Evan Scribner and the Cruel Realities of Relief Pitching

It might seem easy to envy the life of a major league baseball player: even the least famous of them make a lot of money, have a bunch of Twitter followers, and get to play a game for a living. But it’s always been a little tougher for me to envy the life of a relief pitcher, who seems to experience the worst drawbacks of the job with little of the upside. As opposed to starting pitching — where one’s mistakes can be more easily made up for by logging good innings — relief pitching is simply less forgiving, with every mistake massively amplified. And no pitcher has embodied that more in the past two seasons than Evan Scribner.

You might know about Scribner, as he was just traded from the A’s to the Mariners last month. There’s more than that to him, of course, and one of the big things is the number of strikeouts he’s recorded compared to the number of walks he’s issued. That’s important, because we like the K-BB% stat a lot for pitchers: it’s been shown to be the best ultra-simple ERA estimator we currently have. It’s not perfect, but it can be a handy way to get an idea of how a pitcher could perform in the future. There’s even some work that shows it performed better than xFIP, FIP, and SIERA at predicting second-half ERA after a first half was in the books. All of this is just a set up, really, for this table, which shows the 15 best K-BB% marks for relievers with at least 70 innings pitched between 2014-2015:

The list is pretty much a who’s who of dominant relievers, as every guy in the top 10 is closing or has closed for their respective teams sometime during the past two years. Then there’s Scribner, who sticks out like a sore thumb mostly because of his ERA and the fact that he walks no one. Well, not no one, but in major league terms, he walks no one. He has the lowest walk rate among relievers who have pitched meaningful innings in the past two years (20+ innings), and it’s not particularly close; in raw numbers, he’s walked four batters in 71.2 innings.

We could take this article in a certain direction by looking at the batters Scribner has walked, and I might end up doing that the next time he actually does so. But what we should be more interested in, especially given his recent trade, is this:

This explains the ERA, of course, and is one of the reasons why Scribner was just traded for a minor leaguer who, as Jeff noted, has coincidentally never given up a home run in his professional career. The reductionist argument would state that there’s two ways of looking at Scribner: one sees a very good pitcher, whose xFIP was 2.89 over the past two years; the other sees a guy who is simply prone to home runs, and most likely will be in the future. It’s not that the A’s necessarily were entrenched in the latter viewpoint — players are traded for reasons other than being seen as hard to fix — but it was no doubt a part of the equation.

So what gives? Is this just bad luck? Or does Scribner groove a lot of pitches? Answering those questions definitively is always going to difficult, but we can take a few approaches to it.

First, let’s take a look at some of the home runs. Scribner gave up a whopping 14 of them in 60 innings, which is only one fewer than Jake Arrieta has given up in the past two years over 385.2 innings. On April 9th of this past season — in Scribner’s second appearance of the season — he came up against Adrian Beltre. This happened:

There aren’t many other ways to put it: this is an Adrian Beltre home run. And, as it turns out, it was the lowest pitch hit for a home run all year by a right-handed hitter. We could say that’s Scribner’s fault, or we could say it was Beltre’s ability. In baseball, it’s always somewhere in between. This time, however, it was way more toward Beltre.

Then, on April 15th, the A’s were in Houston, and Luis Valbuena stepped to the plate in the seventh inning. On a 1-1 count against Scribner, this happened:

You can’t see exactly where the ball lands, but it’s somewhere in the first few rows of the Crawford Boxes, and Mark Canha’s reaction in left field tells us the ball just got out. Unsuprisingly, if we look at opposite field home runs hit by left-handers this season, this is the fifth-shortest homer. Hit Tracker says this is a home run in zero parks, which seems like the sort of thing you tell a pitcher in the clubhouse after the game to try to make him feel better.

How about one more. Scribner went over a month until his next homer, against Steven Souza on May 22nd:

There’s no way to put positive spin on this pitch, as Scribner just missed his spot and grooved a 3-2 fastball. A lot of major league hitters are going to hit this out. If we want to be optimistic, we could say it was a line drive, and it barely left the park: on a better day, maybe this is a double, or a liner right at the left fielder. Scribner didn’t have a lot of better days this season. Case in point: this was one of the lowest home runs of the year, falling right around the 1st percentile for height among all home runs with an apex of just 49 feet.

We could continue with this exercise if we wanted to: every home run Scribner gave up between mid-June and early August (all six of them!) was under 400 feet, and none of them would have been home runs in all 30 parks. These aren’t the sort of things we want to be holding up as reasons to love what he was doing, but they are part of a larger case for why expecting this sort of home run/fly ball rate in the future is completely ridiculous.

Another angle we can take, following our home runs: does Scribner have a problem with pitching over the heart of the plate too much? Using Bill Petti’s Edge%, we can find out. I’ve taken all pitchers from 2015 who threw at least 500 pitches (there were 439 of them), and calculated Scribner’s overall rank and percentile in the group. Let’s take a look:

Scribner absolutely has a problem pitching over the heart of the plate. He posted the 32nd-highest rate of pitches over the heart (25.0%), putting him in the 93rd percentile of pitchers in our sample. This is certainly one way to not walk anyone, but it also has the unfortunate drawback of people hitting balls very far on regular occasion.

The final answer, as it most often does, lies somewhere in the middle here. Was Scribner unlucky with the types of home runs he gave up? Absolutely. Even if he continues pitching in this manner, we can reasonably expect for some of those balls to not go over the fence. Was the way Scribner pitched during 2015 a big contributing factor in him giving up that many home runs? Absolutely.

This doesn’t even begin to get into how “hittable” Scribner’s pitches are: how much movement they have on them, how they look out of the hand, etc. Those breakdowns are inherently difficult to do on a deep level. What we do know is that Scribner’s four-seam fastball features a higher-than-average swinging-strike rate, and so does his curveball. He seemingly has the tools to be successful, but as we’ve seen many times before, tools aren’t everything.

The Mariners know what they have in Evan Scribner. They know what we’ve seen here and almost certainly a lot more. The A’s do, too, for that matter. Scribner’s strikeout and walk rates point to a dominant reliever. His home run rates point a completely different way. Somewhere in the middle is a solid major league reliever. Because no matter how good your stuff is, sometimes Adrian Beltre is there to go down on one knee and hit it 400 feet. Baseball is cruel, and its crueler to no one more than relief pitchers.