This Week in Meatballs, Whomps per Whiff, and Other Novelty Stats

Hey there, and welcome to a segment that I’m hoping to turn into a recurring feature. Last week, I started delving into the individual event-level predictions built into our pitch grading model, PitchingBot. I made some broad generalizations about the kinds of pitches most likely to be hit for home runs and then looked at which pitchers threw them most often. I gathered some information about those pitches (fastballs, poorly located, in hitter-friendly counts but not 3-0), and tried to figure out what that meant for home run rate.

More specifically, it’s fun to look at these bad pitches, and it’s fun for me to see how few of them actually result in homers. The 50 pitches most likely to be hit for a home run surrendered one homer combined. The top 100 resulted in only three homers, while the next 100 resulted in six homers. There’s a ton of variability, but at its core, baseball is still a game of failure – even when a pitcher does the worst thing they possibly can, hitters mostly don’t punish them.

To that end, I’m going to try a new weekly roundup: various meatball-related items that show who’s been exceptional in one direction or another over the past week. Given that it’s mostly a list of things without a ton of analysis necessary, I’m going to start out by trying to add it on top of my normal schedule, and I’ll also use this to update some of my favorite junk stats (whomps per whiff, Kimbrels, etc.). If it’s popular, great! If not, hey, it’s a long baseball season and you have to keep trying things. Anyway, let’s get going.

Meatball of the Week

We have to stop meeting like this, Randy. When I crowned a single pitch the meatball of the year last week, Randy Arozarena was the batter who received that most-smashable offering, and he took it. This week, he hacked at Beau Brieske’s 3-1 offering, but could only foul it off.

A quick check on why the model thought that pitch was so smashable: a combination of count, velocity, location, and shape. Brieske’s fastball is a bit better than average, but this particular one wasn’t. For whatever reason – maybe he was guiding it in thanks to the count, maybe he just didn’t execute his pitch – it was 3.5 mph slower than his average fastball. It came during a hitter’s count and was thrown to a hittable location without a weird or remarkable shape. That’s the recipe for a crushable pitch, even if Arozarena couldn’t do anything with it.

Italian Grandmother Award for Most Meatballs Served

Valente Bellozo, Miami Marlins

This category is simple: Every week, the Italian Grandmother Award goes to the player who threw the most meatballs over the previous seven days. Valente Bellozo is going to be a regular contender for this honor because he throws a fastball that tops out in the lower 90s and averages 89.7 mph. He throws that fastball quite frequently, and gets by on soft contact thanks to his impeccable command. However, a lot of his fastballs also look like this:

Bellozo has been excellent in his rookie year; through six starts, he has a 2.45 ERA and 3.53 FIP. But the good times probably won’t last. He just got finished posting a 5.87 FIP in Triple-A while allowing 1.82 homers per nine innings. He gave up three warning track blasts to center field and a homer this week while striking out eight over two starts. The Marlins are desperate for pitching at the moment and Bellozo is succeeding, so I’m sure he’ll get tons of rope. But if you’re looking for the most likely pitcher to have a true blow-up start, he’d be my pick right now.

Healthy Diet Award for Most Pitches Without a Meatball

Pablo López, Minnesota Twins

Nothing really to see here. Pablo López made two starts and didn’t throw a single pitch that PitchingBot thought was 3% likely to turn into a home run. He did give one up, but it was to Bobby Witt Jr., and sometimes you just get beaten by a superstar. Honorable mention to Logan Webb, who made only one start last week (104 pitches) and continued to stymie hitters. He’ll feature later on the season-long update; for now, let’s just say that Webb doesn’t make many mistakes.

Now, a tiny bit of research to further explain why I’m looking at these pitches. First, the meatballs this week got crushed. There were 502 pitches that crossed the 3% homer likelihood threshold, and batters hit 21 home runs against them, 4.2% of the time. The model would have predicted 18 homers, so we’re running slightly hot. Pretty much any way you slice it, this week’s homers outstripped their meatball probability, which might be a product of the time of year as much as anything else. I’ll keep an eye on that as the season wears on.

It seems pretty clear that meatball rates don’t correlate overly well with single-game results. Bellozo was great in his two starts last week despite lobbing 80-handle fastballs in there. George Kirby didn’t throw a single pitch that the model categorized as a meatball, and he gave up four homers and eight earned runs.

Season-Long Meatball Standouts

Among pitchers who have thrown 1,000 total pitches in the majors this year, Joey Estes stands alone with 80 meatballs on only 1,369 pitches, a 5.8% rate. Webb is on the other end of the spectrum with only five meatballs out of 2,543 pitches, and believe it or not, Blake Snell is second with only three meatballs in 1,188 pitches. Snell’s more of a walks guy than a homers guy, so even with his abysmal start to the season, he’s giving up only 0.71 home runs per nine innings.

Whomps per Whiff Leaders

This statistic does a pretty good job of highlighting great hitters, ones who combine power with a keen sense of the strike zone. It’s also fairly results-oriented; you can’t get on the list without barreling up a ton of balls, and if you’re hitting a ton of barrels, you’re almost certainly hitting for power. So uh, yeah, this list is made up of all good hitters having good seasons bar one guy:

Stephenson hasn’t been a disaster or anything, he’s just not on anyone else here’s level. Overall, posting a high whomps per whiff means you’re a good hitter. Here’s the entire leaderboard for your perusal.

Kimbrels Leaderboard

A Kimbrel is an appearance with a negative FIP; in other words, an appearance when the pitcher strikes out at least 1.5 batters per inning without allowing any walks, HBPs, or home runs. It’s pretty close to just being a list of good relievers, but the top three are surprising:

Like WPW, Kimbrels mostly tell you what you already know: These pitchers are good. Mason Miller is obviously on there. Emmanuel Clase isn’t far outside the top 10, and the stat undersells Clase because he also allows extremely weak contact. Here’s the whole list if you feel like browsing it.

That’s this week in meatballs, whomps per whiff, and Kimbrels. I hope you enjoyed this semi-rigorous, GIF-heavy, and information-dense roundup of the weird statistics in baseball I’m keeping an eye on.