Aaron Nola Throws Two (Ish) Fastballs

Let’s start this one with the basics. What’s a four-seamer, and what’s a sinker? At their core, they’re both fastballs; the main difference between the two is in the grip. Place the seams perpendicular to your index and middle fingers? That’s a four-seamer – each finger crosses two seams. Place the seams parallel to the fingers? It’s a two-seamer or sinker – one seam per finger.

Of course, you could also define them by their movement. Does it have a ton of tail and not much ride? It’s a sinker. Does it mainly fight gravity with backspin, paired with far less tail? It’s a four-seamer. If you think of archetypical examples of each, it’s easy to tell the difference. Think Clayton Kershaw’s four-seamer – all backspin – and Adam Wainwright’s sinker – boring in on righties’ hands and knees.

Real life doesn’t operate in archetypes, though. Real life is messy. Statcast doesn’t get to stop the game after each pitch and ask a pitcher what he threw, and not every fastball is a textbook definition of its type. Plenty of pitchers throw both varieties of fastball, and they can look extremely similar, even with the benefit of high-speed cameras and piles of pitch data.

Want a practical example? You’re in luck – or, well, not really. I’ve walked you into wanting a practical example with my introduction, and that’s on purpose, because today I want to talk about Aaron Nola’s two fastballs. Nola, like many pitchers, throws a sinker and a four-seamer. Like many pitchers, he releases them from a consistent arm slot – a remarkably consistent arm slot, in fact:

Not only that, he throws them at about the same speed. On average, his sinkers come out of his hand about a mile an hour slower than his four-seamers, but that’s not that big of a difference when you take variance into account. Nola threw his slowest four-seamer at 89.3 mph in 2021. All but nine of his sinkers were faster. He threw his fastest sinker at 95.2 mph – only 33 of his 1,121 four-seamers were faster.

To make matters even more confusing, Nola’s four-seamer isn’t one of those perfect-backspin jobs. In fact, it has nearly as much tail as rise – it’s more of a 45-degree break. That’s due to his delivery, a low-three-quarters motion. Unless you’re Josh Hader, it’s hard to throw a prototypical four-seamer from this arm slot:

Why should we care? Because looking at how Nola differentiates his fastballs can tell us something interesting about two-fastball pitchers in general. Does this matter? I don’t know. But I thought I’d show my work, because a) this is a neat puzzle that Jeff Zimmerman pointed out to me and b) I’ll write about anything in January during a lockout.

Okay, let’s start making my case. Here’s a scatterplot of the long-form movement (PFX_X and PFX_ Z in Statcast files) of all of Nola’s fastballs from 2021:

You can certainly see two clusters there, but it’s really close to being one undifferentiated mass. Throw in a few more dots in the middle, and you might think it’s just a single pitch. The difference between the center of those clusters is enough to matter – roughly six inches – but it’s not an unbridgeable gulf, and pitches cross over each other’s movement definitions all the time. If you’re only looking at velocity and spin, it’s hard to tell the pitches apart:

So again, how do you define the pitches in the middle? One dense point is four-seamers and the other is sinkers, but most pitches are somewhere in between. Here’s the same graph, only done in an uglier graphing tool that can display which pitch type Statcast assigned to each:

Are they differentiated? Sure! There’s certainly overlap between the two, but the two clusters are clearly separate from each other. When Nola goes to his sinker, he’s inarguably getting more sink and tail. When he uses a four-seamer grip, he’s getting more rise and less tail, though his four-seamer still has far more tail than average. They’re even closer than they were in 2019:

Why bother to throw both when they’re so close to each other? Similar break, similar velocity, similar spin and efficiency; there’s just not much to separate the two. But there’s a good reason to split these pitches, and I think that it has a lot to do with why Nola continues using both despite the two pitches looking like doppelgängers.

