Dexter Fowler and the Flare Conundrum

Great news, everyone! Dexter Fowler is back! Take a look at a Cardinals batting leaderboard, and you’ll see this lovely sight:

That’s right -- Fowler has out-hit (and bizarrely out-ISO’ed) Jose Martinez this year. What’s that, you say? Fowler’s a BABIP creation? Well, he’s running a .412 BABIP, so uh, yeah, you may have a point there, friend. That said, Martinez is at .408, so it’s not as though BABIP is the difference here.

Simply put, Fowler’s batting line this year is, at surface level, not too shabby. Yeah the BABIP is high, and yeah the ISO is low, but 11% walks and 23.5% strikeouts is better than average in this day and age. Take a look at the dark magic of xwOBA, and Fowler ‘deserves’ a .327 wOBA so far this year, better than league average. Case closed! Let’s spend the rest of this article celebrating.

.

.

.

Wait, what’s that? Everything’s not great? Alright, fine, let’s do a little more investigation. When I watch Fowler this year, I’m struck by how many flares he seems to hit. It’s a line drive in the box score, sure. It’s not as though it’s a ground ball that sneaks through infielders. Fowler has just gotten a ton of contact that isn’t solid enough to make it to the outfield, but is at the right angle to bloop over the heads of infielders.

A quick perusal of Baseball Savant shows graphically what I already knew viscerally. Take a look at a radial chart of Fowler’s contact so far this year:

That salmon section, ‘flares and burners,’ has more balls than anywhere else. Let’s zoom in on them for a second:

Well, Fowler has certainly made the most of his flares. That’s not all that crazy, either. So far this year, he’s running a .715 wOBA when he makes this particular type of contact. Before you stop and say how ridiculous that is, though, think about this. His career xwOBA on flares is .638. His career wOBA on them is .641. Maybe it’s a fluke that he’s hitting the ball just so, but it’s no fluke that he’s converting these types of batted balls into hits at a high rate.

I say, again, that the case is closed! Here I was, ready to say that Fowler’s contact was weak and only turning into hits because he was lucky, but that just plain doesn’t seem to be the case. Dexter Fowler has *always* lived on a diet of sinking liners, and all of baseball is right there with him. Between 2018 and 2019, the league has averaged a .624 wOBA on the flare/burner segment of contact quality. That’s good contact, is what it is.

Wait just one second though, Ben. Something doesn’t add up here. Fowler’s getting his walks, and he’s flicking balls to the short outfield with just as much efficiency as ever. In the past, though, those flares and burners were only part of Fowler’s recipe. He used to complement those skills with surprising power. With his average exit velocity on balls in the air way down this year, how come his batting line isn’t worse?

Well, there’s something suspicious about those batted balls after all. Take a look at Fowler’s contact quality in 2016, his You-go-we-go Cubs peak:

And then at his 2017, sneaky-effective Cardinals season:

This might not be obvious, but Fowler is hitting way more flares this year, as a percentage of his overall balls in play, than he has at any point in the past. It’s not by a little bit, either -- his 38% flare rate on balls in play so far is higher than it’s ever been, and his next-highest rate was 10% lower. Fowler’s 2019 is essentially the batted-ball version of ‘why don’t they make the whole plane out of the black box?’

This seems, at first glance, like a pretty lucky occurrence for Fowler. Happening to hit some extra not-particularly-well-struck balls at a just-so launch angle and speed might work for a bit, but eventually they’ll turn into just garden variety ground balls and pop-ups. That raises a question, though -- can we be sure of that? What if Fowler is just different now, more efficient at turning medium contact into hits?

I don’t really believe that to be the case, so I did a quick study. I split 2018 into two halves (ish, I used July 11 as a cutoff) and calculated ever player’s rate of flares per ball in play over each half. Then I checked the correlation between each player’s first-half and second-half flare rates. This is a pretty lazy study, honestly -- there are probably better ways to do it, though none that spring to mind at 10 p.m. on a Friday. This should almost definitely have a decent correlation -- even if some of it is luck, there’s surely some talent to it, and guys who hit a lot of flares in the first half are likely to be more talented than average at it. The data look like this:

That’s about as close to no relationship as you’re going to get. The r^2 is 13.5%, and that might be overselling it, because we’re only using samples of players who met minimum qualifying criteria in both halves. If someone started getting unlucky in terms of contact quality and then got demoted, they’d be missing from this list.

Look, this is a bit of an incomplete article. Something looked fishy to me about how Fowler’s hitting the ball. I think, for the most part, that I’m right -- something’s fishy! He probably can’t keep up this crazy rate of perfectly-struck-but-not-hard hits. On the other hand, I’m frustrated that I can’t say for certain there’s nothing there. It’s always hard to disprove a negative, and that’s where we seem to be right now. My intuition tells me that Fowler has been getting pretty lucky, and that he’ll fall back to the pack as the year goes on. The numbers, though, don’t really tell me for certain whether that’s true, or how far he’ll fall.

This is the crux of a lot of early-season analysis. Something looks wrong to the eye test, but there’s simply not enough data to say whether things are different one way or another. For what it’s worth, projection systems are a little more credulous than I am. They’ve bumped Fowler’s wRC+ up something like 3 points to 105, only 3 points below his career wRC+. Their guess is as good as mine, and probably better.

The truth is, most of this probably doesn’t matter. Even if Fowler’s been lucky to get his current results, he’s produced enough that the team is going to keep running him out there to see what he has left. Maybe they’re worried internally about the same contact-consistency issues I highlighted, and maybe they’re not. Either way, though, he’s hit well enough to keep getting more playing time. Would the team be better off banking what they have from him and moving on? Will his exit velocity on air balls rebound to offset an inevitable change in batted-ball mix? Only time will tell, but I feel good for having tried to figure it out.