clock menu more-arrow no yes

Filed under:

Welcome to the Twilight Ozuna

New, 92 comments

Even weirder than sparkly vampires.

Los Angeles Dodgers v St Louis Cardinals Photo by Dilip Vishwanat/Getty Images

Baseball players are, and have generally always been, weird. Not all of them, and perhaps not even many of them — like the humans that constitute generally every profession, most baseball players fall within the meaty part of the personality bell curve. But, there are always the characters. Dock Ellis threw a no-hitter while on LSD. Carl Everett had a well-known feud with the existence of dinosaurs. Zack Greinke lets a 30-cent increase in the price of Chipotle’s guacamole influence his fast food decision-making process. I’m sure you know all about Bill Lee and Manny Ramirez, and if you don’t, well, go rectify that. Hunter Pence exists. You get the idea. Running the gamut from eccentric (Mark Fidrych) to wacky (Turk Wendell), there hasn’t been any shortage of weirdness in baseball.

For some people, perhaps most people, this sort of personality-based flavor is baseball: humans, some kookier than others, playing what is frankly a pretty weird game, for the entertainment of millions of spectators and listeners. For others, though, there’s a bit more: the reams of assorted statistical outputs that plate appearances and games on end generate. If to you, like to me, baseball is really the inextricable linkage of both of these things, and you adore the idea of weirdness in baseball players and their stats, then boy, are you in luck. Because, you see, the Braves’ recent signing of Marcell Ozuna has added both in spades.

This isn’t an article to convince you that Marcell Ozuna is the weirdest baseball player of his generation. He probably isn’t. This isn’t an article that covers every single way in which he’s weird, because A) we’ve all got other stuff to do, and B) I’m not even sure that’s possible to achieve. Instead, this is a trek through his sublime ridiculousness, a journey to the Twilight Zone Ozuna. Doo-doo-DOO-doo doo-doo-DOO-doo. You get the idea. Onward!

Let’s get the obvious stuff out of the way first.

Look, this is a man that made over $12 million in salary the year when this incident occurred. I realize some of us would do far worse for far less, but still. Can anyone explain the process on display here? Anyone? Anyone? Rod Serling? Let’s just go with a momentary lapse during which Ozuna’s brain traveled across the cosmos, seeking the fifth dimension, before snapping back to realizing that he was supposed to be preventing runs by means of catching a baseball hit in his vicinity.

One day, I’ll break free of this accursed Euclidean geometry that contains me!

If you’ve watched baseball over the last few years, one thing you’re probably aware of, if only subconsciously, is that pitchers don’t really put it over the plate like they used to. In 2002, around 54 of every 100 pitches thrown were within (one definition of) the rulebook strike zone. That ratio never got that high again; by 2008 it was just over 50 out of every 100 pitches, and then it fell further still. Between 2011 and 2017, the ratio held steady at around 44-45 pitches out of every 100, and has since collapsed further: 43 per 100 in 2018, and under 42 per 100 in 2019. (Dear pace-of-play handwringers: maybe complain about this instead of everything else? A flip of 12 percent of pitches being optimal swings versus optimal takes definitely doesn’t help a fast-paced game.)

What does this have to do with Ozuna? Well, baseball itself is not the only thing experiencing a collapse in zone rate: over the past five seasons, dating back to 2015, Ozuna’s own personal zone percentage (i.e., the rate of pitches he receives in the zone) has also fallen each and every season. On the one hand, that isn’t super-weird in a vacuum. I ran the numbers, and 46 players (since 2002, when pitch location data started being tracked in earnest) met the same criteria of a year-over-year decline in zone rate over a five-year period or longer. Amazingly, Yadier Molina somehow had pitchers throw him fewer pitches in the zone in Year X+1 relative to Year X for nine straight seasons (2006-2014). But, not to worry, Ozuna’s (lack of well-placed) cheese stands alone here too in one gnarly respect: each of these 46 players, except Ozuna, started their zone rate collapse from a fairly high point. In other words, someone like Molina saw 58 percent of pitches to him in the zone at the start of his streak, providing a lofty pedestal from which to see fewer and fewer strikes. Ozuna, meanwhile, started his streak at an already-low 44 percent. The fewer strikes you see, the harder it should be, theoretically, for pitchers to throw you even fewer. Yet, here stands Ozuna, his zone rate tumbling from 44 percent to below 38 percent in six seasons, with a small consistent drop every year. Only Ryan “The Riot” Theriot and Albert Pujols have experienced similar drops in their consistency and lack of volatility, and both started from higher perches anyway. Can Ozuna make it six in a row? How low can his zone percentage go? More kinda-sorta-weirdness along these lines: Ozuna and Javier Baez are the only players with three consecutive finishes in the bottom 30 for zone rate in each of the past three seasons.

