The Giants Took a New Angle With Sinkers
Ah, sinkers. Wait three years, and the “smart” view of them will change. In the early 2010s, they were the coolest. A few years later, they were a laughingstock, a sure way to make your franchise seem old-fashioned. Between 2015 and 2019, the league abandoned sinkers (and two-seam fastballs, which I’m including in today’s analysis) en masse. In the former, pitchers threw 148,000 sinkers. By the latter, that number fell to 116,000. That trend is still ongoing; 2021 saw only 109,000 sinkers.
Despite the downward trend in usage, sinkers are cool again. When the league switched to Hawkeye tracking technology in 2020, the public could suddenly see the impact of seam-shifted wake, an effect that creates movement that previously wasn’t being measured. Pure transverse spin — like a backspinning four-seam fastball — is one way to create movement. Seam-shifted wake is another, and sinkers have it in spades, though they also generate plenty of movement from spin.
What does that word salad mean? Basically, sinkers drop and fade more than you would expect given the spin that pitchers impart on them. It’s not like sinkers started moving more in 2020 when cameras caught this effect, but quantifying something makes it easier to look for and teach.
The Cubs and A’s have been at the forefront of the sinker movement, and they led the league in sinkers thrown last year. The Giants were in the middle, 15th in baseball in sinkers thrown — but two of their breakout pitching stars this year relied on sinkers, and I think they might be on to something in the way they’re using the pitch, so let’s take a look at what (maybe) makes them unique.
Logan Webb and Alex Wood each had excellent seasons in 2021. The former looks like a breakout star, and the latter shook off years of inconsistency and earned himself a multi-year contract on the back of his resurgence. Both of them relied on sinkers to do so. And not just any sinker, either; both of them tormented opposing batters by throwing their sinker for called strikes over and over.
Wood finished third among starters in called strike rate on sinkers; Webb finished sixth. In aggregate, the Giants finished first in baseball in called strike rate on sinkers and third in called-strike rate overall. Swinging-strike rate is rightly lionized by the analytical community — throwing a pitch past someone’s swing is the surest way to get a strike — but the Giants assembled one of the best rotations in baseball, and I think their use of sinkers has a lot to do with it.
Bear with me here: this article is about to get speculative. I think there’s something unique that the Giants harnessed in Webb’s and Wood’s sinkers, and I think I know what it is: vertical approach angle, which is a fancy way of saying how much the ball is falling when it crosses home plate.
I know, I know: that’s basically just a way of describing all pitches. But plenty of recent advances in pitching analysis have been a matter of quantifying things that previously lived in the realm of “I know it when I see it.” Putting numbers on it lets us test theories about how and why certain pitches work. The basic idea behind looking at pitches by approach angle is simple: batters expect a certain downward angle when they see pitches approaching home plate. If your pitch is falling at a shallower angle — closer to horizontal, in other words — batters will expect the pitch to fall by more than it actually does. If your pitch is falling at a steep angle, batters will expect it to fall less than it does.
What does this mean in practice? Four-seamers high in the zone miss more bats if they’re falling at a shallow angle, because batters swing under them. Likewise, breaking balls that dip below the zone work better if they’re steep, because batters swing over them. Alex Chamberlain covered this logical effect earlier this year.
Why talk about this in an article about sinkers? Chamberlain’s article touches on another interesting characteristic of very shallow (or flat-angle) pitches. Here’s the relevant chart:
This is a swing frequency chart, with blue denoting infrequent swings and red denoting frequent swings. Shallow pitches at the bottom of the zone mostly result in hitters keeping the bat on their shoulder. Why? Because they expect the pitch to drop more, which would take it out of the strike zone. The pitch comes in flatter than expected and holds the bottom of the zone, and there you have it: called strikes.
That said, it’s hard to throw a shallow-angle sinker at the bottom of the strike zone. Sinkers sink, and pitches low in the zone are moving downward, and that’s just a ton of factors working against creating this unicorn of a pitch. Naturally enough, sinkers thrown low in the zone have the lowest VAA on average:
How can you create these take-friendly sinkers? You could do it like Tyler Rogers, the superlative Giants reliever. He’s a submariner, which means his pitches travel toward home plate on an upward trajectory. But most pitchers can’t do that, and they certainly can’t do it with the kind of velocity that Wood and Webb have.
Still, Rogers is instructive of how pitchers can create called-strike-friendly sinkers. For a given point where a pitch crosses home plate, a lower release point will mean a less negative approach angle. It’s a matter of geometry: a pitch that starts at ten feet above the ground and ends at two feet above the ground needs to descend quite a bit, and a pitch that starts four feet above the ground and ends at the same place takes a much flatter downward path.
In other words, lower release slots make for shallower (less negative) approach angles, which make for a higher chance at called strikes. Both Wood and Webb release the ball fairly low to the ground, though they get that effect in different ways. Wood throws from a low three-quarter arm slot, so he’s naturally releasing the ball closer to the ground. Webb has a drop-and-drive delivery, which means he takes a long stride toward home before releasing. The longer your stride, the lower your torso gets, and the lower your torso, the lower your release point, even if you throw from a normal arm slot.
