Brandon Finnegan Improves, Again, Probably

With the switch from the standard PITCHf/x pitch-tracking system to Trackman, a few missteps are to be expected. Dave Cameron, for example, recently pointed out that we have to be careful about reporting velocity right now. Jeff Zimmerman gave us a way to convert old velocities into today’s reality, and Tom Tango offered up way to use current readings to approximate velocities as they were calculated by the previous methodology. For those interested in the simplest possible method, it appears as though subtracting three-quarters of a mile per hour from a pitcher’s reported velocity arrives at roughly the same figure as the previous system would have produced.

But that doesn’t account for the situation in its entirety. If we look at what Brandon Finnegan did in his first (excellent) start, we’ll notice that the movement numbers are also a little off right now. It’s enough to want to throw your hands up and just emote, as I did on the latest episode of our Sleeper and The Bust podcast.

It’s okay to express frustration. It’s cathartic and release is good. Breathe deeply through the nose.

But we also need to pick ourselves up and dust ourselves off. Finnegan did not, as the leaderboards suggest, just record the ninth-most ride on a fastball that we’ve witnessed over the last 15 years. That’s not what was so impressive about the movement and velocities of his pitches in his 2017 debut. But discovering what was impressive can help us better navigate the suddenly unsteady waters of pitcher analysis.

What’s going on with the movement numbers is a little different than what’s happening with their velocity equivalents. There may be some difference due to tracking the ball straight out of the hand versus tracking it from a standardized point in space, but it’s more about calibration. We have a new system of trackers, so we have a new system of errors from park to park. We’re back in 2002, trying to figure out which parks have a hot gun, and which parks have weird movement numbers. The O.C. and The Sopranos are on TV and we’re listening to Outkast’s “Hey Ya” on the radio.

So, in order to create some standard adjustments, smart people will be comparing players across different parks and attempt to equalize the numbers so that we don’t have a leaderboards that are full exclusively of this year’s pitchers. As for calculating the ride, specifically, on Brandon Finnegan’s fastball, we can begin by looking at the numbers his opponent, Jerad Eickhoff, recorded and adjust according. Readings suggested that Eickhoff added 4.9 inches of ride to his fastball between last September and this April. Using that figure as a baseline, then Finnegan’s reported 6.0-inch improvement Finnegan actually does hint at a real change. It suggests Finnegan might have added an inch of ride (6.0 vs 4.9) between last year and this one.

But that just represents a simple test case. Clayton Kershaw, pitching in Los Angeles, added about three inches of ride between last year and this. We can’t just subtract five inches from every pitcher and call it a day. So we’ll work on calibrating the different systems and bring them online. That’ll take a while, though. What can we do in the meantime? What’s our three-quarters of a mile per hour shorthand here?

There might not be one. But we can take some hope from the fact that it’s often best to compare a pitcher to himself — in other words, to look at the pitcher’s pitches compared to his fastball. The fastball sets the tone, and the other pitches play off of it. So if you can define movement within the pitcher’s arsenal, then you can compare it to his previous offerings.

Let’s do that for Finnegan’s changeup, since he had one of the most improved changeups of last year’s second half. Here’s the movement and velocity on his changeup compared to his fastball at three different points: early last year, in September last year, and in his first start of the year. Remember that drop is more important than ride for swinging strikes, and that velocity gap is also good for swinging strikes.

Brandon Finnegan’s Changeup vs. His Fastball
Time Period Velo Gap Horiz Diff Vertical Diff
First Half 16 7.0 2.4 3.9
September 16 9.9 1.8 3.9
April 17 9.0 1.3 5.9
SOURCE: Brooks Baseball
All aspects compared to Finnegan’s four-seamer

The immediate takeaway is that Finnegan has improved his changeup once again. The added ride on his four-seamer has created another two inches of difference between his four-seamer and his changeup, a gap that can lead to swings and misses. He also kept most of his improved velocity gap, which is good news. He did this, which is also good news:

There’s some doubt here, of course, concerning to which fastball we should compare the changeup. Finnegan threw more sinkers than four-seamers on Wednesday, and we could have used that sinker for comparison. If we do, the lefty only improved his drop by 1.5 inches instead of 2.0. But if you look at Baseball Prospectus’ tunneling stats, you’ll see that Finnegan used his changeup and four-seamer in tandem last year just as often as he used his sinker and changeup together (105 vs 107, respectively). The general analysis holds. It’s hard to be more specific right now, anyway.

Finnegan threw the changeup 16 times on Wednesday. It got two whiffs and was put in play four times, all of them resulting in an out. More importantly, it showed that he (probably) kept most of the gains that he made late last year, when he improved the changeup and gave himself four different legitimate pitches. He might have even made it even better.

With a phone full of pictures of pitchers' fingers, strange beers, and his two toddler sons, Eno Sarris can be found at the ballpark or a brewery most days. Read him here, writing about the A's or Giants at The Athletic, or about beer at October. Follow him on Twitter @enosarris if you can handle the sandwiches and inanity.

newest oldest most voted
Ryan Brock

“throw your hands up and just emote”, indeed. It’d be less frustrating if it seemed like MLB/statcast folks were putting more effort into translating/contextualizing vs. the old data, rather than just dumping the change out there and hoping for the best.

Ryan Brock

(… that said, thanks for doing your best to start that process of figuring this out!)