I have had nightmares about the uncanny valley monkey-masked (is it really a mask) man (?) in this gif since first seeing it. I apologize in advance.

Not weird enough for you? Well, okay then. Let’s try something else. Since Ozuna’s debut, he’s batted against relief pitchers 1,365 times, and compiled a .314 wOBA. His career wOBA, meanwhile, is .336. The league’s wOBA against relief pitchers since the start of 2013 is .309, so .314 doesn’t seem that weird... until you realize that the league’s wOBA since the start of 2013, in general, is .316. So, just to recap: since 2013, relievers have suppressed hitting to the tune of around .007; in Ozuna’s career, relievers have suppressed his hitting by .022, or over three times as much.

If that’s clear as mud, we can put it another way. There have been 142 players who have totaled 1,000 or more PAs against relievers since the start of the 2013 season. Of all of them, Ozuna is 22nd on the list with a +.035 wOBA against starters relative to relievers. (Hat tips are in order here to Nolan Arenado, with an unconscionable .403 wOBA versus starters and just a .325 against relievers, and Jed Lowrie, who’s hit starters at just a .310 clip but relievers at a .358 clip.) This puts Ozuna somewhere around the 15th percentile for “guys who really struggle against relievers (relatively),” or alternately, the 85th percentile for “guys who (relatively) mash starters.”

One semi-recent trend associated with increased reliever usage is that the gap between starters and relievers has narrowed. In 2013, when Ozuna was a rookie, the leaguewide wOBA against starters was .319, dropping to .303 against relievers. The gap stayed somewhere around that large until 2018, when for various reasons, it fell to around a fifth of that: .316 against starters, .313 against relievers. It was similar in 2019: .322 against starters, .318 against relievers. Ozuna, though, in his infinite weirdness, doesn’t seem to care about this trend:

The whole thing is just a mess. A baroque, meaningless mess, that simultaneously warrants vivid description and complete sensory deprivation. Abandon all hope of learning anything useful, all ye who enter here:

  • During his first two years in the league, Ozuna was a legit super-weirdo by hitting relievers better than starters.
  • The 2015-2018 period was marked by him being generally much-better-than-average against starters, and way-worse-than-average against relievers. The exception was his breakout 2017, where he just dominated opposing pitchers altogether, but still kept up that huge gap in terms of his versus-starter and versus-reliever performance.
  • But then you get to 2019, and the gap is totally gone.

What to make of this? I don’t think that’s even the right question. This seems far more likely to be devoid of meaning than the Rosetta Stone to Ozuna’s cryptic qualities. But still, it just kind of sits there. It’s unsettling. Who does that?

(By the way, just in case you’re wondering about other past-and-future Braves using this same metric, there’s not much to see. Pretty much the only consistent things I could find were Ender Inciarte hitting starters better than relievers, though the effect size is inconsistent year-to-year, and Tyler Flowers hitting relievers better than starters with a consistently large effect. The rest was a hodgepodge of small and/or erratic effects. Nick Markakis has generally hit starters better, but that flipped for him in 2019. Freddie Freeman consistently flip-flops which group he punishes more every single year.)

Don’t mind Marcell, nothing to see here...

The stuff that’s been mentioned to date is perhaps weird, and mostly inconsequential. A non-pattern pattern of hitting relievers? Eh, that doesn’t matter nearly as much as aggregate production. A constantly-declining zone rate? Sure, it’s interesting and kind of novel, but what matters is whether and how that trend affects his hitting, not the mystical ball-deflecting gizmos he has in his pockets. That time he climbed the wall and faceplanted? Silly and brutal, but it was just one play. So let’s get into weird stuff that matters.

