About a year ago, I wrote about the ridiculous, in-progress occurrence that was Charlie Culberson’s 2018 season (see here: https://www.talkingchop.com/2018/8/27/17783348/charlie-culberson-mockery-xwoba-statcast-atlanta-braves). At the time that post went up, Culberson was, by comparison of his wOBA and xwOBA marks, the luckiest player in baseball. Through 250 PAs, he managed a very good .359 wOBA on the basis of a quite bad .282 xwOBA. Over the season’s final month, his xwOBA and wOBA both fell, and the gap got a little smaller — he had an .077 gap in late August, and ended at .065. Not that it mattered, though: the .065 mark was still the luckiest such occurrence in baseball last year; the only players with greater positive gaps between their wOBAs and xwOBAs all had fewer than 80 plate appearances all season. In addition to that, there was also the other mind-boggling aspect to Culberson’s season. I’ll just quote myself to save time:
In short: no player with a substantial xwOBA/wOBA gap in a season’s first half saw that gap increase in the second half… except 2018 Charlie Culberson. I also went back to examined every player with a first-half gap of .051 or more. None of them increased their gap in the second half; only one (Eddie Rosario, 2015) matched his gap. The rest saw declines, usually massive ones. Charlie Culberson is the only player in the Statcast era to increase his substantially large first-half xwOBA-wOBA gap in the season’s second half.
Not only was Culberson a luck wizard, he was a relatively consistent one, at least as far as 2018 was concerned. Fast forward to 2019, and...
Charlie Culberson is doing it again
Okay, so he’s not quite doing it again to the same extent, but he’s still doing it. Through August 16, Culberson had amassed a .346 wOBA on the back of a .299 xwOBA. That’s not the biggest gap in baseball, like it was last year, but it’s still a pretty egregious .047. Among the 411 players with 100 or more PAs in 2019, he’s still in the top 20, i.e., above the 95th percentile in wOBA-xwOBA gaps. (Note: in the games since August 16, the gap has fallen a bit more, down to the low .040s.)
If you take all the players with 200 PAs in 2018 and 100 PAs so far in 2019, only six of the 296 are in the 90th percentile or above in wOBA-xwOBA gaps in both years. Only Culberson has been above the 95th percentile in both years. While it’s not entirely clear how he might end this season, and how much further playing time he’ll get given the acquisition of Adeiny Hechavarria to tide the Braves over in Dansby Swanson’s absence, Culberson can, at least for now, add another probability-defying merit badge to his magical cloak: he did something incredibly unlikely in two halves of the same season, and now he’s doing another pretty wild thing in two consecutive seasons.
For those of you that want to look at this and go, “big deal, I bet xwOBA outperformers do so in subsequent years, so what?” well, don’t say that! The below is a chart of 2018 xwOBA outperformance on a percentile basis (i.e., 0.5 = average, or basically no outperformance) and the same for 2019. Maybe you have the ability and/or keen eyesight to discern a pattern here, but I definitely don’t. (The correlation coefficient is less than 0.1.) The red dot is Charlie Culberson. He’s the most extreme. You probably figured that already.
There are many ways to quantify that Culberson is indeed “doing it again.” Here are a few that I like:
- Let’s leave Culberson aside and take the other big xwOBA outperformers in 2018, and consider their average xwOBA outperformance, again on a percentile basis (0.5 = no outperformance; less than 0.5 = underperformance; more than 0.5 = outperformance). The top five are, in aggregate, 75th percentile outperformers. The top 10? Only 53rd percentile! The top 20? 55th percentile. For every Tony Kemp and Nolan Arenado (90th-plus percentile outperformers in both years), there are Austin Hedgeses, Joey Wendles, and Harrison Baders (all 90th-plus percentile outperformers in 2018 and below the 15th percentile in 2019) aplenty.
- Here are the leaderboards for consecutive-season xwOBA-wOBA gaps, minimum gap .050, minimum 300 PAs and playing time in both seasons, in the Statcast era: (1) Charlie Culberson, .059, 2017-8 (2) Charlie Culberson, .058, 2018-9. That’s it. There are some other guys around .040, but no one crests .050, except Culberson.
- Here’s a listing of notable 2018 xwOBA overperformers and their fates in 2019, aside from Culberson: Yoenis Cespedes — has not played in 2019; Steven Duggar — 87 wRC+ has cratered to 62 since he’s now undershooting his xwOBA by .010, optioned to the minors, recalled, then injured; Ryan O’Hearn — from 153 wRC+ to 47, spent six weeks in the minors; Alen Hanson — from 86 wRC+ to 0, traded, DFAed, outrighted, released; Tyler White — 144 wRC+ to 71, DFAed, traded; Scooter Gennett — 125 wRC+ to 58, traded. Note that these guys aren’t cherry-picked, this is literally just the 2018 xwOBA overperformance leaderboard. And then you get to Charlie Culberson: 108 wRC+ to 102.
In short, he’s doing it again. Maybe not quite as much, and maybe it’s no longer quite as novel, but it’s still happening.
