Much a-Du about something that is probably nothing
Oh, hey! We’re technically in the offseason now. It’s kind of funny, one day you go to sleep with the Braves having won the final game of the baseball season, and the next day, you’re probably pivoting to roster construction. So it goes. This post, though, isn’t meant to weigh in on any side of a discussion about whether the Braves should retain Adam Duvall for the 2022 season. Rather, it’s just a series of things I found interesting, but seemed out of place to talk about as they were happening given how rapidly the Braves changed their fortunes in the regular season. So, with that said, it may not even be particularly relevant to the Braves (or anyone else), but if you like slight twinges of baseball weirdness, read on.
Before anything else, take yourself back to last offseason. The Braves had some lineup holes that needed patching. Whether or not to tender Adam Duvall was an open question, and Marcell Ozuna was a free agent after a very successful, albeit shortened, 2020 campaign. The Braves had a lot of potential free agent options, though many were of the platoon-ish kind, including (amusingly in retrospect), Joc Pederson and Eddie Rosario. One popular framework for a corner outfield spot for the prospective 2021 Braves was a platoon featuring Pederson and Duvall. Pederson had a career 128 wRC+ against right-handed pitching, with a similar 131 mark over the past three seasons. (Since 2015, the xwOBA and wOBA for him against righties were not that far off, in the 2018-2020 period his wOBA of .365 substantially outpaced his xwOBA of .345 in that split.) He made sense as the strong side of a platoon, as keeping him away from southpaws could easy mitigate his main weakness.
As for Adam Duvall, though… sure, he hit right-handed. And, he had a reputation as a lefty-killer, a platoon guy in the very specific sense of that term. It’s not that the thought was outlandish or anything…
…but his splits against lefties also may not have been quite what you’d expect. From 2014-2020, Duvall only put up a 106 wRC+ against southpaws. In that same span, the collective set of all MLB right-handed batters put up a 103 wRC+ against southpaws. In the broad view, Duvall was a platoon bat only in the sense that he hit lefties better than righties (92 wRC+ for his career through 2020, which is identical to the right-on-right split for MLB as a whole in that period). Sure, Duvall crushed lefties in both 2019 and 2020, but in fewer than 100 total PAs. He had 126 such PAs in 2018, and scuffled — his 2018-2020 aggregate against southpaws was just a 104 wRC+. (In xwOBA terms, too, none of these year ranges really show much deviation, so I’m not omitting any particularly relevant information here; his wOBA and xwOBA are within .006 of each other across these various time horizons.) What’s the point of all of these paragraphs? Mostly just to say that the Pederson/Duvall platoon, had it come to pass, wouldn’t have been about two guys that definitely crush opposite-handed pitching. It would’ve been shoring up Pederson’s weakness with a guy who hits lefties at a decent clip, and that’s it. Of course, none of this ended up being directly pertinent to the 2021 Braves, but it’s still an interesting footnote to the overall annals of how we think about player skillsets.
Why bring any of this up? It goes back to the middle of the 2021 season, when the Braves were mired in some particularly nasty doldrums. Lamenting the absence of Duvall was, for some, the style of the time, especially because Duvall was tormenting his former team. In his Miami uniform, Duvall faced the Braves in 11 games. He homered against them in his first game, and then hit two homers in his second game. Later, there was a pinch-hit double, and a game with a homer in each of the next two series against the Braves. In total, Duvall hit five homers and put up a 214 wRC+ against the Braves in 43 PAs against them. He had 23 percent of his Miami longballs in the 13 percent of his PAs that came against the Braves. Meanwhile, his overall wRC+ on the season was just 102. To put it in starker relief:
- Adam Duvall, 2021, against the Braves: 214 wRC+
- Adam Duvall, 2021, against everyone that wasn’t the Braves, as a Marlin: 85 wRC+
- Adam Duvall, 2021 as a Marlin overall: 102 wRC+
The best revenge apparently isn’t a life lived well, it’s apparently being pretty bad except against the team that non-tendered you. (And no, Duvall’s xwOBA as a Marlin was within .001 of his wOBA, he wasn’t getting thumped by outcomes or anything.)
