St. Louis Cardinals: Cracking the code of Tyler Webb’s success

ST LOUIS, MO - AUGUST 21: Tyler Webb #30 of the St. Louis Cardinals reacts after giving up a grand slam to Matt Davidson #64 of the Cincinnati Reds during the sixth inning at Busch Stadium on August 21, 2020 in St Louis, Missouri. (Photo by Jeff Curry/Getty Images)
ST LOUIS, MO - AUGUST 21: Tyler Webb #30 of the St. Louis Cardinals reacts after giving up a grand slam to Matt Davidson #64 of the Cincinnati Reds during the sixth inning at Busch Stadium on August 21, 2020 in St Louis, Missouri. (Photo by Jeff Curry/Getty Images) /

When it comes to the bullpen of the St. Louis Cardinals, there isn’t one player who gets more hate than Tyler Webb. It could be the specs, it could be that he looks like a middle-aged dad, or it could be that his stuff just doesn’t look all that special.

However, over the past two seasons, Webb has quietly been one of the best relievers on the Cardinals. In 2020 specifically, he led the bullpen in appearances (21) and ERA (2.08).

This was just a 21.2 inning sample, but if you expand it out over the past two years, Webb has a combined 3.28 ERA over 76.2 innings with a 1.043 WHIP. He doesn’t strike people out much, and through traditional advanced metrics, his stuff isn’t all that good.

From 2020, his fastball and sinker velocity averaged right around 90.5 mph and those plus his curve were clearly nothing special from a spin-rate perspective. However, the results were all fantastic from the outcomes and expected point of view.

Before you claim a small sample size, the 2019 expected and advanced statistics say the exact same thing.

Webb truly is an enigma. What we will attempt to do here today is to bridge the gap between what we see from a results side of things and what the advanced metrics say about his pitching arsenal. How in the world does Webb do it?

In order to do this, we’ll need to take a brief aside from the wonderful world of Tyler Webb to talk about physics and the recent breakthroughs in advanced metrics in baseball. 20 years ago, the advanced statistics in baseball were nothing compared to what they are now. Given advances in radar technology and computers, the amount of detailed information publicly available about the sport we all love has been growing at an amazing rate.

What we already know about the physics of baseball

It all started with pitch location and movement, then the introduction of Statcast in 2015 added more wrinkles. With Statcast, spin rate data became available which opened up a whole new world of research opportunities.

It was really around this time that people much smarter than myself began to dig in to use this technology to explain just what happens between the time a pitcher releases a baseball and when it gets to home plate.

As Ben Clemens of FanGraphs details well in this article, Dr. Alan Nathan of the University of Illinois was one of the first to publish papers on why spin rates matter. Dr. Nathan also published this paper on how spin axes could be calculated from spin rates.

Look around for Dr. Nathan on The Hardball Times and you’ll be busy for hours. Since 2015, Trackman radar has been measuring spin rates and while the spin rate is important, it isn’t even close to everything that matters with a pitch. The way I like to think about spin rate is kind of like potential energy. A low spin rate can put a cap on how much movement a pitch could have, but a high spin rate doesn’t matter unless it can be converted into motion efficiently.

Speaking of efficiency, the terms “spin efficiency” or “active spin” are thrown around a lot in these conversations, but all it means is what percent of the spin is directed straight at the catcher when the ball is released from a pitcher’s hand. If the ball is angled towards either dugout even a little, it loses efficiency. The fun part is that a less efficient pitch isn’t necessarily a bad thing.

In more scientific terms, there is both transverse and axial spin. If a ball is 100% directed at the catcher, it is pure transverse spin. Sorry to steal Ben Clemens’ metaphor, but pure transverse spin is like a tire spinning perfectly on the axle of a car. The axis that it is being spun around is perfectly perpendicular to the motion. When a cylinder or sphere has transverse spin it can generate lift through something called the Magnus Effect.

Basically, when a cylinder or sphere is spinning in a fluid (like air), there is a force created perpendicular to the flow direction. That is why efficient four-seam fastballs “rise” and 12-6 curveballs drop so much. Both are best with highly efficient spin to get the most out of their spin rates. The only reason a four-seamer “rises” and a 12-6 drops is because they are spinning in different directions so the lift vector is pointed in opposite ways.

The other half of this is side spin or gyro spin. The best way to think of this type of spin is like a bullet or football spiraling. Because any of this spin of a baseball is not directed into the velocity flow, gyro spin doesn’t generate lift the way that transverse spin. Up until recently though, there was not a great way to know what this gyroscopic spin did for the movement of the baseball. Modeling the effects of Magnus from gravity, spin rate, and spin axis were the best that was available.

Seam-Shifted Wake

Before the 2020 season, Statcast systems in each ballpark made the switch from Trackman to Hawkeye radar, revolutionizing the data available. Now with 12 cameras around the ballpark measuring everything that happens, a new way to look at pitches is here.

