OAA Has Come for Emmanuel Clase

It’s rough being a reliever. Your whole career is a small sample. Emmanuel Clase has been one of the best pitchers in baseball since pretty much the moment he set foot on a major league mound in 2019, but over his entire career, he’s thrown just 338 innings. Our leaderboard says that total has been bested in 534 different player-seasons. That’s 534 times that one single player in one single season threw more innings than Clase has over his whole career. Clase set a personal best by throwing 74 1/3 innings last season, and on the individual season leaderboard, that total put him in a 79-way tie for 20,484th place. A small sample size means high variance. Over his entire career, Clase has never finished a season with an xFIP below 2.18 or above 3.42, which is pretty stable for a reliever. But after running a microscopic 0.61 ERA last season, his ERA is currently a so-big-you-can-see-it-from-space 5.51. Five-run swings are decidedly less stable.

When things go wrong to this degree, it’s usually because a combination of factors have conspired to make it happen. When you’re as good a pitcher as Clase, it takes both luck and skill to get results this bad. Our focus today will be on the extraneous factors. You know what else is subject to wild variations in short samples? Defense. And defense is letting Clase down in a big way. We’re here today because Mike Petriello asked me to look into something. Petriello is Major League Baseball’s Director of Stats and Research, and it’s my understanding that as such, I am legally required to investigate any statistical anomalies he assigns me. Here’s what he sent my way:

When Clase was on the mound in 2024, the Guardians racked up 5 Outs Above Average. They were great defensively. This season, even though he’s only pitched roughly one-fifth of the innings he did last year, Guardians fielders are already all the way down at -4. That’s an absurdly big swing. How is that even possible? Is it just luck?

That’s why we’re here, but let’s start at the beginning because there is real stuff going on under the hood. As has often been the case, Clase’s velocity was down a bit in the first month of the season. What’s new is that his location has also changed significantly. As Lance Brozdowski noted a few weeks back, he’s not busting righties up and in with his cutter like he did last season.

Clase is also catching a lot more of the plate with his slider, rather than burying it off the corner like he did in 2024. As a result, he’s pretty much only hitting the middle of the plate and his glove side. Last season, he owned the inside corner, too.

All of a sudden, batters have a lot less plate to cover, and that makes their job much easier. (A quick note: This is 100% speculation, and I don’t think it’s the case, but I should mention that Clase hit Jonathan India in the head with a pitch during the first weekend of the season. It’s at least worth considering whether that has made him slightly hesitant to bust hitters inside the way he once did.)

Clase’s cutter has always been great at both earning whiffs and sawing off batters. This season, he’s getting more whiffs than ever, but he’s also throwing fewer called strikes because batters are swinging more often. His strikeout rate is down and his walk rate is up. When batters do make contact, they’re hitting fewer groundballs and pulling more fly balls. Clase’s launch angle, hard-hit rate, and barrel rate have all increased from last year.

Those are real issues that need to get ironed out. But there’s plenty of goofy stuff too, and that’s why we’re here. First, there’s batted ball luck. Among pitchers who have thrown at least 10 innings, Clase’s .442 BABIP is the third highest in baseball. That brings us to our next issue. Clase was lucky last season, and he’s not so lucky now. Here’s how he’s fared on balls in play according to Statcast’s expected stats.

Last season, when batters put the ball in play off Clase, he allowed an actual batting average 80 points lower than his expected batting average. Sometimes fortune just smiles upon you.

This season, thanks to the harder contact and higher launch angles, Clase’s expected batting average is 54 points higher than it was in 2024. His actual batting average is 137 points higher than that expected batting average, for a total swing of 171 points. To be fair, because Clase is allowing more pulled balls in the air and expected stats ignore spray angle, we should probably be expecting him to underperform them by a bit. But 137 points is not a bit. Any way you slice it, the swing from -80 to positive 137 is an awfully big one, and that brings us back to the defense behind Clase.

We have to start with the fact that the Guardians are just plain worse on defense this season. They traded away all-world defender Andrés Giménez. Center fielder Lane Thomas is hurt. Shortstop Brayan Rocchio and third baseman José Ramírez have both seen their defensive metrics plummet. In all, the Guardians ranked second according to DRS and fourth according to Statcast’s Fielding Run Value (FRV) last season. So far this season, they rank 13th and 14th, respectively, in those two metrics. But strange as it sounds, defense is about more than just getting to the ball, and as Mark Simon just noted for Sports Info Solutions, the Guardians are quite good in the field at all the other stuff. Still, Petriello’s question specifically related to OAA, so that’s what we’ll look at here. And according to OAA, which just measures range, the Guardians have fallen from ninth to 29th. They’re at -10 OAA.