One key item here, and it’s one that arguably could have started this article, is the extreme weirdness of Ozuna signing a one-year deal in the first place. For those that need a refresher on Ozuna’s top-level production:

  • 2017: 679 PAs, 143 wRC+, -4.3 Def, 5.0 fWAR, 6.1 rWAR
  • 2018: 628 PAs, 107 wRC+, -2.6 Def, 2.8 fWAR, 2.9 rWAR
  • 2019: 549 PAs, 110 wRC+, 0.0 Def, 2.6 fWAR, 2.2 rWAR
  • Career: 112 wRC+, 1.7 Def/600, 3.2 fWAR/600, 3.0 rWAR/600
  • Steamer, 2020: 2.6 fWAR in 634 PAs
  • ZiPS, 2020: 2.8 fWAR in ~500 PAs
  • Age: 29 for entirety of 2020 season

Take a step back, and look at those numbers. Ask yourself whether it makes sense for that player to sign a one-year deal. And yes, I know, I know, Ozuna bet on himself and signed a one-year deal despite longer offers. But still, isn’t that kind of weird? Ozuna wasn’t Bryce Harper or Manny Machado, but he was an above-average player hitting free agency before turning 30. Last offseason, Michael Brantley got a two-year deal with a little less AAV. Brantley had a better walk year, but was previously very injured. Andrew McCutchen got a three-year deal with a similar AAV, and is four years older than Ozuna, with a further-in-the-past dominant season. A.J. Pollock is three years older than Ozuna, has been consistently injured, was further removed from his breakout year, and still got a (really weird) multiyear deal from the Dodgers well above Ozuna’s financial guarantee. I’m not saying Ozuna was wrong to bet on himself — that’s his prerogative. It’s just odd to me that no team came calling and obviated that decision by giving him something more substantial. In a market where Nick Castellanos (younger but similar-to-worse production, no Qualifying Offer attached) is getting a guaranteed $64 million with a ton of opt-out flexibility, and Mike Moustakas (older, but similar to Castellanos otherwise) is getting a Castellanos-sized deal, something just feels off. That’s nothing new for Ozuna, though. It just adds to the Ozuna-ness of it all.

Things only get stranger when the peripheral stats, those things you expect teams-in-the-know to look at heavily when making decisions for the future, absolutely adore Ozuna. In 2019, Ozuna was in the 90th percentile or higher in exit velocity, hard-hit rate, and xwOBA. In 2018, he was no lower than the 84th percentile among those key measures. Same in 2017. Even in 2016, his lowest percentile measure among any of these three metrics was 67th. Since Statcast started measuring stuff (2015), Ozuna has never been below the 55th percentile in any quality of contact measure or xStat, and he’s only gotten better at all of these things as the seasons rolled by.

Some of you probably already know what’s coming, given that this is a (very long) discussion of Ozuna and his weirdness. In short: Marcell Ozuna is (freally frickin’) weird because he somehow, some way, underperforms his xwOBA, leading to far more pedestrian results than are arguably warranted. When Charlie Culberson goes all voodoo-magic-that-mocks-xwOBA-and-leverage-splits, it’s awesome and exciting. When Marcell Ozuna does the opposite, it’s really just kind of sad. Not Kendrys Morales sad, but sad nonetheless. Why bring up Kendrys Morales? Well, you’ll see:

The tables above convert wOBA into run value (based on the wOBA-wRAA relationship between 2015 and 2019), and estimate how much run value was actually lost by xwOBA underperformance. The tables are cumulative, such that the top table encompasses the greatest time period and the bottom table is 2019 alone. Since Statcast data have been available, Ozuna is a top-20 value loser due to underperforming his xwOBA. Over just 2019, and both of the last two years, no player has lost more offensive value due to underperforming his xwOBA. See, it’s sad. Or, if you’re Ozuna, maybe it’s more “bruh, really?” than sad. You’ll have to ask him.

I’m not going to re-litigate all the possible explanations for this sort of thing here. They’ve been covered. I don’t think they really clarify enough to hoist Ozuna out of weird-and-sad territory. Suffice to say, though, it’s been covered.

  • Here’s Ben Clemens/Fangraphs. He refers to the “slice” theory...
  • ...covered by Zach Gifford/Birds on the Black. Do you find this compelling? Or is it fitting an explanation onto something fairly noisy and prone to have something like this happen to someone in a two-year (or five-year, to a lesser extent) period? It’s notable, for example, that Ozuna didn’t show up in this examination of spin effects on (a lack of) fly ball distance though said examination is dated at this point.
  • And here’s Mike Petriello/MLB.com. There’s a lot of great stuff here, which indicates that unlike Culberson’s voodoo, the real story for Ozuna is that his grounders are really easy to defend, and drag down his batting line more than someone with an average assortment of grounders.

