Many of the Ways that Tyler Clippard Is Unusual

I’m going to let you in on a little secret that might not actually really be much of a secret. The most difficult part of this job isn’t the writing or the analysis. At least, as far as I’m concerned, the most difficult part of this job is finding ideas, and finding them consistently. Once you have an idea, everything else can follow, but the thing about ideas is you’d like them to be original and, if you’re lucky, good. And interesting! Interesting is a big one. Maybe interesting and good ought to be categorized together.

For a while, I’ve personally been interested in Tyler Clippard. I’ve considered on several occasions writing about him, and about him specifically, but on every one of those occasions, I’ve talked myself out of it, because it just never seemed relevant enough. Generally, people haven’t woken up and thought, today I’d like to read in depth about Tyler Clippard. So I’ve had this idea on the back-burner for ages. But now? Now is the time to strike, since Clippard just got dealt from the Nationals to the A’s for Yunel Escobar. Tyler Clippard, to me, has always been interesting, but now he’s both interesting and topical, so, here goes nothing. Let me try to explain to you why Clippard is such a weird reliever.

(1) High fastballs. You already know about the goggles. You probably already know about the throwing motion, to some extent. Hell, if you’ve been watching Clippard closely, you even already know about the high fastballs. But maybe you don’t appreciate the full magnitude of his tendency. Let’s create a line, two and a half feet above the front of home plate. This represents, basically, the vertical middle of the strike zone, and now let’s define high fastballs as any fastballs at least two and a half feet up. So, last season: which pitchers threw the highest rates of their fastballs up? So, this is high fastballs / fastballs, not high fastballs / all pitches. With help from Baseball Savant:

1. Tyler Clippard, 75.7%
2. Michael Kohn, 70.9%
3. Jake Odorizzi, 70.8%
4. Madison Bumgarner, 69.5%
5. Todd Redmond, 68.8%

It’s Clippard by a relative landslide. Clippard threw three-quarters of his fastballs above the middle of the strike zone, and second place was five percentage points behind. A heat map of said fastballs:

Of course, this isn’t just a one-year thing. Clippard didn’t only recently decide to start climbing the ladder. Over the whole PITCHf/x era — setting a minimum of 500 fastballs — Clippard shows up with the highest high fastball rate. That leaderboard:

1. Tyler Clippard, 73.7%
2. Joaquin Benoit, 71.4%
3. Daniel Bard, 70.9%
4. Jake Odorizzi, 70.6%
5. Heath Bell, 70.0%

Clippard throws almost all his fastballs upstairs. When a fastball isn’t located upstairs, my guess is it’s probably a mistake. As a consequence of all those high fastballs, Clippard generates —

(2) Tons of fly balls. This is one of those baseball fundamentals: the higher a pitch is, the more likely it is to be hit in the air, should contact be made. Clippard hasn’t thrown exclusively fastballs, but he’s worked off of the pitch, and, well, we’ve got ball-in-play data going back to 2002. Since 2002, Clippard has posted baseball’s second-highest fly-ball rate, at 56%. The only guy with a higher number: Troy Percival, at 57%. Clippard generates fly balls like Chris Young, which is to say, he generates fly balls to an extreme degree. And Clippard doesn’t just give up fly balls that reach the outfield. He also generates —

(3) Tons of pop-ups. During the summer, Chris Young talked to Eno about the merits of pitching up in the zone. While pitchers have been cautioned against being too aggressive up, because of the dinger risk, those are also difficult pitches to catch up to, and they frequently generate poor contact if thrown just right. We here at FanGraphs use the Baseball Info Solutions definition of an infield fly, which requires that a ball end up within 140 feet of home plate. Since 2002, 861 different pitchers have allowed at least 500 balls in play. Clippard has posted the very highest rate of infield flies per ball in play, at 9.0%. The average has been under 4%. And, as you know, Clippard has done that mostly as a National. Now consider that he’s going to Oakland. From Clem’s, a ballpark comparison:

More space there for easy outs, instead of foul-ball strikes. And in our 13 years of data, the highest single-season infield-fly rate is 11.5%, posted by Louis Coleman in 2011. Clippard, that same year, finished at 11.4%. Perhaps this season we’ll see a new high. The ballpark certainly ought to help.

Combine regular fly balls with infield fly balls and you get —

(4) Hit prevention. Fly balls go for hits less often than groundballs or line drives. Infield fly balls go for hits virtually never. So, let’s go all the way back to 1969, when the mound was lowered. We’ll set a minimum of 400 innings thrown. Here are the lowest BABIPs:

1. Troy Percival, .230
2. Tyler Clippard, .236
3. Catfish Hunter, .239
4. Andy Messersmith, .241
5. Roger Nelson, .245

You know what happens when you run a super low BABIP? You out-perform your peripherals-based run estimators. Troy Percival’s career ERA- beat his career FIP- by 17 points. Clippard, so far in his career, has a difference of 20 points. And his FIP- is another seven points lower than his xFIP-, because he also has managed to limit home runs. Clippard has a career xFIP- of 99. More or less league-average. Not special. His career ERA- is 72. Randy Johnson’s career ERA- is 75. WAR doesn’t give Clippard proper credit, because his peripherals don’t totally capture what he really is.

And this is separate from the previous points, but —

(5) New splitter. Clippard’s always had the fastball and a good changeup. He’s mixed in a curveball in the mid-70s. He’s long thrown a splitter on the side, but it wasn’t until late in 2013 that he decided to bring it into games. Last year, he threw it about 10% of the time, subtracting from his fastball usage. As such, last year, Clippard was the only pitcher in major-league baseball to throw at least 10% splitters and 10% changeups. It makes for an unusual repertoire, but Clippard likes the different look that the splitter provides. The fastball’s always up. The changeup doesn’t have a lot of sink. The splitter? It sinks. It’s a putaway pitch, for Clippard to use when he wants a whiff or a grounder. 37% of swings against the splitter last year missed. The changeup’s rate was 39%.

And while Clippard threw the splitter about 10% of the time overall, he threw it 28% of the time with two strikes. He threw it almost exclusively with two strikes. For the most part, it worked, when thrown well:

And, for the most part, it worked, when thrown less well:

Clippard now is a reliever with four pitches. When he had a previous rough stretch, he blamed it on losing his curveball. Figured he couldn’t get by on fastball and changeup alone, and the curveball is a tricky pitch to have control of consistently. He thinks the splitter is easier, with a bigger margin of error, and by diversifying his repertoire, Clippard might avoid going the way of Ernesto Frieri, a once-effective high fastball thrower who’s now coming off an ERA over 7. Frieri’s had the strikeouts, but he hasn’t been as good at limiting quality contact. It’s one of Clippard’s strengths, and the splitter should help every pitch that he throws.

All together, Tyler Clippard is one of the more unusual pitchers in the game today. He’s certainly extreme, in a variety of ways, and now that he’s going to Oakland, his pop-up tendency should only get a boost. There’s nothing at all wrong with the move the Nationals made. Clippard’s entering his walk year, and the team needed some middle-infield depth. But now the Nationals will be exactly this much less interesting. That doesn’t matter even a little bit to them, but they and I have different perspectives, out of necessity.