“Different day, different arm,” is one of those things you’ll hear a pitcher say. You get up on the mound on a given day, and you try to figure out which pitches are working, what parts of your body are barking, where you can actually intentionally throw your pitches. It’s understandable, given the complicated mechanics required to throw the ball so hard, with so much movement — but it has implications for those who would attempt to place a number on their true-talent ability.
We know about this difficulty when it comes to pitching. Projections try to put a number on the true ability of a player, but pitching projections lag behind hitting projections. Even when a stat — like exit velocity — becomes meaningful in similar samples for hitters and pitchers, it behaves strangely for pitchers. It becomes meaningful quickly but isn’t quite predictive, either — maybe because pitchers add pitches, change the script, and become different more quickly than hitters. Maybe because their true talent shifts often.
Maybe true talent for hitters shifts more than we think, though. At least when it comes to their actual ability to express that true talent due to health reasons.
It’s not quite the same for a hitter as a pitcher, but it certainly sounds true when you hear a veteran say that the only day on which a baseball player is 100% is the first day of Spring Training. And that’s not even true of everyone. And that has cascading effects on their abilities.
Take Carlos Correa.
Let’s look first at his exit velocity for the year.
Nobody’s chart is a straight line, but there are a couple big dips in this constellation. It gets a little dicier when you look at that in tandem with this chart of his launch angle over the course of the year.
When I asked Correa about those dips in performance, and his dropping launch angle, the answer was… confusing. You’ll have to read between the lines a bit.
“Some of those things people don’t know,” Correa said last week. “Some parts of the body are hurting so you have to lay off some things and deal with some things. It’s something that people don’t know, but obviously you know.” When I stopped him and asked him what was hurting, he stopped short. “Nothing. Just stuff that goes through the year, you have to try to manage. It’s a long season, everyone goes through ups and downs and everybody gets hurt at some point.”
But we know that Correa is dealing with a sprain in his shoulder, because we have some postgame notes on the subject. Let’s go back to that exit-velocity chart and annotate it with some injury notes. In each case, I’ve chosen the first occasion in a string of updates on which the injury in question was mentioned at his RotoWorld player page.
Correa rolled his ankle fairly severely earlier in the season, on June 9. He sprained his shoulder sometime around September 7. Because these are rolling averages, you’d expect any effect to show up a little while after those injuries. In both cases, his exit velocity took a tumble after the injury. In both cases, his launch angle took a tumble, too.
If you’d like to compare swings between a “healthy” Correa early and late with your own eyes, you might discover less aggression and bat speed in his average swing this month, and maybe a slightly different follow through. Here are two examples, both against Texas lefties at home, on similar-ish pitches. First, against Cole Hamels in May:
And then, against Martin Perez this month:
An advanced scout might spot this. They’d notice the dip in exit velocity and launch angle, take a look at the swings with a more trained eye, and make some recommendations to the team based on these factors.
Some might find that frustrating. It would be based on small-sample numbers, and any recommendation to treat a player of Correa’s caliber as anything less than excellent — it could surely come around to bite his opponents. Because Correa has a true talent that’s 20-25% better than the league according to projections, not 10% worse than the league, like he’s been showing in September.
On some level, we already know ignoring injury notes is folly and should be baked into projections. Jeff Zimmerman found that we under-project the slugging percentage for players who’ve previously played through pain — by as much as 14 points of slugging — mostly because our projection systems can’t know that the player was anything less than his best self.
The most recent example of this might be Robinson Cano. After playing through a hernia injury on the way to the worst season of his career in 2015, Cano was projected into a .442 slugging percentage by Steamer, and then went on to beat that projection by 80 points so far this season. The numbers saw his 2015 as poor play because they couldn’t see why. We could, though, and we could find a way to include that knowledge of his injury into the projections, particularly if we found a way to reliably turn player news bits into data.
Our projections would be better but they still wouldn’t catch it all. We’d still miss the daily nicks and cuts that never make it into a game preview, game reaction, column, or news roundup. And there’s plenty of those. “Different day, different body,” you could say.
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.