That’s right, although Clase has pitched just 1% of the team’s total of 1,428 defensive innings, the -4 OAA the team has incurred while he’s on the mound make up 40% of the team’s overall mark. That’s -4 outs in just 16 1/3 innings!

To figure out what’s going on, I watched all 56 of the batted balls Clase has allowed this season. I witnessed a smorgasbord of the things that can go wrong when a bat meets a ball. There was luck, there was skill, there was something for everyone. First, we’ve got the errors.

After the errors came the plays that probably should have been scored as errors, but instead went as hits.

Next on the totem pole, we’ve got the balls that just sneaked through the infield and the balls that just fell in front of outfielders. They’re not errors. They’re not even plays that should be errors. They’re just balls where a half a step or so would have made all the difference. These are the plays that separate a good defense from a bad one, and so far, almost none of them have gone Clase’s way. That’s what happens when your team OAA drops to 29th place.

In those three clip packages, I showed you nine different plays. That right there is almost enough to represent Clase’s entire nine-run swing in OAA. However, we’re not done just yet. I have purposely held back one key factor. That factor is Clase himself, and it’s a really important one. You see, Clase induces tons of comebackers. Look at this spray chart. Everything is right back up the middle.

Over the course of his career, 5.8% of the batted balls Clase has allowed have been hit right back to him. Among players with at least 500 batted balls in the Statcast era, that puts him in the 89th percentile. If you look at the percentage of balls that went to either the pitcher, shortstop, or second baseman, he’s at 41.1%, fourth highest out of the 725 pitchers in the sample. When Clase is on the mound, you should expect batters to hit the ball up the middle.

In large part, that’s because Clase’s cutter is an extreme pitch that gets extreme results. To show you just how specifically weird the contact he induces can be, here are five clips from 2023. They’re all Baltimore chops that took huge bounces off home plate. That happened to Clase five different times that season. More than a quarter of the balls hit to him were Baltimore chops! Who does that?

This season, all those comebackers have cost both Clase and the Guardians. Clase is having an abysmal year with the glove. Seriously, picture an abyss. Now picture Clase all the way at the bottom, trying to field comebackers down there. Did it look like this?

This was ruled a throwing error, but I think reasonable people can agree that there actually should’ve been two errors on the play: the first when Clase booted the ball, which gave former teammate Andrés Giménez first base, and the second when Clase threw it away, which gave Giménez second. Either way, the play cost the Guardians two bases.

It’s not just the errors, though. Clase can’t seem to field the ball cleanly. A surprisingly high number of baseballs have bounced off either him or his glove this season. All four of the balls below are from this season, and only two of them ended up as outs.

In other words, if you hit a comebacker at Clase, this year, you’ve got a 60% chance of reaching base and a 20% chance of ending up on second. Only one of those balls was ruled an error, but the other two definitely cost the Guardians outs. Clase has a career mark of -6 DRS. Since 2019, if you look at every pitcher who’s thrown at least 100 innings, that puts him in just the 12th percentile on a per-inning basis. This season, he’s only been docked -1 DRS, but he’s definitely struggling more than he has in previous years. In 2023, Clase made one error covering first, but batters went 3-for-19 when hitting the ball to him. The three hits were those three Baltimore chops that bounced so high that he never had a chance. In 2024, Clase made another error covering first and one regular fielding error, but he retired five of the six batters who hit the ball to him. This season really is different.

I definitely don’t mean to lay all the blame at Clase’s feet. First of all, fielding the ball as a pitcher is extraordinarily hard. You’re focused on making your pitch, and you do not have much reaction time at all. Second, his miscues aren’t nearly enough to make up the entire nine-out swing. Third, there’s still a big element of luck at play here. Last year, the Cleveland defense put up -8 OAA while Tanner Bibee was on the mound, but this year, Guardians fielders have been right at the league average when he pitches. That’s an eight-out swing, nearly as big as Clase’s, but in the other direction (and in considerably more innings). That kind of randomness happens, and right now it’s happening specifically to Clase.

The Guardians have made errors behind him and missed gettable balls. Moreover, the loss of Giménez hurts Clase more than any other pitcher. Over the course of his career, 18.1% of his balls in play have been hit at the second baseman, the second-highest rate in the Statcast era, and he just lost the best defensive second baseman in all of baseball. The one thing you didn’t see in all the clips I’ve shown you was a spectacular play, a diving stop, a tough play in the hole. I was really looking for one, but it just hasn’t happened yet. I would guess that the Guardians will clean up their defense some, if for no other reason than that Rocchio and Ramírez are due for positive regression. Still, if Clase wants better defensive play while he’s on the mound, he might have to figure out how to do it himself.