What’s eating Dansby Swanson?

Disclaimer 1: This post is long. It also has a lot of images (charts, tables, etc.). Sorry if it takes a while to load.

Disclaimer 2: If you’re expecting a silver bullet explanation that answers the title question, I’m afraid you might leave this post disappointed. This is really just a sojourn through a lot of hitting data, and I’m afraid there’s no single super-satisfying answer.

Coming into the season, the Atlanta Braves roster did not feature an overwhelming array of bright spots. Freddie Freeman looked prime to devour the National League with his bat, newly-extended Ender Inciarte is a treat with the glove and has had average success with his bat, and a number of rotation arms featured some interesting upside. But, much in the way that our lizard brains love shiny new things, there was also a ton of excitement about rookie shortstop Dansby Swanson getting his first full season in the majors.

Swanson made his major league debut in mid-August of last season and ended with a 107 wRC+ and 0.8 fWAR in 145 PAs (about 3.3 fWAR/600). A good chunk of that value was buoyed by a .383 BABIP and the fact that one-third of his home runs were of the inside-the-park variety, but it was still a pretty exciting showing for a 22-year-old with just over 700 professional plate appearances in his lifetime. Coming into 2017, the projections were more skeptical of sustained success: Steamer had him at an 89 wRC+, while ZiPS was at 92. (IWAG, loving his minor league numbers, pegged him at 108, in a large deviation.) Suffice to say, whatever pessimism Steamer and ZiPS were injecting into the universe, what’s actually ended up transpiring has been far, far more dreadful.

Through Thursday’s game, Swanson has put up a miserable .138/.169/.188 line, which yields a .160 wOBA and a -6 wRC+. If you’re wondering what a -6 wRC+ means, well, it means that Swanson has produced 106 percent fewer runs than an average player, or something fairly close to that. Awkward. Swanson has walked just three times, while striking out 22 times (over a quarter of his plate appearances). It also probably does not make anyone particularly happy that, entering play on Friday, Swanson has the worst wOBA and wRC+ among all players qualified for the batting title. Life is misery and full of pain.

So, Swanson has been bad. We know it, Freddie Freeman knows it, the Mets announcers were talking about it during Thursday’s game, and you know that no one feels it more than Swanson himself. The Braves don’t appear to have any interest in demoting him and letting him work out his issues in the minors, so this is something he’ll have to battle through while facing major league pitchers, most or all of whom are actively trying to win meaningful games for their teams.

Below is a somewhat haphazard collection of the problems that have plagued Swanson in April 2017. Some are old. Some are new. Most are ineffably tinged with that ever-present bugbear: small sample size. It’s still April, and we’re still at the point where one good game can move the needle a bunch, and a good series can give a statistical profile a whole new flavor. But, what’s done is done, and it hasn’t been very pretty. Let’s take a look.

---

Sliders

No, not the TV show, but this very popular breaking pitch has given Swanson a bunch of trouble. (This post from Tomahawk Take is two weeks old, and just as salient as ever.) But, the thing is, this is not a new development. Let’s pretend 2017 hasn’t started to happen yet. Here’s Swanson’s 2016 line against sliders:

One, that’s a lot of sliders. Two, that’s a lot of whiffs. Three, that’s a really problematic ball-in-play profile against sliders. Okay, let’s go back to the present day.

Yikes. It got worse.

Some other problematic stats: 1) Swanson has whiffed on 37.5 percent of the slider’s he’s swung at; and, 2) only eight percent of the sliders he’s put in play have been line drives.

It appears that, in general, Swanson has a bit of a breaking pitch problem (that is, throw curves in with the sliders), but has feasted (relatively) on fastballs and changeups. Cutters elicit more whiffs from him, but he hits them well when he does make contact.

Pitchers will just keep throwing him sliders, though, and he’ll have to adjust. Such is life for a young guy in the majors.

---

Outside Pitches

This one is interesting, because it’s bandied around a lot, but doesn’t seem to be as much of a problem. Again, let’s look only at 2016 first:

Pitchers definitely love pitching him down and away, but if the ball gets belt-high, it’s not a huge issue. The exit velocity is also telling: it’s definitely worse, but not egregiously worse. Okay, back to 2017:

Similar approach from pitchers, with perhaps even more of an emphasis on outside pitches, but he’s not faring terribly, especially if those outside pitches are up a bit. Compared to some of the other things, I’m not sure this is worrisome in and of itself, though you can tell that he’ll probably be in for a frustrating night if a pitcher can just pound that corner low-and-away over and over again.

---

Balls in Play and Luck Tigers

(Why tigers? Because they’re the opposite of dragons, apparently.)

Let’s be clear - this section is in no way going to argue that bad luck on balls in play is the entirety of Swanson’s problems so far this season. But, they’re still some part of his problems.

