie Shaffer up to bat. Shaffer crushed the ball a projected 393 feet away to the warning track, and Jones was unable to track it down, allowing the Rays to break a 2-2 tie in what would become a 5-2 Tampa Bay victory.

Now, let's see a play from earlier this month against the Blue Jays, when Kevin Pillar hit a similarly-angled ball a projected 398 feet away, also to the warning track. While the ball did have about a half-second more hang time, the big difference here is that Jones was positioned 330 feet deep. So instead of having to run 90 feet to try to track down the ball as he'd been unable to do against Shaffer, he needed to run only 68 feet for Pillar. Not only did Jones manage to get there, he made it look so easy that he blew a bubble while hauling it in.

When we look at the type of ball we've termed "Barrels," the high-value batted balls that generally fall for a hit more than 80 percent of the time with a slugging percentage near 3.000, we can see that in 2016, the Orioles allowed the eighth-highest slugging percentage last year, and now they're allowing the fifth-lowest slugging percentage this year. What that's saying is that at least in the early going, those extremely dangerous batted balls aren't hurting Baltimore as much as they did last year.
Of course, Catch Probability shows that the Orioles' outfield are getting to the seventh-fewest number of balls of any team, collecting just 80 percent of chances. And our brand-new "Sprint Speed" metric, which measures how fast an outfielder runs in feet per second over his fastest one-second window, suggests the same. Jones's sprint speed mark is 27.2 feet per second, almost exactly league average. Smith, Trumbo and Hyun Soo Kim all came in between 25-26 feet per second, well below average. As Jones said, this is not a particularly athletic outfield.
"I've always played to my comfort, my knowledge of hitters, my knowledge of the counts, my knowledge from being out there for a lot of innings, seeing a lot of hitters, a lot tendencies," said Jones. "But I understand data. They feel my defensive metrics will get better if I back up."
It worked for Fowler, another notoriously shallow center fielder. It doesn't work for everyone. Still, for Jones and the Orioles, this step back seems like a step forward, at least very early in the season.