At the end of the 2017 season, the Chicago Cubs bullpen ranked around the middle of the pack in terms of WAR (4.1, 14th), and probably projects to finish around the same spot in 2018. Of course, we’ll have to wait until more projections are released to really dig in, but there’s at least some hope they could do much better.
This offseason, the Cubs have lost Wade Davis (Rockies) and Hector Rondon (Astros) to free agency, before signing Brandon Morrow and Steve Cishek to replace them at the back end of the pen. Among the near-losses, however, was the best bullpen surprise of 2017, Brian Duensing.
On January 17th, the Cubs announced that Duensing would be returning on a two-year, $7 million deal and I was sufficiently enthused.
Out of 155 qualified relievers last season, Brian Duensing's …
0.7 WAR ranked 65th
3.39 K/BB, 59th
3.41 FIP, 54th
48.6 GB%, 50th
27.3 Hard-hit, 45th
2.74 ERA, 34th
This is a quality signing for just two years and $7M (especially with that low AAV for luxury tax reasons).
— Michael Cerami (@Michael_Cerami) January 17, 2018
After Duensing’s re-signing, Brett explored the downstream impacts his return to the bullpen could have on the rest of the roster, and they were both positive and decidedly non-zero – particularly with respect to the ever-present vacancy at the back of the rotation:
With Duensing in the fold, the Cubs not only now have a much-more-solidified bullpen, but they’ve got an additional full-inning guy, who also happens to be a lefty. In a world where Montgomery is in the rotation, having another full-inning lefty in the bullpen could be critically important. And now the Cubs have that, meaning that the Duensing bullpen signing is actually something of a nice hedge against the risk that the Cubs don’t sign another starting pitcher before this offseason is up.
Clearly, we like the guy and the signing.
But as I was considering his 2017 season, which was, inarguably, very good, I was wondering just how much better Duensing’s performance really was than his previous seasons, and if that’s going to be repeatable in 2018. So let’s take a little peak behind the curtain and see what was so different last year.
NOTE: Duensing became a full-time reliever in 2013 (he was previously a starter with the Twins), so all of the following data/analysis is dating back to that season.
As a reliever last season, some of Duensing’s biggest statistical achievements were due to marginal improvements in his batted ball data. For example, he was able to drop his career 22.7% line-drive rate (well above average, in a bad way) into a much stronger 17.9% rate (solidly below average, in a good way). Similarly, he was able to pop up his ground ball rate from a middling 44.1% for his career to a solid 48.6%.
And while his fly ball rate remained mostly flat, his infield fly ball rate dropped off quite a bit. In fact, that might be one way he can subtly improve at the margins based on naturally regression alone. But we’ll put a pin in it for now, because it’s at least plausible that whatever he’s done to get more grounders is leading to fewer infield pop-ups.
Relatedly, Duensing’s contact-management was pretty flat (a little less soft contact, a little more hard contact, but generally better than average there) vis a vis his career, so we’ll slip past that area, too.
Looking at a more traditional area, we see one of the most notable differences between his 2017 season and the four years of relief that preceded it: tremendous changes in his K/BB ratio.
2013-2016: 1.86 K/BB
2017 season: 3.39 K/BB
While Duensing cut down on his walk rate by quite a bit (8.7% to 7.0%), the rise in strikeouts really stands out: 16.7% to 23.7%. Let’s see if we can find out why.
According to his plate discipline data from 2017, Duensing was getting batters to chase pitches out of the zone less often than usual, which was a bit unexpected. They are swinging at pitches in the zone less often than his career, too, though, so it seems like he may be catching batters off guard by quite a bit, and that’s leading to more indecision at the plate (overall swing rate is down), which, in turn, is leading to more strikeouts.
In addition, batters used to be making contact with 69.8% of pitches Duensing threw out of the zone, but in 2017 that number dropped considerably (60.8%)! With a higher swinging strike rate overall, it seems Duensing did, indeed, find a way to induce more whiffs (contact rate is down from 80.8% to 76.9%).
These aren’t league-leading numbers, mind you, but clearly there was some pretty significant improvement. And with that in mind, I suspect there might’ve been a change in his pitch-mix …
2013-2016 Pitch Mix:
2017 Pitch Mix:
Four-seam: 28.1% (+4.3%)
Sinker: 20.7% (-11.8%)
Change-up: 13.2% (+2.2%)
Slider: 24.5% (+2.1%)
Curveball: 13.3% (+7%)
First of all, with a pitch mix this robust and evenly utilized, it’s no shocker the Cubs wanted to keep Duensing around. He’s got a full starter’s arsenal and can pivot to any one pitch as needed/appropriate.
But more to the point … wow! There are some notable differences, eh?
Last season, his first with the Cubs, Duensing dramatically reduced his reliance on the sinker, and started leaning more heavily on his four-seamer … which he also happened to be throwing nearly a mile per hour harder. In conjunction, he stepped up the usage of his change-up, which was now showing a greater velocity gap in relation to his fastball. Wonderfully, that new combination of pitches made each of those three more effective!
Change-up (’13-’16): 2.3
Change-up (2017): 3.1
Four-seamer (’13-’16): -1.8
Four-seamer (2017): -0.6
Sinker (’13-’16): 3.6
Sinker (2017): 6.4
Duensing saw a similar rise in productivity out of his slider, while his curveball actually ranked a bit worse than the previous four seasons (you might expect him to steer away from that a bit next season). It’s actually not all that uncommon to see a different pitch mix lead to increased value of other pitches, even ones that are used less. No one pitch is thrown in a vacuum, so the way they play off each other is important (as you can see) – consider the massive improvement in the value of Duensing’s sinker compared to the 11.8 percentage point drop in usage. That’s crazy!
So where does this all leave us?
Well, in a very good place. We now know that Duensing’s improved results last season, on a peripheral level, were probably due to an increase in strikeouts and ground balls, and a decrease in walks and line drives. But instead of those peripherals coming out of nowhere, we can say that some added fastball velocity and a new pitch mix (which kept more batters off balance than in the past) may have been the root cause.
So as long as Duensing can keep his velocity up and continue to adjust his mix to stay a step ahead of hitters, he should be able to enjoy another successful season out of the Cubs’ bullpen.