Why is Luke Weaver so Effective?

(Author’s note: This is my first piece for VEB. It’s truly an honor to write here. As a lifelong Cardinals fan, VEB immediately became my go-to daily baseball reading when I discovered sabermetrics, and I’ve always admired the writers as well as the community around them. I hope I’ll live up to these standards, and that people will enjoy reading my writing here as much as I’ve enjoyed the great authors this site has featured over the years.)

One of the great joys of baseball is watching a dominant starter. They’re almost like superheroes. Carlos Martinez is a blur of arms and legs who throws every pitch ever invented. Justin Verlander looks like a movie star and throws a million miles an hour. Noah Syndergaard is nicknamed Thor! It doesn’t leave much to the imagination. Watching Luke Weaver pitch is… not this. I’m not saying he looks awful up there. He just looks very plain. He’s medium height, right-handed, doesn’t throw THAT hard. He doesn’t have an interesting delivery. He doesn’t even really have interesting hair- that’s the very definition of replacement level goatee. If there was a casting call for ‘generic major league journeyman,’ Luke Weaver would certainly look the part.

The results, though- the results are basically the opposite of the visuals. It’s easy to get hung up on the results of his last three outings- that Mets appearance in particular is the kind of game that leaves a bad impression in the brain. The fact of the matter is that even with that recent bump in the road, his pitching line remains exceptional. I wouldn’t go so far as to call him elite, but over the last two years, he’s been 20% better than league average by FIP-, on par with the 2017 seasons of Michael Fulmer, Jacob deGrom, and Chris Archer. If K% is your preferred metric of dominance, he’s right there with 2017 Verlander, Carlos, and Jon Lester. That’s impressive company. So what gives? All these guys look the part of aces, and Luke Weaver is just quietly putting up the same stats while looking like Jeremy Hellickson. I endeavored to find out.

To start, I built a composite of major league pitchers who resemble Weaver. I started with three central facts about Weaver- he’s a 6’2” righty, he throws a 93.3mph-average fastball, and he is a fastball/changeup/curve guy who throws a good mix of the three with a lot of changeups. To create a list of comparable pitchers, I looked through a list of starters who threw at least 60 innings in 2017 (more methodology notes at the end). I looked for righties between 6’1” and 6’3” with velocity within 1.5mph on either side of Weaver. For pitch mix, I took a more qualitative approach. I looked for people who threw 10% or more changeups(or splitters), a breaking pitch (or two breaking pitches) slower than their offspeed pitch, and a good amount of fastballs. This is admittedly kind of a rough cut, but I ended up with 15 names I feel okay about.

It is, to put it lightly, a pretty varied list. The best player on it is undoubtedly Max Scherzer, and that’s a pretty great name to see on a list of comparables. The bottom of the list is probably Jesse Chavez, but it might also be Nick Martinez or Vince Velasquez. The point is, the bottom of the list is not great. To get an idea for what this pitcher type is like, I built a composite player by weighting each pitcher by their innings pitched. The tables below compare them to 2017-2018 Luke Weaver. I’m excluding 2016 because his stuff was significantly different then- I want to look at pitchers who look like what Weaver is now, not what Weaver was then.

Some batted-ball luck aside (with only 90 IP in the last two years, I’m okay ignoring that for now), Weaver outperforms the blend pretty handily. Put simply, just a generic guy with stuff and stature like his wouldn’t get up to his level. Fine then, that makes sense. If Weaver were performing at the level of the pitchers who most resemble him- slightly better than league average run prevention with a league-average strikeout and walk rate- he’d be a useful player, but much less interesting. Let’s find some things that distinguish him from the pack.