Swanson’s BABIP is .179. That’s the seventh-lowest BABIP among all players currently qualified for the batting title, putting him in the bottom four percent. Here are some other percentile ranks among this same group:

• Line drive percentage: 47th
• Avoiding infield pops: 78th
• Hard-hit percentage (Fangraphs): 18th
• Avoiding soft-hit percentage (Fangraphs): 70th

Those numbers aren’t saying he’s absolutely creaming the ball, but putting up one of the lowest BABIPs this season with those parameters speaks to some definitely luck tiger-ing going on.

Of course, we don’t need to only look at those stats, especially now that we have Statcast data! Swanson’s 88.3 mph average exit velocity is pretty middling. You can sort by every other stat on that leaderboard, and you’ll see that Swanson’s been throwing a pretty average batted ball profile out there: he’s hit balls in the air a little weaker than average, balls on the ground a little harder than average, and his rate of barreling up balls (i.e., hitting balls in a way that often has a good result) is also pretty quotidian.

Possibly one of the coolest things that Statcast does is track balls in play and give each ball a hit probability based on its exit velocity, launch angle, and distance traveled. (Note: one thing that Statcast doesn’t do is apply location to this calculus. So it gives you a sense of how well a ball was struck, but not how easy it was to catch given the defensive alignment used by the pitching team, or the unique contours of the ballpark, etc. For example, Adonis Garcia hit a home run down the line earlier this season that had a hit probability of just 33 percent.)

At one point, I started to track this manually, because my own eye test seemed to suggest that Swanson was getting really hosed on balls in play. The reality, well... you decide:

The black solid line is Swanson’s actual hit total, while the blue dotted line represents what Swanson’s hit total should have been by adding up the partial “hit probability” percentages for every ball he’s put in play. Right now, the gap is about four hits; at one point a few games ago, it was around five hits. Among these balls in play have been two lineouts that go for hits around 75 percent of the time that were both caught; on the other side of the ledger, Swanson tallied an early-season single that only goes for a hit 13 percent of the time.

Of course, the sad truth is that four more hits would not transform Swanson’s batting line into milk and honey at this point. Even if we were to magically bestow upon him five more doubles in lieu of five outs, he’d have a wOBA of around .237. That’s pretty much the same wOBA that A.J. Pierzynski had last year. Gross.

But don’t just take my word for it: Baseball Savant actually calculates “expected wOBA” based on these same ball in play characteristics, and lets you make charts with them. Swanson is the highlighted dot below. Sure, the expected wOBA of around .230-.240 is better than the current wOBA of .160, but neither number is really acceptable for a major league hitter.

(Special thanks to BravesRays for pointing me to these data, so that I could stop calculating this sort of stuff manually.)

Long story short: yes, Swanson has had some bad ball in play luck, but it is hardly the complete and total explanation for his offensive ineptitude so far.

---

Various Plate Discipline Stuff

In all honesty, I’m not too sure what to make of this: it’s still early, and it’s speculative at best. But, there may be something there, or not. You decide.

Swinging at strikes (also called z-swing) is good. Swinging at balls (also called o-swing) is generally pretty bad (unless you’re annoying and foul them off repeatedly, and then you’re just annoying). If a player swings at strikes and doesn’t swing at balls, good things will probably happen for that player in the long run. If a player swings at balls and doesn’t swing at strikes, something has gone terribly wrong. That’s why the gray-shaded part below is kind of scary.

This is a tiny sample. And it’s a tiny stretch of rolling averages. But you can see a story start to form. Dansby Swanson grew more aggressive over his first ten games. He was especially on point at swinging at strikes, for all the good it did him, but he also swung at a greater proportion of pitches outside the zone. Then, something strange happened. He started swinging at fewer strikes... but was still swinging at a lot of balls. That’s no good. Most recently, he’s really dialed back the swinging in general.

This chart won’t tell you anything you can’t figure out from the prior chart, but it just adds some more context to the story. Swanson started off seeing about an average number of pitches per plate appearance (the average was 3.88 in 2016), then got really impatient, and has rebounded to be perhaps overly patient. His most recent game was an interesting microcosm of this tension:

• First PA: walked on five pitches (only his third walk of the season!);
• Second PA: works full count, hits into forceout on seventh pitch (with bases loaded, none out!);
• Third PA: swinging strikeout on six pitches; and
• Fourth PA: swinging strikeout on four pitches, with the fourth pitch being a fastball high and out of the zone.

You can’t say he’s not out there trying to make adjustments, and I think these charts shed some light on that. Unfortunately, some of those adjustments may be works in progress, and it doesn’t seem like anything has clicked yet. Still, he’s out of that gray-shaded danger zone for the moment, and perhaps being a bit more patient in the future will prevent pitchers from exploiting him with sliders down and away off the plate. (Last season, it seemed like he dealt with that issue by letting that pitch go for a ball enough that he could sit on a fastball in a deep count and do something with it. Not sure if that’s what he’s trying to do now, but at this point, it couldn’t hurt.)

---

The Big Finish

Alright, I admit that without this section, I wouldn’t have bothered making this post. It’s still not a smoking gun. I just find it weird and fascinating, and I’m really, truly sorry that I had to encounter this bit of information at the expense of Swanson’s suffering.