So, naturally, when the idea of finding reinforcements for Atlanta’s rent-to-shreds outfield came around in July… Duvall was certainly a target, maybe even a popular one… but about the same level of slam dunk-level improvement as he was as an offseason platoon candidate: better than nothing, but not exactly a savior. (To which notion, of course, as you know, both Duvall and the universe hoisted a big ol’ middle finger after the Braves acquired him anyway at the Trade Deadline.)
There was also something else going on. Remember that image above about Duvall’s track record against lefties? Well, in 2021, before he was dealt to the Braves, Duvall was doing something even weirder. Or, well, was having something weirder happen to him. The following two charts show Duvall’s platoon splits for each season. One chart is wRC+ (outcomes), the other chart is xwOBA scaled to league average (without adjusting for park, just so it’s set to have a 100 equals league average thing so it’s easier to compare).
In short: in his 2021 tenure with the Marlins, Adam Duvall had a reverse split (very dramatic by outcomes, still borne out by inputs) for the first time since 2018 (when he was basically just equally meh-to-bad depending on whether you look at inputs or outputs), with a notable reverse split for the first time since 2015, when he had very few PAs overall.
Players with reverse splits are awkward. For one, reverse splits don’t really seem to have much foundation in reality, with smaller-sample reverse splits eventually evening out in the expected direction most if not all of the time. For another, their appropriate usage gets messy. If Duvall really was his 2021 version, who hit lefties poorly and righties okay… when would you use him? You couldn’t use him as the weak side of a platoon, but at the same time, you could just find a lefty who hits righties well. These thoughts were pertinent in terms of the Braves possibly acquiring Duvall in 2021 — even if they did, how would they use him?
This strange, half-season reverse split led to a further line of questioning: why was it happening? This answer, too, was somewhat puzzling.
We’ll just go ahead and start with “the answer” here, such as it is.
Long story short, Duvall has basically existed as a major league hitter by hitting fastballs. That’s not weird, that’s kind of the default, but it definitely encapsulates his career. Duvall’s 2019-2020 re-emergence with the Braves came because he pounded fastballs — especially in 2020, when he went berserk and posted .400+ xwOBAs on anything in that category, including a mind-scorching .500+ mark against fastballs of all types from lefties. Notably, in those seasons, he was terrible against non-fastballs in general, more terrible than he was when he wasn’t pounding the snot out of fastballs, too. There’s a few quirks here and there and I won’t go over the entire table (for example, what was up with Duvall and four-seam fastballs from lefties in 2019, given what he did to non-four seamers from lefties?), but the gist is that when Duvall feasts, he feasts on fastballs…
…which is what made his 2021 Marlins stint so weird. If you look at the top table (against all pitchers), his performance on fastballs declined to its lowest level since 2017, even lower than his disappointing 2018 season. This happened with Duvall getting fastballs from righties (with a modest, not particularly meaningful improvement against non-fastballs), but was eye-popping in terms of how poorly he handled fastballs from lefties, against which he was straight-up horrible as a Marlin (again, with a modest but not super-useful improvement against southpaw non-fastballs).
Basically, this is the explanation for the weird reverse split: Duvall, who had most recently killed lefty fastballs to a nearly-incomprehensible degree, was now awful against them. He actually got worse against fastballs from either arm, but the lefty drop was so precipitous as to give him a reverse split.
When you think about why a player might struggle against fastballs, a concern might be bat speed, which falls into the “time comes for all of us” category, i.e., aging. Duvall is no spring chicken — while he seems like he hasn’t been around that much given that he’s still under team control for 2022 and has fewer than 2,750 career plate appearances (by comparison, Ozzie Albies has 2,440 already), he just completed his age 32 season. Given what we know about player aging, it wouldn’t be unusual for a player’s early 30s to signal his decline phase, and for that phase to feature an inability to catch up and damage fastballs, even fastballs from opposite-handed hurlers.