Charlie Culberson isn’t doing it again
But wait, that’s not the entire Legend of Charlie Culberson, is it? Of course not. After all, he didn’t get the Charlie Clutch moniker because of his ability to defy expected ball-in-play outcomes. We got the Charlie Clutch moniker because of this:
A leverage index of 3.00 or above is basically “the game on the line.” My rough estimate is that around five percent of game situations have anything like this, and even fewer have some of the 4s and 5s in the snip above. In those situations last year, Culberson failed three times and came through five times. In a game where failure is generally the mode outcome, that’s crazy-exciting. (It’s still crazy-exciting even if you consider that in two of them, he was more of a passive participant. A fifty percent success rate is still phenomenal.)
There’s also this:
And really, when it’s all said and done, this:
Yes, I will take a .350/.458/.750 slash line in high-leverage situations, thank you very much, 24 PAs representing the total sample or not.
Fangraphs has a “clutch” statistic which compares how well a player does in higher-leverage situations versus lower-leverage situations. Clutch above zero means the player performed better in the former than the latter; Clutch below zero means the opposite. Clutch isn’t predictive in any way, but it’s a fun descriptive stat that tells you about the impact (positive or negative) the player’s at-the-plate outcomes had depending on the situation in which they occurred.
Culberson wasn’t the most clutch player in 2018 (hello, Gleyber Torres, crazy 2.28 seasonal Clutch score), but he was up there. He finished in the top 20 with a 1.09 mark, a 93rd percentile outcome among players with 300 or more PAs. It was by far the highest mark among his teammates. Fast forward to 2019, though, and...
Okay, but those are just a few situations. Plus, he had a couple of walks in there. What about the situational splits?
Hmm. But surely the leverage splits will show his mojo lives on, right?
Or not, I guess. Sure, it’s just 14 PAs, but what a bummer.
On the season, Culberson’s Clutch metric for 2019 is -1.14, basically a plain reversal of his 2018 success in this regard. His biggest WPA-garnering offensive play of the year was a double in a three-run game on April 23. It brought the Braves within two, but the Braves wouldn’t do anything else in the inning and would end up losing by one. Aside from his two high-leverage walks, the highest leverage index for any hit he’s had this season is just 1.88 (an RBI single in the eighth with two outs, down by three); the Braves ended up dropping that game to the Royals by a run as well.
Of course, Culberson did make that awesome, life-saving throw from left field to stymie the Marlins, which revived the Charlie Clutch banner for a moment. But offensively, in a complete turnaround from 2018, he’s been anything but.
What Charlie Culberson has (and hasn’t done)
Leaving all the game-state and situational stuff aside, Culberson is actually a pretty fascinating evolution in his own right, when comparing his 2018 and 2019 statistics. Despite him still being an xwOBA-mocking wizard, he’s made real strides offensively. He’s raised his average launch angle three degrees, moving from a grounder-oriented hitter to one that’s more generic in terms of his batted ball profile. He’s raised his exit velocity over four miles per hour, going from weak contact in aggregate (430th out of 480 players with 50-plus balls in play in 2018) to average oomph in this respect as well (229th out of 436 players with 50-plus balls in play in 2019). His xwOBA has improved by nearly .030, though it’s still well below average despite the much better quality of contact because of his problematic plate discipline: Culberson walks way less than most hitters, and strikes out way more — the latter of which puts a damper on his mystic pseudo-ability of getting balls to drop in more frequently than warranted.
His gains have come across the board, as he’s improved his xwOBA against fastballs, breaking pitches, and offspeed pitches in unison. He’s hitting breaking balls really hard (90-plus mph average exit velocity, 45th among the 400 or so players that have put 20 of them in play so far this season), but he still whiffs so much on them (40 percent of the time) that it’s been a piecemeal improvement, though an improvement nonetheless.
The two tables below give you a sense of the mostly-across-the-board improvement between 2018 and 2019 for Culberson that has resulted in his xwOBA gains.
He’s making more contact in the zone, transforming grounders into liners, and barreling pitches more often. He seemingly had a skill at thwacking pitches over the left-field boards last year, but has actually gotten better at pulling the ball so far this year. There’s a lot to like, relative to the 2018 version of Culberson... at least until you get to the chase rate and the egregious whiff rate when he chooses to swing outside the zone (which is all the time). For some context, Culberson’s chase rate is not too different from Austin Riley’s (they’re 24th and 30th, respectively, on Fangraphs’ leaderboard for o-swing rates among players with 100 or more PAs in 2019), and his o-contact rate just barely avoids falling in the bottom 50 (using the same set of 400 or so players). That’s clearly the next step for Culberson as a developing hitter, and while the main narratives surrounding him might be about his xwOBA defiance and whether or not he’s actually “clutch” (he was, then he wasn’t, same old story), his work is actually cut out for him pretty well: maintain/improve the quality of contact from this year while reining in his swings and whiffs. Or just keep clutching whatever ancient amulet he found that bestowed upon him his mythical 2018 season, and hope it does the trick once again.