Alex Colome led Major League Baseball in saves with 47 in 2017. He did that despite seeing his strikeout rate drop from 31.4% in 2016 to 20.6% in 2017. While he had a 3.24 ERA, his xFIP ERA was all the way up to 4.32. I already viewed him as a risk because of his likelihood of being dealt, but when DVR and I were talking about closers on our SiriusXM show this week, he named him as one of the established closers that he's most worried about. He cited Colome's 11.9% swinging strike rate, which was 85th among all qualified relievers last year. As a point of reference, Craig Kimbrel led all relievers with a 19.8% swinging strike rate, followed by Kenley Jansen at 18.2%.
But why do we care about swinging strike rate (SwStr% on Fangraphs)? Hopefully we've advanced enough in fantasy baseball analysis to persuade you on the merits of strikeouts, particularly with closers, so I won't belabor that point. A pitcher's swinging strike rate is a measure of how often a batter swings-and-misses at his offerings. It's the best measure I know of that distills the very essence of the game - batter vs. pitcher. The pitcher isn't reliant upon his fielders, the weather conditions, and for the most part, his ballpark. The ballpark factor occasionally comes into play with this - Jered Weaver with his three-quarters delivery blending in with the fake rocks in Anaheim Stadium comes to mind as a notable exception. Even the importance of pitch-framing and umpiring is minimized - sure, if a pitcher is getting certain strikes called opposing batters might be more inclined to chase borderline pitches, ones that he'd be most likely to miss. But even that factor is indirect and in the case of the umpire, unlikely to repeat more than once or twice. So the pitcher that has a strong SwStr% is most likely to record more strikeouts and thus be an effective closer.
To that end, here's a list of this year's projected closers, ordered by their SwStr% last year, along with his rank on my tiers article and his NFBC ADP.
Closer | SwStr% | Rank | ADP |
Craig Kimbrel | 19.8% | 2 | 44 |
Kenley Jansen | 18.2% | 1 | 37 |
Roberto Osuna | 16.8% | 5 | 78 |
Hector Neris | 16.6% | 20 | 139 |
Ken Giles | 16.4% | 14 | 92 |
Edwin Diaz | 16.1% | 7 | 83 |
Brandon Morrow | 15.9% | 18 | 226 |
Luke Gregerson | 15.8% | 29 | 376 |
Felipe Rivero | 15.8% | 9 | 86 |
Wade Davis | 15.5% | 15 | 97 |
Sean Doolittle | 15.4% | 10 | 131 |
Greg Holland | 15.3% | 19 | 127 |
Cody Allen | 14.9% | 4 | 99 |
Arodys Vizcaino | 14.8% | 13 | 147 |
Corey Knebel | 14.1% | 8 | 60 |
Raisel Iglesias | 14.0% | 6 | 96 |
Blake Parker | 13.8% | 17 | 214 |
Aroldis Chapman | 13.6% | 3 | 64 |
Brad Hand | 13.3% | 11 | 117 |
Blake Treinen | 13.2% | 21 | 199 |
Fernando Rodney | 12.3% | 25 | 240 |
Brad Brach | 12.2% | 24 | 230 |
Alex Colome | 11.9% | 12 | 119 |
Kelvin Herrera | 11.5% | 28 | 208 |
Juan Minaya | 11.5% | 30 | 334 |
Jeurys Familia | 10.2% | 22 | 168 |
Archie Bradley | 10.0% | 16 | 187 |
Alex Claudio | 9.9% | 27 | 321 |
Mark Melancon | 9.6% | 23 | 171 |
Shane Greene | 8.7% | 26 | 239 |
Brad Ziegler | 8.5% | 31 | 390 |
I have a couple of reactions to the data. One, I probably should be more aggressive in adding Hector Neris as my second closer. There are a couple of risk factors, the biggest being his job security - it took the Phillies a long time to commit to him as the closer last year, and now he has a new manager in Gabe Kapler who is a bit of a wildcard, though we know he's a fan of incorporating analytics. Will that translate into not having a traditional closer set-up? I'm also a little alarmed at how low Aroldis Chapman rates. His 13.6% SwStr% represents a big drop from his previous five years. I might have to remove him from second tier of closers, especially when you factor in how deep the Yankees bullpen is. Meanwhile, I've already downgraded Colome since publishing my previous article, on the basis of both his trade risk as well as his performance risk.
Swinging strike rate isn't a super-stat, however. It doesn't account for managerial whim, previous injuries, or the elusive guile that one often needs to succeed as a closer. I'm also still learning some of the limitations of the stat - how repeatable is it, what is an acceptable level vs. just comparing relative comparisons, etc... But hopefully this snapshot is a starting point for you to continue investigating a few angles to use at your draft.