The Curious Case of the Curveball in the Nighttime

Monday night, Michael Wacha made a cathartic first start with the Mets. Over five solid innings, he struck out four while allowing only one run on a Mitch Moreland solo shot. He walked away with the win, his first in more than a year, and gave Mets fans hope that they might cheat the injury gods and assemble an acceptable rotation. But wait! Michael Wacha was last seen being terrible. It’s time to do some digging. The game is afoot!

We start this investigation, like so many others of sudden pitching competence, with the fastball. But alas, there’s nothing to be gleaned from it. Wacha averaged 94.3 mph on the pitch, a hair higher than last year’s season-long average but only a hair higher than last July’s mark. Allowing for the fact that the switch to Hawkeye might come with some calibration errors, we can rule out a newly lively fastball accounting for the fact that the Red Sox looked flummoxed.

Or can we? Why not spiral deeper, hunt further for fastball clues? His spin rate is up by nearly 150 rpm. Mayhap that’s the culprit. Mayhap indeed — but in my opinion, it’s not likely. Spin is one of the things to be most skeptical about in the new system. Perhaps skeptical is the wrong word; maybe we should be skeptical of the old measurements. The Hawkeye system measures spin directly with high-speed cameras, while the old radar-based system imputed spin from other factors. The point is, spin is going to be a tricky thing to tackle for a good while.

In every other respect, Wacha’s fastball was uninteresting. It had less ride than 2019’s edition, missed fewer bats, and was overall a subpar offering. Moreland’s homer was on a first-pitch fastball that Wacha wasn’t fine enough with:

In any case, digging too much into the fastball misses the point. Results are still treacherous this early in the season, and the raw building blocks of the pitch don’t seem meaningfully different. Onward.

What about Wacha’s cutter, a pitch he used extensively in lieu of fastballs? It has a slightly different shape this year; less ride, and therefore more vertical drop, almost a hard slider. In some ways, it’s a pale facsimile of teammate Jacob deGrom’s slider — four miles an hour slower and with less break, but the same general idea.

Still being mindful of the small sample, the cutter looked perfectly adequate on Monday. It generated two whiffs, which should make the tiny sample sizes we’re dealing with evident. The pitch avoided loud contact, always a plus, and was generally inoffensive. Most importantly, Wacha used it far more than he had in previous seasons, 36% of the time. Not only is that higher than any previous season high, it’s a higher rate than he’s ever thrown in a game before.

The key to that is that more cutters means fewer fastballs, and Wacha’s fastball simply doesn’t work as a primary pitch at the major league level. This isn’t a new observation — it was already clear in the numbers. The last time he had a positive run value on the pitch was 2017, when it sat 95-96 instead of 93-94, with an extra two inches of ride. That pitch generated whiffs on 20% of swings, nearly double the rate of his last two-plus seasons. Short of rediscovering his old form with the pitch, burying it deep in his arsenal and using it only when necessary seems to be the best option.

To pad the rest of his arsenal, Wacha needed something slower. The cutter isn’t even that good, either. It’s just a different, and likely less exploitable, look. Luckily, his best pitch is a changeup, and he rode it in Boston, throwing it a full 35% of the time. That pitch is still lethal; it sneaks in on right-handers with deceptive arm-side break and tantalizes lefties before absolutely vanishing. Thanks for trying, Andrew Benintendi:

Better luck next time, J.D. Martinez:

It is, again, a small sample, but batters came up empty on 43% of their swings against Wacha’s changeup. That’s not a career high for a single game by any means, and it’s not a career-high usage rate either. Wacha’s changeup has always been great, and throwing it a third of the time is an absolute no-brainer.

The weird part of all of this is that none of these pitches have really changed from last year. Bad fastball, okay cutter, great changeup; none of it has changed. The window dressing of swapping four-seamers and cutters is nice, but it’s hard to imagine that could move him from a pitcher who looked broken by the tail end of last season to a credible third starter type. Should we just write this off to a single start early in a season where pitchers seem to have an early edge?

I have one last piece of evidence to introduce that I think will shed some light on my newfound Wacha enthusiasm. Like the Sherlock Holmes story I clunkily referenced in the title, the key change Monday wasn’t something we saw, but rather something we didn’t see. Or, okay fine, saw exactly once:

Ahh! Cover your children’s eyes! That pitch belongs nowhere near a major league stadium, and it’s the only curveball that Wacha threw all day. It at least had the benefit of being so bad that the worst that could happen was a ball, moving the count from 0-0 to 1-0. That’s bad! He shouldn’t throw that pitch! But the damage ended there, because he simply didn’t throw any others.

Wacha’s curve is bad. He’s a practitioner of the first-pitch hook that pitchers with underqualified breaking pitches often lean on, and opposing batters weren’t kind to that poor pitch. Sometimes he stole a strike, or missed the zone. When he threw it in the strike zone and batters swung at it, however, this happened far too often:

Two strikes. Lefty batter. There exists no better situation for a changeup, and yet Wacha went to the curveball. Sometimes merely having a curveball makes you want to use it, and that was almost always a bad instinct for Wacha.

At the end of the day, one start isn’t enough to say many things. You can see new velocity, or a new pitch. You can see a different motion, or a new location on the mound. The actual results, however, need to be overwhelming to be compelling, and Wacha’s start was certainly not that.

But absence can be evidence. Not doing something is a choice, one that seems likely to pay off in this case. Ten percent of Wacha’s pitches in 2019 were curveballs. That was only 227 pitches, but he managed to be a whopping 5.3 runs worse than average over those pitches, between loud contact and a minuscule chase rate. Replace those with an even mix of the rest of his pitches, and he would have saved roughly five runs.

That might not sound like much, but if you do some gorilla math and just lop those runs off of his ERA, it would have lowered it by nearly 0.4. It probably won’t be that easy or that straightforward, but imagine learning one simple trick that could lower your ERA by almost half a point. You’d put Driveline to shame in days, pitchers knocking down your door for the cheat code that could get their careers over the hump from okay to good. Wacha seems to have found that.

The lone curve this year is only the second time he’s thrown one curveball or fewer in a game since the beginning of 2017. The one other time? His final start of the season last year, which is a promising sign. It’s not rocket science. Throw it less, and profit. Next time you watch a Wacha start, watch carefully for what you don’t see. It might be the difference between a valuable pitcher and a replacement level clunker.