If you’ve been reading much of my stuff recently, you guessed it: it has to do with vertical approach angle. See, Nola’s slingshot delivery means that all of his pitches start out low to the ground. His average release point was just over five feet off the ground; the only starters who checked in lower were Alex Wood and Freddy Peralta. That means his fastballs are well-suited to do one of two things: miss bats at the top of the zone or steal strikes at the bottom of it.

Importantly, a flat approach angle isn’t the only thing that helps you miss bats at the top of the zone. Getting rise on your pitch is important, too – and in fact, the two are correlated. When Nola is hunting swings and misses, he throws his four-seamer almost exclusively. If you draw an imaginary line 2.75 feet over home plate, 76% of his fastballs above that line were four-seamers. They were meaningfully better than the sinkers, too; they generated more whiffs per pitch and per swing.

When he ventured down in the zone (2.25 feet over home plate or less), things changed. There, he still throws more four-seamers – he throws far more four-seamers overall, as you can tell by the scatterplots above – but roughly a third of his low fastballs in 2021 were sinkers.

Here’s where things get a little weird – the low-in-the-zone, flat-angled sinker doesn’t really pop for Nola like it does for, say, Logan Webb. In fact, if you look only at how good each pitch is at generating called strikes low in the zone, Nola’s four-seamer is better. Better at missing bats up in the zone, better at drawing called strikes down in the zone – why even throw the sinker?

Nola throws the sinker because you can’t get a called or swinging strike on every pitch. Sometimes batters put the ball in play, and that’s always been a problem for Nola’s four-seamer. For his career, he’s allowed a .382 wOBA on contact (.395 xwOBA). That’s worse than average, one of the few blemishes on his sterling track record. His sinker does better – .369 wOBA on contact and a .354 xwOBA. Opponents barrel up his four-seamer nearly twice as frequently. The reason why isn’t a surprise – the pitch lives right on the 45-degree line, which is a recipe for solid contact, while the sinker gets far more grounders.

The damage-on-contact problem is even worse when Nola ventures down in the zone. It’s not a massive sample, but opponents have hit for a .413 wOBA on low four-seamers from him, and the expected stats agree – .422 xwOBA and 8.8% barrel rate. At only a 33.5% groundball rate, it’s not hard to guess the culprit – the best part of pitching down is keeping the ball on the ground, and Nola can’t do it with his four-seamer because it rises into bats.

Sacrificing whiffs for contact suppression isn’t always a good tradeoff, but it is when you’re throwing fastballs low in the strike zone. The solution here seems pretty straightforward to me, and I think it’s one Nola generally uses. When he’s attacking up in the zone, he can just grip it and rip it; his standard four-seamer plays well that way, even with its rather generic movement profile, because of his delivery. When he wants to attack low in the zone – based on a matchup, to set up a pitch, or simply because he thinks that’s the best location for the situation – he can simply throw the same pitch, but rotate his grip to give it sinker action.

Some might argue that Nola’s sinker isn’t really a sinker. It doesn’t really look like one; he gets a bit of help from seam-shifted wake, but not a ton, because he throws it with just as much transverse spin as his four-seamer. It doesn’t get an outrageous amount of drop. There’s really not much to make it stand out as a pitch. It mostly looks like he took a good four-seamer and threw it with a two-seam grip.

But if you want to attack low in the zone, and you’re not Walker Buehler, you need to do something different with your fastball. For Nola, that’s sinking action. It might not be hugely differentiated from his fastball. It might not be a vintage Wainwright or Dallas Keuchel sinker. In fact, as you can see from this graph of initial spin axis and subsequent deviation, it’s quite literally just his four-seamer turned sideways. That means seam-shifted wake makes the ball deviate from its initial path on its flight homeward:

But that’s really all you need sometimes. Does Aaron Nola throw two fastballs? Not really, but also yes. That’s how a lot of pitchers with sinkers and four-seamers operate – it just so happens that Nola’s delivery makes them both quite useful in different ways.