I’m not sure we should expect a silver-bullet explanation for Ozuna’s underperformance, as sometimes these things just happen. Even so, there not being one uniform reason for Ozuna being such an xwOBA underperformer is kind of weird in and of itself. This isn’t a Kendrys Morales situation, and it’s not developing into one... or is it? (Because if it is, then Ozuna and the Braves might have some problems, given that shifting and defensive positioning just destroy Morales’ very-hard-hit-but-mostly-on-the-ground profile.)

This is probably going to resolve itself in one of two ways. Either Ozuna will do something to stop hitting predictable grounders / stop slicing or top-spinning the ball so much / something else, and the xwOBA gap will disappear, and everyone will rejoice, and he’ll probably do a dance.

Or, it’ll persist, and the enigma of Ozuna’s underperformance will prompt further thinking and analysis, but it will continue to be sad for him and the Braves. Sure, there are probably other variants, but those don’t have much to do with baseball (e.g., he transforms into a yeti) and seem at least marginally less likely. Whether the crux of Ozuna’s potential offensive underachievement goes away or not is definitely one of the most interesting 2020 storylines for the Braves at this point.

We started this article with perhaps the most famous Ozuna gif to date: his failed audition for the Broadway version of Spider-Man (luckily he didn’t get hurt, which you can’t say for many of that production’s cast members). Yes, his decision-making was weird in that moment, but it’s far from the only puzzling aspect of his defense.

At this point, there are three commonly-used defensive metrics for outfielders: UZR, DRS, and OAA. UZR and DRS are pretty similar, and are expressed in runs above/below average. OAA is a Statcast measure that doesn’t quite work like UZR and DRS: 1) it’s not run-weighted, and is instead expressed on the basis of outs; 2) it’s based only on the catch component of outfield defense, whereas UZR and DRS include other components such as runs prevented via throwing the ball; and, 3) it’s measured based on calculations regarding starting distance, distance to the ball’s landing point, and hangtime, which means it’s more precise at determining how likely a given ball was to be caught, given the fielder’s starting position (UZR and DRS are much more approximate in this regard, given that they do not do these calculations).

UZR and DRS relate to one another pretty well. Here’s the plot of outfield DRS versus outfield UZR for every player-season with 1,000 or more innings, going back to 2015.

(Note: this analysis was initially done wrongly thinking that OAA was available from 2015-on, to work with one dataset across all three metrics. However, OAA is only available from 2016-on. While not shown, the 2016-2019 data for UZR vs. DRS feature essentially the same conclusion as shown above and discussed here; the correlation is actually even stronger.)

If you do just the range components of these metrics, the results are pretty much the same and the scatterplot looks basically identical to the one above. Over three-fourths of player seasons feature either an above-average UZR and DRS, or both below zero (below average). This same percentage holds when you assess just the range component. It would be fitting with the theme of this article that Ozuna would be a weirdo whose UZR and DRS didn’t match, but that’s not the case. In his five seasons, Ozuna’s UZR and DRS only really seem at odds in a few places and times. In 2015, he had -3 DRS and +1.1 UZR. In 2017, he had +8 rPM (the range component of DRS), and -1 RngR (the range component of UZR). Really, that’s about it. His 2017 is really the only season that stands out: a gap of seven runs between his DRS and UZR is one standard deviation from the mean difference (which is less than a run); a gap of eight runs between his rPM and RngR is around 1.2 standard deviations from the mean difference in range components (which is less than 1.5 runs). In every other season, the difference between Ozuna’s UZR and DRS has been five runs or fewer, so basically, nothing to see here, as Ozuna himself demonstrates in the gif below.

(In fact, this exercise used data from 104 outfielders. Of those 104, Ozuna is one of just six whose average difference over this period is basically zero runs. His average difference in range component alone is also on the lower end, though not quite that minuscule.)

But, this is really just two of three metrics, and the third one is really different from the first two. And, there’s a giant red flag that happens, pretty much right away:

  • In 2016, things seem aligned. Ozuna was a below-average outfielder, with particular issues in range. OAA was not quite as negative, but the general premise was consistent.
  • In 2017, Ozuna was an above-average outfielder, though UZR had his range as a slightly negative. OAA was generally positive, which is confusing in light of RngR but seems reasonable otherwise. (Remember that 2017 was Ozuna’s one UZR-DRS disagreement year.)
  • In 2018, Ozuna was again an above-average outfielder, with particularly heady range marks from DRS. OAA, though, had him as a marginally below-average defender. That’s puzzling.
  • In 2019, Ozuna remained a somewhat above-average outfielder, though he was closer to average by DRS (which had him as the better defender as recently as 2018). But, his OAA tumbled to a pretty low mark. In fact, only 11 outfielders posted worse OAA marks than Ozuna in 2019.