So, basically, at the time the Braves were in the market for outfield help, you had Adam Duvall, who:
- Had killed the Braves, but was otherwise hitting pretty poorly;
- Randomly had a reverse split, including one borne out by quality of contact measures; that in and of itself was the result of
- An inability to hit fastballs, especially from lefties.
The Braves acquired Duvall anyway on July 30. At the time they did, his platoon split was getting even more exaggerated: Duvall posted a .377 xwOBA/.413 wOBA against righties in July, but just a .228 xwOBA/.185 wOBA against lefties.
Do you know where this is going? You might think you do, but it’s a little weirder than that. After an aggregate 102 wRC+ with Miami, Duvall didn’t really reinvent his own particular wheel with his old mates, posting a 106 wRC+ as a Brave in 2021. His walk rate and strikeout rate between the two team stints are almost identical.
But the splits! Did you think this was going to be a redemption story, one of reversion to the norm? Well, it isn’t, not entirely. In a Braves uniform, Duvall’s 2021 featured an even more exaggerated reverse split — he was 67/114 as a Marlin in wRC+ terms against lefties and righties, respectively… and somehow, 58/121 as a Brave. What did revert itself was the inputs — but somehow Duvall found the motherlode of xwOBA underperformance against lefties in a Braves uniform. I’ll just throw the charts here.
And as for fastballs, well…
There’s a couple of stories here, but the most eye-popping one is how Duvall just absolutely destroyed four-seamers from righties after the trade. Duvall hit 16 bombs after the trade, of which 15 came off righties, and eight came off four-seamers — Duvall hit 50 percent of his post-trade homers on a pitch type he only saw a third of the time. There was a similarly gigantic leap forward in crushing lefty four-seamers, too, though not enough to reverse his season platoon split overall. So, if you were worried about Duvall’s fastball problems and resulting reverse split being an incipient sign of aging, don’t be, I guess. The other story is about how Duvall absolutely terrorized non-fastball stuff from lefties as a Brave, while making some real improvements against those sorts of offerings from righties as well. Beyond just obliterating fastballs, being able to contend with other pitches is probably the main way Duvall can still find another gear as a hitter. With (hopefully) his Marlins-based fastball issues behind him, I wonder whether gains against those pitches in the latter two months of 2021 are a sign of things to come.
Duvall collected 64 PAs throughout the Braves’ successful title run, but hit just .220/.266/.407. His overall playoff line was actually really similar to his 555 regular-season PAs: actually fewer strikeouts but more pop-ups, and the killer: a more “normal” HR/FB of around 13 percent instead of the 20-plus percent he posted in the 2019, 2020, and 2021 regular seasons. The xwOBA was similarly bad (.279 wOBA, .274 xwOBA). But, in these playoffs at least, Duvall did play to type: in 11 PAs against lefties, he hit three homers and added a walk and a hit-by-pitch, for an insane .756 wOBA and .656 xwOBA. Adam Duvall didn’t terrorize lefties in the regular season, but oh boy did he eviscerate them in October. He also crushed fastballs — .450 wOBA/.418 xwOBA against four-seamers, .378 wOBA / .357 xwOBA against fastballs of all stripes (though weirdly that’s mostly just propped up by the four-seamers), and a .178 wOBA / .189 xwOBA against everything else. His weird Marlins tenure aside, Duvall returned to himself as a Brave, and then became the ultimate version of that in the playoffs.
Will Duvall be back to rake some more in a Braves uniform in 2022? Who knows. If he isn’t, he gave the Braves another thrill ride — coming over in the midst of a strange, reverse-splits-and-fastball-problems tenure with the Marlins, rectifying it after changing uniforms, and then becoming an ultimate fastball-and-lefties destroyer as he helped the Braves win some rings and hoist Rob Manfred’s “hunk of metal.”