Trevor Williams might have a strong weapon at his disposal

Losing a game where your opponent puts up 14 runs is never easy to deal with. Even though it was the first game of a four-game battle against the St. Louis Cardinals, it could have been enough to set the tone for the entire series. De facto staff ace Trevor Williams wasn’t about to let that happen Friday night when the Pirates played their second game of the series at Busch Stadium.

Williams’ start arguably changed the momentum of the series, despite the fact that the Pirates only managed to score two for him (and as many in the following game). With five strikeouts and a wale, he did surrender nine hits, those only amounted to one earned run.

Coming into Friday night’s start, Williams ranked 18th (min 200 pitches thrown) in Command+ with a 118. If you aren’t familiar with this metric, I highly recommend you go and read The Athletic’s article on the foundations of Command+ by Eno Sarris.

Williams’ pitch chart from Friday exemplifies how he’s been so effective at commanding the strike zone.

His pitch ecosystem is on point- high four-seam fastballs, down and in sinkers (RHB side) and sliders (LHB side), with his changeup peppered in the lower quadrant of the strike zone.

Speaking of which, through his last three starts, Williams’ changeup and slider velocities have paired up fairly well, the significance of which I’ll get to shortly.

Anyone who is familiar with my work over at Pitcher List or Royals Review knows I’m a tunneling fiend. I like to look for ways that pitchers can get filthier and more deceptive using this tactic. I will also concede that there is little evidence as of yet that supports the effectiveness of tunneling, but its not exactly being denied, either. Its hard to imagine that it doesn’t provide some level of benefit when you consider what the concept consists of.

Here are a couple of quick visual examples using overlay’s I made a while back.

Let’s talk about how the changeup and slider can work off each other. Further down, I’ve posted the tunneling data for both pitch pairs from 2018 and 2019, the latter of which has a minute sample.

When I talked to Williams, he stated he is making some adjustments on his slider.

“What I have been working on the most in tightening up my SL” Williams said. “I think that was the biggest step forward in my last few starts. If I tightened that up then my four-seam, two-seam, changeup and slider all will look similar. It’ll enhance my curveball separation more too.”

Some of Williams’ 2019 metrics deviate a little from what we have in 2018 and that sample isn’t the greatest either. It should be noted that I don’t believe there are studies on how long it takes tunneling data to stabilize, mainly because the metrics are so new.

What does all this mean? Well, here is a crash course in tunneling metrics (all info taken from this article at BP).

RelDist (release distance) is how far apart, on average, the release points on back to back pitches are. The league average (as of data collected by Baseball Prospectus in 2017) is 2.6 inches. Each of Williams’ pitches are very tight on release, especially the CH>SL to RHH. PreMax is the distance, in inches, between back to back pitches at the decision making point, somewhere between 100 ms after release and 167 ms before they reach the plate. PreMaxTime tells us at what point, in milliseconds, the second pitch is at the largest perceived distance from the first.

FTimeDiff is basically how far apart, velocity-wise, back to back pitches are in seconds (ie- .0042 is 4.2 ms). PlateDist tells us how far apart (in inches) back-to-back pitches are from each other at home plate. Lastly, PPMax Ratio is essentially the distance between the what the batter sees prior to the tunnel point and the true distance at the plate. The higher the number, the later the break (‘late break’ is a contentious term).

Did I put you to sleep? Still with me?

Anyway, you can see most of William’s data on the changeup and slider is better than average.

Have a look at an example:

But wait...is this what the batter is really seeing? Surely it looks different standing at the plate? Well, let’s take a look using Baseball Savant’s 3D pitch tool from the perspective of Jose Martinez.

What we want to look at is the far right green trajectory and the yellow trajectory next to it. Both pitches, traveling at difference of about 2 MPH, break apart shortly after what is referred to as ‘the commit point’ (about 169 ms before it reaches the plate).

Referring to the velocity chart earlier, both pitches traveling at speeds that close appear identical for a period time. Its important because your mind’s eye interprets a 95 MPH pitch differently than an 85 MPH pitch (which could go without saying).

Here’s where things get interesting. If we look into the average spin axis of both pitches, it shows that Williams’ change and slider create a conflicting tilt. His changeup averages about 237-degrees while his slider is around 120-degrees. That difference creates an opposite break which further helps facilitate an effective tunnel as long as both pitches can keep on the same trajectory as long as possible.

As he pointed out, if Williams can get his slider to where he wants it, especially considering his release points are as tight as they are, he’s got a really tough pitch combo to lay on hitters.

The tunnel data charts show Williams isn’t using the combo much but it would be interesting to see what the results would yield should he start attacking hitters with that pattern in random parts of the count. It doesn’t have to be a mainstay of his repertoire but it could be a very powerful weapon to pull out when the heat is on.

Thanks to Eno Sarris (Command+ data), the content at Baseball Savant, Brooks Baseball, and Baseball Prospectus, and Trevor Williams for speaking with me.