2016’s Best Pitches by Results

While the 2016 campaign is over and the flurry of moves after the season has come to a halt for the moment, a whole year’s worth of data remains to be examined. Today’s post is an easy one and a fun one. Let’s find the best pitches that were thrown regularly last year.

Before we begin: the word “results” appears in the headline, but I’m not going to use results judged by things like singles and doubles and the like. The samples gets pretty small if you chop up the ball-in-play numbers on a single pitch, and defense exerts too much of an influence on those numbers. So “results” here denotes not hit types, but rather whiffs and grounders.

I’ve grouped all the pitches thrown last year, minimum 75 for non-fastballs, 100 for fastballs. I combined knuckle and regular curves, and put split-fingers in with the changeups. So the sample per pitch type is generally around 300 — a lot less for cutters (89) and a bunch more for four seamers (500) — but generally around 300 pitches qualified in each category. Then I found the z-scores for the whiff and ground-ball rates on those pitches. I multiplied the whiff rate z-score by two before adding it to the ground-ball rate because I generally found correlations that were twice as strong between whiff rates and overall numbers like ERA and SIERA than they were for ground-ball rates.

The caveats are obvious. Pitches work in tandem, so you may get a whiff on your changeup because your fastball is so devastating. This doesn’t reward called strikes as much as swinging strikes, so it’s not a great measure for command. On the other hand, there isn’t a great measure for command. By using ground-ball rate instead of launch-angle allowed, we’re using some ball-in-play data and maybe not the best ball-in-play data.

But average-launch-angle allowed is problematic in its own way, and ground-ball rate is actually one of the best ball-in-play stats we have — it’s very sticky year to year and becomes meaningful very quickly. Whiff rates are super sexy, since a swing and a miss represents a clear victory for the pitchers over the batter — and also because there’s no room for scorer error or bias in the numbers. And while the precise way in which pitches work in tandem remains obscure in pitching analysis, we can still learn something from splitting the pitches up into their own buckets.

First, the overall top 15. It’s all relievers, of course, because relievers can come in throwing as hard as they can, they can use only their very best pitchers, and they can be leveraged to face only the batters who are predisposed to struggle against them. In all other ways, relieving is the toughest job in baseball, but in this one way, they get an advantage.

For the second year in a row, Zach Britton’s sinker is the best pitch in baseball. It’s an awesome pitch, matched only in velocity and sink by the sinkers thrown from the right arms of Sam Dyson and Blake Treinen. Remarkable, given how differently the three pitchers find that movement, but still a rare combination of gas and movement. There are 14 sinkers that get double-digit whiff rates, and only one other sinker that gets more than 15% whiffs, so Britton stands alone.

That way in which pitches might work in tandem is relevant to the second name on the list. Sergio Romo’s slider is good but not great these days (only slightly above average by whiffs at 17%), but it’s in everyone’s head because he throws it so often. In certain counts, though, Romo is just as likely to throw the sinker as any pitch: in 1-1 counts last year, batters actually saw the four-seam and two-seam as often as the slider. That would be surprising, considering Romo led baseball in slider percentage last year.

Then you get the gas. Michael Lorenzen, Aroldis Chapman, and Jeurys Familia throw the ball hard. It’s what they do. Throw in Alex Colome on that mini-list — and Carl Edwards, Jr, too, although he also adds great spin to the picture. These are the pitches you expect to see on this list, though even here you learn that Lorenzen has a better fastball than you might expect.

The phrase “perennial surprise” is an oxymoron, but it characterizes Darren O’Day’s appearance on this list every year. His fastball lost half a tick of gas, and half an inch of rise, so it dropped a few rungs from eighth to 13th last year. It’s still a remarkable pitch, given the deception he can provide by sometimes throwing a rising fastball from that arm slot.

Luke Gregerson’s slider will forever keep him in the big leagues, it seems, but now it doesn’t get the most whiffs on his own team! Ken Giles is coming for the crown. Newcomer to this list and our hearts, Edwin Diaz throws impressive heat, but it’s the slider that makes him a star. That might be relevant to any discussion of his future as a starter? If it was all about the fastball and he was destined to lose gas by moving to the rotation, you might worry. This way, you know he has an elite slider, and the question of starting can return to being one about his changeup and his health and mechanics.

Last on the list is a guy who shows us why we do this sort of thing: Josh Osich. The 28-year-old had a forgettable year in the Giants pen, at least when you look at traditional results. He barely struck out more batters than he walked, and added a home-run problem to boot. However! He was a top-five ground-ball guy among relievers with at least 30 innings, and that was because of his excellent 96 mph sinker, which has top-10 movement in both directions among sinkers that averaged over 94 (66 total). After offseason surgery on his knee, maybe his command will improve, and his pitch will shine. At the very least, there’s more going on with Josh Osich than a near-five ERA.

We’ll get to the starters next, but before you get to a starter on this list, you see Gavin Floyd‘s curve, Ryan Tepera‘s cutter, and Felipe Rivero’s changeup. So, yeah, it makes sense to separate the starters and the relievers. Just like it makes a little sense to separate out each pitch type.