clock menu more-arrow no yes mobile

Filed under:

Carlos Carrasco keeps getting better

Almost every month the Indians hurler has shown a new level of excellence.

MLB: Tampa Bay Rays at Cleveland Indians David Richard-USA TODAY Sports

Carlos Carrasco wasn’t quite excellent this past Sunday. By Baseball Reference’s Game Score it was his eighth worst start with a 42, though the 6.1 inning, 5 run, 9 K outing he posted belies how excellent he was for the majority of the game. Two of those runs were because reliever Neil Ramirez gave up a couple ringing wall balls, but baseball blames the man who allowed the man to get on base.

Prior to the game — and indeed prior to that 7th inning — Carrasco’s ERA was it’s lowest since sample sizes normalized in early May. He’d been on a run for quite a while now, and despite that little hiccup, Carlos Carrasco just keeps getting better.

It hasn’t been quite linear — he hasn’t just gotten better and better with each game — but month by month this season Carrasco has just turned over another, better leaf. It seems like so long ago we were worrying about his first start of the year, and worrying if he’d lost his way with the departure of Mickey Callaway. He didn’t have a great July, but generally, it’s been improvement with each passing month:

Carrasco month-by-month splits

Mar/Apr 21.9% 4.8% 17.1% .210 0.95 .238 76.7% 3.81 3.79 25.5
May 26.9% 6.2% 20.7% .259 1.28 .330 68.4% 3.64 3.24 45.8
Jun 29.8% 7.1% 22.6% .276 1.42 .400 65.2% 2.46 3.52 41.2
Jul 33.0% 4.1% 28.9% .261 1.15 .357 85.5% 3.27 2.56 45.0
Aug 32.5% 3.3% 29.2% .216 0.95 .313 79.6% 1.71 2.15 40.8

It (sort of) calls to mind Corey Kluber’s second half of 2017. In the first half he’d been his usually great self with a 2.80 ERA and striking out 27.2 percent of hitters while walking only 6.6 percent. But once the All-Star break happened, he kicked it into overdrive, culminating in a September where he pitched 43 innings and allowed four earned runs. Fun little thing about those four runs, all four came on three total home runs, two in one game against he White Sox.

But as for Carrasco, he doesn’t have the same abject change in his approach that Kluber did. Remember, Kluber happened to pack the best pitch in baseball last year, and he threw it a ton in August and September:

Normally, if a pitcher starts getting better and better, you tend to think they’re doing something different, whether obviously in Kluber’s case or maybe something more subtle. At least when it comes to pitch use, Carrasco is the same pitcher he’s been all year:

I did trim out his first start of the year and his most recent start, just because they skewed the month-by month view. Overall there’s some slight variation, but nothing as stark as when Kluber shot into space last season. And Carrasco isn’t quite on that trajectory either – maybe planetary orbit.

In an effort to try to explain this though, perhaps Baseball Savant can help illustrate changes Carrasco’s made. I picked an arbitrary time cutoff — July 20th, which also happens to be the beginning of the second half — and took a look. Their Pitch Heat Maps seem quite illustrative to me – while they don’t give you granular information, they paint a picture of how, in general, a pitcher is locating pitches. The general goal for a pitcher is to get it out of the center of the zone. Carrasco’s pre-break pitches create this picture:

Baseball Savant

And after the week off — because the voters rudely if rightly snubbed him — this picture was painted:

Baseball Savant

It doesn’t tell a whole story because no one stat or graph or anything does in baseball or life, but it speaks to some kind of trend. If you were to look at the Detailed Zone Charts — which I could have used but I find less pretty, so here’s a link to his pre- and post-break charts — you’d see a similar trend with the raw pitch numbers. Carrasco is simply grooving fewer pitches in the second half. Which is good. It lets him give up fewer hits, less damage, less runs, and so on. His release points haven’t moved much at all. How is he so effective?

That’s how we get to my favorite chart Baseball Savant makes, their Pitch Selection by Count charts. I find them intensely neat because pitch sequencing and that whole pitching meta-game the most fun of the brain wars going on in baseball every day. Before the break, here's how Carrasco attacked hitters:

As you might expect, he features the fastball early and in hitter’s counts, and once he gets ahead he goes off-speed. Makes sense. Then we get to the second half:

Again, there’s no obvious change. But of note is a relative movement toward throwing non-fastballs in all counts, along with a marked movement towards the curve in pitcher’s counts and some much broader mixing in hitter’s counts.

This doesn’t tell the story of how he’s not grooving pitches quite so much, except maybe he’s just not throwing quite as many fastballs so the pitch naturally finds its way out of the center of the zone. it’s a bit of a curiosity. Sometimes players just “do better”, maybe because they feel better, or are doing one small thing a bit different, or who knows what. The location mystery is a hard thing to parse. But it’s working.

I’ve written in the past about how Carrasco probably has the most complete arsenal on the Indians — or did, until Bauer decided to develop a top three or four changeup and top ten slider in the game. Carrasco’s value lies in pitch mixing, and it appears that’s where he’s gone in the second half, to considerable effect. More notably, not throwing fastballs — which he still packs a great one — any more than necessary, especially when he’s ahead. It’s the pitching that way when he’s behind that might have the most impact though. It’s not something that will show up in pitch use charts because a lot of things don’t change that much. It’s just about when he throws it.

As with most of his starts, it’s not a dramatic shift, it’s just consistent, steady movement toward a very good outcome. That it’s leading to his ending the season at his very pinnacle, that’s just icing on the cake.