Beginning with the research of Dr. Barton Smith of Utah State University and others, there was a realization that seam orientation not only matters but shows that there was a whole other force happening here that can fill a knowledge gap. This gap is between the calculated and measured axis of a pitch.

With Hawkeye, there is no longer a need to calculate the spin axis from the movement and spin rate. It can be directly measured. In doing this, it was noted that the calculated axis ended up being different than the observed axis. Smith and others have been doing a ton to research this other movement, but as of now, it is commonly referred to as seam-shifted wake (SSW).

At the base of it all, it is all about how the seams on a baseball can be used to alter a ball’s flight. For pitchers, this isn’t something new. Jared Hughes, Trevor Bauer, and many many more pitchers both professional and otherwise have talked about how different grips change the way the ball moves, even if they throw it the same way on the same axis.

The difference now is that the effects of gyro spin and this side force can be quantified and therefore used as another tool to analyze pitchers.

Friday was a big day for seam-shifted wake as Clemens’ wrote on it, as did Eno Sarris of The Athletic. A couple of weeks ago, Mike Petriello of also posted on SSW and how it has aided pitchers like Shane Bieber, Lance Lynn, Kyle Hendricks, and more. If my explanation of all this doesn’t make sense, I urge you to check out theirs. There is no gatekeeping here for this data. The world is a better place if more people understand what this stuff is and how it works.

Like efficiency, more SSW does not always equal better.

This is all a huge can of worms that is only beginning to be opened.

As a senior in aerospace engineering, all we have learned about for the past three-plus years is fluid dynamics. What I’ve consistently found is that these concepts that have horribly complex equations and things happening behind the scenes normally have easy ways to understand them in real life.

Seam-shifted wake sounds complex, but it isn’t anything more than now being able to measure changes that we knew were happening already between a pitcher’s hand and the plate. New models can be built to track the path and movement of a baseball now. Academically, finding a way to mathematically explain a common physical event is significant, but if you don’t care about the math behind it, your eyes will glaze over.

I’m not trying to minimize the work that people with doctorates in physics and other degrees have done on this, but at the end of the day, seam-shifted wake is just a way to say that pitch grip and orientation out of the hand matter.

Back to Tyler Webb

One of the new tools on Baseball Savant’s website is a comparison of the spin-based movement vs the observed movement. This spin axis tool allows anyone in the world to see the difference between the observed movement axis and the spin-based movement axis.

Pull up Webb’s comparison, and it’s easy to see that seam-shifted wake helps out Webb.

Starting with the spin-based movement on the left, it is clear that Webb first does a pretty solid job of spin mirroring with all of his pitches. Spin mirroring is where pitchers try and pair the axis of a pitch that spins one way with another pitch that spins the other way. In practice, batters can see the axis of the pitch, but it’s more difficult to see the direction. When this happens, a batter can think they are getting a fastball that is actually a curveball. On a chart like this, the axis of a perfectly mirrored pitch will be 180 degrees off of the other.

More from St Louis Cardinals News

Webb’s fastball, sinker, and changeup all come out within an hour of each other, with his curveball almost perfectly “mirroring” their spin but just in the opposite direction. That means they’ll look relatively the same out of his hand. The best mirror is with the changeup and curveball, but the sinker and fastball are close.

This mirroring alone is enough to explain at least some portion of Webb’s success. Digging deeper, Webb averages 88% and 89% efficiency on his sinker and four-seam. This means that he has 12% and 11% gyro spin. With that gyro spin, seam shifted wake pulls the sinker to a lower angle to the left, and the fastball to the right. This is evidenced in the right chart.

In practice, this will cause the pitches (that leave at the same angle) to tail and cut in different directions.

Perhaps the most interesting thing about this is that smack dab in the middle of these two cutting and tailing pitches is his changeup. His changeup somehow has a 99% efficiency, suggesting it doesn’t benefit from SSW at all. This doesn’t mean that it doesn’t have any horizontal movement, but it does present an interesting case study as it stays really close to the same axis.

Add in the curve on nearly the identical axis the other way and there is a good chance that all four of these pitches look very similar out of Webb’s hand.

Here is a reel of all of Webb’s pitches in this order: four-seam, sinker, changeup, curveball. It is almost impossible to tell the axis of the pitch here, but you can see the cut of the four-seam, the tail of the sinker, the drop of his changeup, and then there’s the curveball.

There is more than meets the eye going on there with Webb and even if SSW doesn’t explain all of Webb’s success, the effects of SSW with his ability to mirror the spin of his pitches does make his arsenal play up. There are more things that could make Webb as good as he is with a worse arsenal like sequencing or control, but the main card on Savant’s home page for Webb doesn’t tell the whole story about his pitches. That is the point of all of this.

As I said in the beginning, seam-shifted wake could explain part of why Webb is good even when most things in traditional advanced statistics say he shouldn’t be as good as he is.

Next. Four unsung World Series heroes. dark

I am sure I still have more to know and understand about all of this, but for now, I’d like to think we can have a little more appreciation for Tyler Webb and his arsenal.