Release Point

I’ll put the conclusion at the start of this section- Weaver is nothing short of amazing when it comes to disguising release points. Here are the horizontal and vertical release locations of every pitch Weaver threw in 2017 (all release point graphs provided by Fangraphs):

Here’s Scherzer:

For an idea of how bad this can be, here’s Velasquez:

Weaver releases all his pitches in the same one-foot-square box, more deceptive even than Scherzer. The guy he most resembles in this regard is Aaron Nola, pretty elite company to be in. I couldn’t find a good way to quantify release point spread, but suffice it to say that pretty much every one of the effective pitchers in Weaver’s cohort is exceptional at disguising their stuff. If you’re looking for a reason someone might have pedestrian stuff that plays up, this is probably the first place to look. Hitting is all a game of recognizing differences, and making the hitter wait a split second longer to tell the difference between pitches can make the difference between a hittable 93 and a pitch that produces late swings and whiffs.

Changeup Command

Here’s where Weaver throws changeups to lefties:

Put simply, that’s an impossible pitch to square up, particularly in conjunction with the disguised release point. 33% of his changeups produce either a foul ball or a swing and a miss, stacking up comparably against Scherzer and Nola (32%). All this is despite throwing 25% changeups, more than anyone he compares to except Matt Andriese; a number that would have put him fourth among qualified starters last year. It’s not just an out pitch, though. When he’s behind or even in the count, he basically never throws it for a ball:

And when he gets to two strikes, he buries the ball to lefties:

Just to prove it’s not a fluke, he does the same to righties:

With this kind of location (albeit better to lefties than to righties), it’s no surprise that his changeup usage spikes to a massive 40% with two strikes. It’s also no surprise that hitters just can’t do much with it. When they do manage to put a bat on a Weaver changeup, their average exit velocity is a putrid 81.9mph. Putting all of this together, the only three pitchers in the sample to allow a lower xwOBA with their changeup were Scherzer, Carrasco, and Chase Anderson, a delightful trio.

Death to Lefties

Honestly, this might be a rehash of the above point, but I think it’s a major reason for his success. The way pitching is going these days, getting out righties as a righty is doable. This is a major reason why it’s much easier to be a reliever than a starter. Getting out lefties as a righty, though- that’s the brass ring. Would it surprise you to hear that only Anderson and Lance McCullers allowed a lower xwOBA to lefties than Weaver (among his comparables)? Honestly, it probably wouldn’t. It follows with the filthy changeup from above. The best way for a right-handed pitcher to attack a platoon disadvantage has always been with an effective changeup, and Weaver’s changeup is by far his best pitch. Slight tangent here: you may recall that while the Red Baron was in on Luke Weaver, Keith Law notably continued to rank him as a poor prospect because he was likely to end up in relief. I understand that a lot of that was based on arm action rather than his pitch mix, but I find it hard to believe that a righty with acceptable velocity and an absolutely jaw-dropping changeup could top out as a reliever. Pitchers with the tools to get opposite-handed hitters out just have a huge leg up on sticking as starters.

If you wanted to touch on only the best parts of Weaver, we could stop here. This deceptive guy with an unhittable pitch who crushes lefties sure sounds like a great pitcher. And honestly, I’m tempted to leave it at that. There are areas to improve on, though. There’s a reason that I haven’t mentioned his curveball once so far. That reason is that it’s just not a very good pitch yet. It needs more speed, more break, or both. In his disastrous start against the Cubs earlier this year, for example, he threw nine curves. The Cubs swung at five, resulting in a foul, a whiff, and three balls hit over 100 miles an hour. Pick your favorite synonym for abysmal.

If I keep coming back to Nola, that’s because he seems like the pitcher Weaver maps best to, and he’s throwing a similar-speed curve, albeit with an additional 6 inches of run. To take the next step to being elite, Weaver needs to develop his curve into another pitch he can miss bats with, rather than a show-me pitch that doesn’t really do much. That said, after this deep dive into how effective he already is with his current approach, I’m not betting against him. He may already be experimenting with one breaking ball change this year- his curveball velocity is up 2.5mph so far this year. It hasn’t really worked yet, as hitters still aren’t missing it, but If that starts clicking, look out. Nothing could make Luke Weaver LOOK like an ace, but a plus curve might make him pitch like one.

(Quick methodology note: I picked 60 IP minimums to get a wide sampling of pitchers without grabbing any monumentally tiny samples- at ~250 batters faced the stats should be at least marginally stable. You can see a list of my pitcher comparables and their raw 2017 data here.)