Well, I’m puzzled. How about you? The question to ask is whether this sort of UZR/DRS versus OAA disagreement is normal, or totally off the wall. Answering that helps us understand whether Ozuna’s defensive metrics are bonkers, or whether there’s just a lot of variation across UZR/DRS and OAA, even if UZR and DRS generally align. Before we do so, though, we should consider the ways in which we might expect UZR/DRS and OAA to differ.

One such way is through run-weighting. OAA doesn’t care if the ball is behind or in front of a fielder, as it treats all outs the same. To UZR and DRS, though, missing a ball that is more likely to result in extra bases is more harmful than letting a single fall in the shallow outfield. OAA is published on a broad directional basis, and we can see that it generally suggests he’s worse going back on the ball than coming in:

Given this, we’d expect (especially in 2018, where he was +2 in but -2 back) that Ozuna’s UZR/DRS should be worse than his OAA, since he performs worse on the more heavily-weighted plays. But, that’s not really the case: Ozuna’s RngR and rPM are generally far more positive than his OAA, and in 2018, where the difference between his going in and back was most stark, he posted a super-high rPM of +12. So, this neat and tidy explanation won’t do.

The other possible alternative is that it’s all about positioning. This is harder to examine, as the difference between OAA and RngR/rPM can really just be called “positioning” in a glib fashion. In other words, if a player has worse OAA than RngR/rPM, he may be positioned well to prevent runs in the UZR/DRS paradigm, but not actually particularly skilled as an outfielder. Conversely, if a player has better OAA than RngR/rPM, he may be letting balls drop in for hits and runs, but it’s not the fault of his skillset, but rather where he tends to start plays. Unfortunately, there’s not much to clue us in on this front, either. We know from Statcast that between 2018 and 2019, Ozuna’s “average” start position in left field was unchanged: 293 feet deep, -29 degrees radial angle, both seasons. Furthermore, we can see that between 2018 and 2019, the gap between his RngR/rPM and OAA may have widened (from a four-run to 13-run drop, to a nine-run to 13-run drop), without any obvious aggregate positioning change. I don’t think we have many good answers here; instead, it’s all just very “Marcell Ozuna is weird, y’all,” which is thematically appropriate for this article but unsatisfying. So, let’s go to the data...

We can convoke these data in a few ways. One such way is not to hive anything off year-to-year, but look at it over the entire 2016-2019 period. If we do this, we get something like this:

  • UZR versus OAA: R-squared of 0.38; out of 89 outfielders, 62 percent have UZR in the same direction as OAA; average difference (acknowledging that UZR is in runs and OAA is in outs) of 0.1. By this measure, Ozuna is an outlier: he has +10.6 UZR but -8 OAA; Kole Calhoun is the only outfielder with a higher UZR and lower OAA than Ozuna. Ozuna is weird.
  • DRS versus OAA: R-squared of 0.45; out of 89 outfielders, 71 percent have DRS in the same direction as OAA; average difference (acknowledging that DRS is in runs and OAA is in outs) of 0.9. By this measure, Ozuna is very much an outlier, as only 17 outfielders have positive DRS (Ozuna = +15) and negative OAA (Ozuna = -8), and only Jon Jay/Justin Upton have more exaggerated differences in this regard (DRS= +7/+8; OAA = -20).
  • RngR versus OAA: R-squared of 0.38; out of 89 outfielders, 67 percent have RngR in the same direction as OAA; average difference (acknowledging that RngR is in runs and OAA is in outs) of -1.5. Ozuna isn’t quite a huge outlier here — over 16 percent of the sample has positive RngR (Ozuna’s is +2.5) and negative OAA, and his gap isn’t really that big compared to guys like Justin Upton (+15 RngR, -20 OAA), Jon Jay (+6.4 RngR, -20 OAA), and Kyle Schwarber (+0.1 RngR, -26 OAA).
  • rPM versus OAA: R-squared of 0.40; out of 89 outfielders, 65 percent have rPM in the same direction as OAA; average difference (acknowledging that rPM is in runs and OAA is in outs) of -1.5. Ozuna’s rPM (+14) is much, much higher than his RngR, so he’s definitely pretty weird (along with, again, Justin Upton, who has +16 rPM and -20 OAA). Aside from Upton and Jay, you have to find someone with half as much rPM as Ozuna (Mike Trout, of all people) to find another player with a negative OAA.