For most players (with standard Judge and Altuve exceptions), the bottom of the strike zone is around 1.5 feet off the ground. Want to draw some called strikes at the bottom of the zone? You’re going to want a relatively shallow VAA. That means a low release point, which is where Wood and Webb come in.
Sinkers in the 1.25–1.75 feet off the ground band cross the plate with an average -7.5 degree approach angle. Giants pitchers bucked that trend, as their average called-strike-seeking sinker crossed the plate with a -7.2 degree approach angle, the third-shallowest team mark in the league. Both Wood and Webb checked in shallower than the team average: -6.8 degrees for Wood, -7.1 for Webb.
Batters repeatedly believed that sinkers from both of them would fall below the zone. They were repeatedly wrong. Here’s the visual evidence. First, Wood uses his low arm slot to fool Max Muncy. He has a great eye, but that necessarily means he’s a little passive, and this pitch takes advantage of that tendency. Even a more aggressive hitter might have left the bat on their shoulder here:
Webb gets a similar effect, but he does so with a different approach. Here, he bewitches Ketel Marte with a perfectly located sinker:
That pitch had massive horizontal movement, but it was over the plate more or less the whole time. The bigger problem, from Marte’s perspective, was that he expected it to fall off the plate. Instead, it just held the bottom of the zone. Like swinging strikes, called strikes often come down to deception. In this case, the deception is simply different; a pitch ending up an inch higher than the batter expects is the difference between a taken strike and a ball.
If this all sounds kind of like nonsense to you, you’re not alone. I’m still in the “is this even a meaningful effect” phase of research. But I really do think that despite how strange it sounds — wait, you want your sinker to be flat? — there’s something to throwing the pitch with a shallow approach angle.
Here’s a simple way of thinking about it. Take all sinkers thrown to that at-the-knees zone — between 1.25 and 1.75 feet off the ground. Bucket them into three groups: roughly average, shallower than average, and steeper than average. The shallower the pitch, the better the result (using linear weight run values):
Type | Run Value/100 Pitches |
---|---|
Shallow | -0.60 |
Average | -0.42 |
Steep | 0.49 |
There’s a lot more to this, for sure. Webb, for example, doesn’t just throw a shallow-angle sinker. As Luke Hopper noted this summer, he throws one that gets more drop than any other starter’s in all of the majors. Tons of drop with a shallow approach angle? That sounds like sorcery to me, but the numbers don’t lie. There’s every chance that Webb’s success owes more to the combination of angle and movement than to either one alone.
Likewise, Wood’s success might have something to do with his changing release point; he dropped down significantly this year, which made his sinker shallower but also changed the angles on basically every pitch. It’s not some small change, either. Here he is in 2018:
And 2021:
That’s a huge change, and while getting a few extra called strikes on his sinker is certainly a benefit, there’s no way it was the reason for the new release point. And Wood’s sinker is completely different than Webb’s sinker in many ways, despite sharing its shallow trajectory. It generates almost no seam-shifted wake, with a path home that’s basically true to the spin he imparts on the ball out of his hand. It doesn’t sink much, either, while Webb’s fastball drops ten inches more on its path homeward. It doesn’t even get many grounders. This is absolutely not the case of the Giants teaching two pitchers the same pitch.
In fact, if you take anything away from my analysis here, I think it should be that the Giants are willing to get to the results they want in many different ways. Shallow sinkers? They’re definitely into them, and they seem to understand that called strikes are the reason that pitch pops. But whether you get there by dropping and driving like Webb, or with release point and spin like Wood — or by underhanding it like Rogers — they don’t seem to mind. They’ve created a solid pitching staff and found interesting ways to prevent runs without turning into a cookie-cutter pitcher academy. It’s a cool feat, and I’m anxious to dig further into how they did it.
Ben is a writer at FanGraphs. He can be found on Twitter @_Ben_Clemens.
This is amazing. In my experience, people usually talk about flat VAAs and four seamers living at the top of the zone. This is very much the inverse of that, and something I wouldn’t have put together on my own. (Of course, the Giants have some of the best people in the world for that kind of stuff, so it makes sense they’re figuring stuff out like this)
I’m actually a little confused by this. To me, the mirror of flat VAAs and four seamers at the top of the zone is deep (?) VAAs and two seamers at the bottom of the zone. So: looks like a strike, moves out of the zone. The opposite of that pairing would seem to be flat VAAs + four-seamers at the bottom of the zone or deep VAAs + two seamers at the top: looks like a ball, but comes back to the zone. And that would seem to be the more effective, numbers-based version of what the article proposes.
But maybe I just need to think about two seamers and four seamers in isolation.
And I suppose that latter pairing would have the disadvantage of running back towards the barrel, as opposed to running away less (like with flat two seamers at the bottom: they’re still dropping and still grounder-inducing).