First, let’s look at an overall, league-wide heat map for contact rate.

Major league hitters are pretty good at making contact on pitches inside the zone, and worse on making contact on pitches inside the zone. This should not be that surprising. Hitters find it easier to make contact away from the corners, for the most part. Not surprisingly, down the middle is the easiest location for a hitter to make contact.

Let’s check in on Dansby Swanson (2017 edition).

Uh-oh.

76% contact on pitches right down the middle? How is that possible? (By the way, this was decidedly not an issue in 2016.) So, you think, “Look, Swanson is struggling. I’m sure all sorts of batters have issues with even pitches down the pipe when they’re mired in a slump.” To which I say... well... if only.

Devon Travis, the only other full-time regular that currently sports a negative wRC+? Nope, he’s doing just fine on making contact with pitches down the middle (94%). The reanimated remains of Jose Reyes, currently playing third base for the Mets? Pretty much league average at it, despite being terrible at everything else (88%). What about Byron Buxton, another young player who’s struggled massively at the plate and with making contact? Sure, he’s got problems, and an 80 percent contact rate on pitches down the middle is fairly awful. But it’s not 76 percent...

But, don’t worry, Dansby Swanson has some company on this Braves team as far as this ignoble stat goes. 75% is awfully close to 76%.

Any guesses? Think hard. Real hard.

Okay, now scroll down.

I’m so sorry, Dansby (and everyone).

Of course, the problem goes beyond just making contact with pitches down the middle. Surprise, surprise, there’s also a problem associated with actually doing something with them even when a juicy pitch comes by.

Consider Exhibit A:

I didn’t intend for the chart to work out that way, it just happened. But essentially, 80% of down-the-pipe pitches Swanson has seen have not been fruitful. A fifth were just watched, which is mostly fine, if not optimal. A fifth were spoiled. Two fifths were put in play, but only one fifth resulted in hits. And, of course, a fifth were missed entirely.

But, hey, a .500 BABIP on pitches down the middle is exciting, right? Well, consider Exhibit B:

This awesome-looking chart is not presenting a replica of a baseball field. Instead, it’s showing launch angle and exit velocity, and color-coding zones based on that quality of contact. Of the ten pitches down the middle that Swanson has put into play, he’s gotten five hits. But three of those hits came on topped pitches (beaten into the ground), when that’s exactly what shouldn’t be happening on a pitch down the middle. The other two hits were weakly-hit flares, another unexciting outcome. Of course, the one ball that Swanson creamed, in the “barrels” dark red territory, was caught in left field by Travis Jankowski. Another three pitches were routine flyouts that Swanson got under.

There’s a lot of opportunity for damage on these pitches, but by whiffing on some, fouling others, and making unimpressive contact on a bunch more, Swanson has tanked his own batting line in a particularly cruel way. Is this a timing issue? A hitting mechanics issue? A swing plane issue? I don’t know, and I’m not qualified to answer, or even speculate. But, whatever it is, it’s definitely an issue, and one that Swanson can’t afford to have linger much longer if he wants to improve his offensive output.

---

I have no idea what’s going to happen next. Theoretically, Swanson will continue to participate in that multifaceted waltz known as “hitters adjusting to pitchers who then adjust to the adjustments and on and on forever.” Yeah, it’s not a very catchy name.

With that said, though, Dansby Swanson has put himself in a somewhat of a hole as far as his overall season line. While the Braves probably really only care about his performance going forward, and not his aggregate line that includes what he’s done to this point, in actuality, he doesn’t need to exceed expectations by that much to get back to respectability by season’s end. If he finishes the year with around 600 PAs, he only needs to hit at a 103 wRC+ clip from this point forward to end with a full-season wRC+ of 89 (the Steamer pre-season projection). If he hits at an 89 wRC+ going forward, his full-season wRC+ will be around 76.

Of course, in order to pull his batting line up, he’ll actually have to hit at one of those better rates going forward. History, meanwhile, has not been very kind to starts as bad as Swanson’s.

The last time that a full-time player had a negative wRC+ in April and finished with a wRC+ in the 70s was 2005, when both Yadier Molina and Jack Wilson did so. Molina had a -5 wRC+ in April and finished with a 71 wRC+ in his sophomore season, while Wilson converted a -1 April wRC+ into that same 71 wRC+ mark. Of course, Swanson also has a few games in April left to start pulling up his batting line.

Speaking of seasonal turnarounds, the last time one of the league’s worst hitters in April went on to post an above-average batting line was in 2011, when James Loney accomplished the feat. Loney’s 2011 was a completely bizarre season: his monthly wRC+s were 27, 118, 122, 23, 196, and 175. Still, that 27 mark was the second worst in baseball that April, and Loney went on to finish the year with a 110 wRC+ and 2.1 fWAR.

If he’s going to rebound, Dansby Swanson has his work cut out for him. But, though rare, it has been done before. Let’s hope he gets going soon.