What to make of this? Basically — Ozuna’s high level of disagreement between his UZR/DRS defensive metrics and his OAA is indeed fairly odd. It’s not the most extreme instance of disagreement we can find, though it is fairly notable in terms of players who are rated more highly by UZR/DRS than OAA. If you widen the scope to include those players who are rated far worse by UZR/DRS than OAA, his overall deviation doesn’t seem very notable. For example, Billy Hamilton has amassed +58 OAA since the statistic has been publicly available... compared to just +14.6 UZR, +13 DRS, +3.3 RngR, and amazingly, -2 rPM. Ender Inciarte is another player with a super-high OAA mark that dwarfs his awesome-but-still-not-nearly-as-high run-weighted metrics, though this may just be a measure of scale rather than an “issue.”

The other way to think through this potential discrepancy is not to aggregate across four years, but use the player-season as the analyte. Doing this gives us 171 outfielder player-seasons with their own corresponding UZR, DRS, RngR, rPM, and OAA to match up and consider.

Overall, you might expect that with smaller samples, the results would match up less well, but they really don’t. The r-squared values are all pretty much exactly where they were up above, right around 0.4 (i.e., about 40 percent of the variation in OAA can be explained by the variation in UZR, DRS, RngR, or rPM... or vice versa, if you prefer to think of it that way, since we’re not talking about causality here). The average difference between one of these measures and OAA is generally less one, which is kind of a neat idea as far as validating them. They’re all pretty consistently in the two-thirds range in terms of having the same sign (i.e., above-average in both a metric and OAA, or below-average in both). They correlate basically okay.

So where do Ozuna’s last four seasons fit into all this? I’m glad I asked. Ozuna’s 2016 and 2017, as you could probably tell by the table above, aren’t really notable in this regard. Ozuna’s 2018 and 2019, though? Oh boy.

In 2018 (refer to the table above if you need a refresher), Ozuna was very positive by DRS/rPM, slightly above average by UZR/RngR, and had -1 OAA. In 2019, Ozuna was marginally above average by DRS/rPM, reasonably above-average by UZR/RngR, and horrific by OAA with a -8 mark.

In terms of rPM, no player’s OAA deviated more, in terms of being worse than his rPM, than Ozuna’s 2018 than Yasiel Puig’s 2017 (+13 rPM, -1 OAA). Ozuna’s 2019 is similarly deviant. Only Justin Upton (2017, 2018) and Kevin Pillar (2016, 2017) had similar consecutive-large-discrepancies where OAA was much lower than rPM. In terms of RngR, Ozuna’s +4.6 compared to his -8 OAA in 2019 is the third-biggest such discrepancy (again, only looking at cases where a player’s OAA is below his RngR). If you average the discrepancy between RngR and OAA with the discrepancy between rPM and OAA, Ozuna has the third-most-deviant season (2019), and the 15th-most deviant season (2018) in terms of OAA being worse than these metrics. It’s not quite Leonys Martin’s 2016 (-9 rPM, -2.6 RngR, +13 OAA, RIP Seattle’s outfield positioning), but it’s pretty close.

Or, in other words, the above. The Ozuna seasons are the red dots. Two are pretty normal! 2018 definitely bucks the trend, but it’s not that weird. That’s what 2019 is for. I look forward to having uncomfortably caveated discussions about Ozuna’s defensive prowess for the Braves in 2020 — the partisan debates between “his UZR/DRS are propping up his value even though he’s not actually a good defensive outfielder” could get heated. Or, what if it flips, and he suddenly underperforms his OAA? That’d be even worse! Of course, he could just be normal and save us all the headache, but that wouldn’t be very Marcell Ozuna, would it?

In 2020, Marcell Ozuna will probably do Marcell Ozuna things, like the ones in this gif. The rest of us, meanwhile, will be like Ozzie Albies: stunned, bewildered bystanders. A potentially-above average player with this much potential for statistical shenanigans? A part of me can’t wait. A part of me can’t sleep. If you die in Marcell Ozuna’s dreamscape, you die in real life. Or at least develop a headache. You’ve read nearly 5,000 words about Ozuna at this point. Go take a nap.