Where Kevin Pillar Tops Joey Votto

The Angels and Brewers recently made a trade, with Milwaukee ending up with Jett Bandy. I’m going to guess you didn’t linger on the deal, if you heard about it at all. It’s not a move that’s going to change the landscape of either league. But I saw an interesting tweet on Tuesday:

David Stearns says Bandy has plus raw power that hasn’t necessarily shown up in his stats. Playing time split with Susac/Pina remains TBD.

— Adam McCalvy (@AdamMcCalvy) December 13, 2016

Jett Bandy: plus raw power! But for the most part invisible, at present. This is fairly typical scout-speak, and if you just concentrate on Bandy’s profile, there could be something there. He’s an extreme fly-ball hitter, and he stands 6’4. He turns 27 in March. He could be a really strong guy.

Thinking about the tweet, however, got me thinking about Statcast. What might we be able to learn about raw power from baseball’s new tracking technology?

Most of us get our Statcast information from Baseball Savant, and there’s a link there to a Statcast leaderboard. It includes, of course, two years of data, and one of the columns included is maximum exit velocity. Almost all of the time, the focus is on average exit velocity. Sometimes even average exit velocity within a narrow band. But maybe that maximum can tell us something. What that really is is a hitter’s observed batted-ball-speed ceiling. Seems like that’s another way of measuring raw power.

There’s a difference between raw power and game power. Raw power is more like what a hitter can do, while game power is more like what a hitter does do. Taking a look here, over the last two seasons, there are 489 hitters with at least 100 batted balls. By maximum exit velocity, Jett Bandy actually ranks just 455th. He’s down there by names like Skip Schumaker and Zack Cozart. Bandy hasn’t really killed a ball even once. It makes you wonder.

On the other hand, again, this is a player’s observed ceiling. The smaller the sample size, the less likely it is for that observed ceiling to match the actual ceiling. The maximum holds greater significance as a player’s playing time increases, and Bandy has yet to clear 200 batted balls. The Brewers might still be on to something, and you could conceivably learn even more by tracking minor leaguers and batting practices. Reps are reps and swings are swings, right? Raw power doesn’t really have to be guessed at anymore. There are ways to have it measured.

But now let’s shift away from Bandy for a few minutes. I promised this would get to Kevin Pillar and Joey Votto. They’ve played regularly in each of the last two years, so they’ve built large sample sizes. Votto is one of the best hitters in the world! Pillar, I suppose, might also qualify as one of the best hitters in the world, but you’d have to really relax the standards. Pillar is one of the, I don’t know, 1,000 best. Votto is probably one of the 10 or 20 best. Anyway.

If you focus just on exit-velocity ceiling, Pillar gets the edge. This might surprise you. Pillar has topped out at 113 miles per hour. Here that is:

Votto, meanwhile, has topped out at 109.3 miles per hour. He has a lower observed ceiling, and while three or four ticks might not mean much to you, look at this list. Comparing Pillar and Votto since the start of 2015, Pillar owns the 12 hardest-hit batted balls. That means something! That speaks to what Pillar could, in theory, accomplish.

The big difference is in what you might call efficiency. Now, for this part, I’m using corrected Statcast information from our own Jeff Zimmerman. Actual Statcast misses about an eighth of all batted balls, most of them poorly hit. Zimmerman has tried to account for that by using estimates for those unproductive and unmeasured batted balls. The results aren’t perfect, but they’re more complete and less misleading than what Baseball Savant is presently able to publish.

I’m defining efficiency very simply: It’s just average exit velocity / maximum exit velocity. I’m not sure how much this adds, but it’s at least a new way of looking at things. The average is about 76.8%. One standard deviation is right around two percentage points.

Kevin Pillar shows up at 72.6% efficiency. That’s easily worse than average — Pillar, relatively speaking, makes a lot of imperfect contact. You knew this, but, here you go.

Joey Votto shows up at 80.7% efficiency. That’s easily better than average, and it’s another indication of how Votto makes a habit of squaring the baseball up. Though Votto’s peak exit velocity is almost four ticks lower than Pillar’s, Votto’s average exit velocity is more than six ticks higher. One of these guys is a line-drive machine. One of these guys is Kevin Pillar.

Another unsurprisingly inefficient hitter: Jonathan Schoop. Watch him hit a ball almost 115 miles per hour!

Schoop shows up at 72.7% efficiency. Anyone who’s watched him for a while knows he could obliterate any one given pitch, but most of the time, he doesn’t do that, either because his swing is just off, or because he’s gone after something out of the hitting zone. Schoop isn’t close to as good as he could be.

And some other efficient hitters: Joe Mauer, 81.5%. Ryan Howard, 81.4%. Dustin Pedroia, 80.8%. Justin Turner, 80.8%. Pedroia has shown an even lower exit-velocity maximum than Bandy, but by comparison he spends more of his time making nearly-optimal contact. That’s why Pedroia has a shot at a Hall-of-Fame career.

I’m not totally comfortable embedding longer leaderboards, because Statcast comes with certain spot-errors, and I can’t get rid of them all. For example, Jonathan Schoop’s maximum exit velocity shown on Baseball Savant is incorrect. I eliminated that data point, but I couldn’t conceivably do that for everyone. So it’s kind of a case-by-case-basis thing, and I’m not even certain this information clarifies something we didn’t know or couldn’t guess.

But at least Statcast continues to open new doors. In theory, we can start getting real readings on each player’s power maximum. And then we can see how often players hang around close to that ceiling. If you limit it to game results, it could take a while before a ceiling feels legitimate. We do, however, have time. And teams themselves have even more opportunities to gather their data. All of it’s good for something.