Mid-season Arsenal Scores

About a month ago, I started searching for the league’s underthrown pitches. Pitches that despite inducing elite swing-and-miss, ground ball, and pop-up rates, are thrown with scarcity relative to other inferior offerings. It was as an enlightening a topic to research, as it dealt with untapped potential, as it was a fun series to write. Though to be honest, there wasn’t a whole lot of actionable fantasy advice to be gleamed. But in the process of writing those pieces, I had to grade each pitch. And arsenal scores, a subject of interest to the RotoGraphs community over the years, were just a stone’s throw from away from the groundwork I’d already laid.

The pitch grades I used were comprised of three components converted into Z-Scores: whiffs, ground balls plus pop-ups, and called strike-to-ball ratio. To each Z-Score, I applied weights (below) based off a linear model and summed them to arrive at an overall Arsenal Score. I’ve provided two sets of results, one in which each pitch is weighted evenly and another that weights each offering based on frequency thrown. I’ve pretty much taken the work that Eno and Alex had done previously, added an element of control, and refined the weighting based on a correlation between the aforementioned components and Baseball Prospectus’ True Average.

While there are strong correlations between the outcomes evaluated (whiffs, ground balls, pop-ups, and called strikes), there is still a significant amount of the variation left unaccounted for. Perhaps, not unrelated, we’re still at the point where we depend on outcomes to evaluate pitches, relying on hitters’ decisions and talents instead of an offering’s innate qualities alone.  These are some of the drawbacks with this methodology.

One other thing before we look into a few of the interesting names to make the list: it’s too exhausting to keep scraping the PitchF/x leaderboards for each article so this data is, admittedly, a week old. The table below doesn’t account for Clayton Kershaw’s recent outing against the Diamondbacks during which he struck out 11 nor does it include Jon Gray’s latest turns since his activation, though in fairness Gray still would fall short of the 50 pitch threshold on half his repertoire.

Still, in the never-ending quest to separate talent from results, this is a helpful exercise and one that I hope you can find use for.

Noteworthy Arsenals

Alex Wood – at some point I need to stop being surprised by Alex Wood. Since the start of last season, he’s thrown 141 innings with a 2.55 ERA, 2.53 FIP, and 2.88 xFIP on the back of funky deception and three above average pitches. This season, Wood features the league’s 3^rd best sinker, 17^th best curve, and 25^th best changeup.

Always one to generate ground balls, he’s now doing so at a near league-leading clip. And the whiffs? Well, he’s first on the Dodgers in strikeout rate, which should tell you all you need to know. And going back one year, Wood’s K-BB% ranks 10^th in the league among starters who’ve thrown as many innings. As long as he stays healthy, Alex Wood is as good as any.

Robbie Ray – I have to admit, I’m still not aboard the Robbie Ray Train. I’ve never been able to get past his command and homer problems. But to be fair, neither has he. Ray is walking more batters and inducing fewer ground balls than ever before. And there doesn’t there seem to be an easy fix tinkering with his pitch mix either. His two most underthrown pitches are his four-seamer, which is merely average at generating grounders, and his curveball which he actually throws fairly often; it’s considered underthrown not because he throws it less frequently than the average pitcher does but because it’s so effective, he should throw it as often as possible.

Perhaps Ray has found a way to manage elevated contact but with an average Fly Ball and Line Drive Exit Velocity in the 75^th percentile (in this case, high is bad), I’m not too eager to wait around to find out. Ray’s elite arsenal score is supported almost entirely by incredible swing-and-miss and stuff, which undeniably has value. And perhaps, in this inflated offensive environment, you can stomach an ERA more in line with his FIP or xFIP. Just don’t be surprised to see a sharp ERA correction in the second half. Ray should net a haul in return. Those with pitching to spare, and I realize there are just a handful of you, might consider selling high.

Lance Lynn – despite his placement on this list, I’d caution against buying Lynn. His fastball and sinker velocity dropped precipitously a few starts ago starts and though it returned in his last outing, he’s the unlucky protagonist in a long and storied injury history. So while those two games may prove nothing more than a blip, I’m still concerned enough to wait and see.

Jimmy Nelson – if you haven’t already, I’d suggest listening to this latest episode of The Sleeper and the Bust. Paul and Eno do a great job of analyzing changes to Nelson’s breaking pitches that explain his unforeseen success (sincere apologies to anyone who foresaw said success but I know you’re lying so, I forgive you). After being inexplicably cut in one of my Ottoneu leagues a few weeks ago, I’ve been riding him ever since. Nelson’s post-hype breakout has been one of the truly rewarding stories in this pitching-starved environment.

Clayton Kershaw – a commenter last week pointed out that Kershaw was too low on this list. How could I argue otherwise? I mean, I tried but I admit I felt silly doing so. This is one of the limitations with any purely quantitative framework – some results just don’t pass the smell test. That doesn’t mean your whole model sucks, just that you need to buttress your numbers with some common sense. But as I pointed out in my response, as good as Kershaw has been, he hasn’t been the same Kershaw to which we’ve grown accustomed. This isn’t a knock nor is it to say he won’t “turn things around.” And yes, I recognize how ridiculous I sound, hence the “air” quotes that actually aren’t “air” quotes because they appear on the screen in front of you.

But we’re looking at Kershaw’s lowest strikeout rate, K-BB%, and whiff rate since 2013 and his lowest ground ball rate since 2011. And the homers are alarming. Hitters are averaging 1.24 bombs per nine innings off Kershaw; not only is that significantly above the league average, it’s nearly 150% higher than his career rate through last season. Check out Jeff Sullivan’s piece on this very issue here.

Now, that doesn’t tell the whole tale. Obviously. When it comes to contact management, Kershaw is on the right side of average for most Statcast metrics and most everything else. Honestly, talking about Kershaw’s arsenal is a waste of time. He’s extraordinary. I single him out for no other reason than to illustrate the limitations of scoring arsenals based on just a handful of outcomes. But since you came here for fantasy advice, here it is: for anyone playing in the 1% of Yahoo! leagues where Clayton Kershaw is available, consider picking him up. I’d even burn a high waiver priority to do so.

Most Underowned Arsenal

Trevor Cahill is groin-grabbingly underowned. I’ve mentioned him here for his slider and again here for his entire arsenal. In his first start back from a shoulder injury, Cahill tossed 4.1 shutout innings, striking out four and inducing grounders on 82% of the balls Cleveland hitters put in play. Vintage 2017 Cahill. No, he didn’t dominate but he was likely on a strict pitch count and facing one of the league’s hottest lineups. Still available in 78% of Yahoo! leagues, PICK HIM UP.

On another note, this is my last article at RotoGraphs, at least for a while. I’m moving onto other things and as much as I love writing here, I can no longer put in the requisite time to churn out pieces capable of satiating the voracious analytic appetites of this site’s devoted and discerning readership. It’s been a blast and a truly cromulent honor to contribute to my favorite website for the past two years. Hopefully, I get to do so again soon. Thanks for the positive comments and for keeping me honest when I deserved it.