The good news is that Wade Davis, despite battling back and shoulder woes, has still been very good over the past month, allowing just one run in 12 innings. Kelvin Herrera hasn't been all that effective, but he has a good short-term excuse in the chicken pox diagnosis that interrupted his season, and it shouldn't be a concern in October. And deposed starter Jeremy Guthrie has allowed four homers in just 10 2/3 relief innings, which hardly reflects on the regular relief cast and crew.
The bad news? It may be time to worry about closer Greg Holland.
While you might accurately note that Holland's 2.45 ERA/1.78 FIP over the past month hardly seem like concerns, we'll file this under "a picture is worth a thousand words," because here's Holland's velocity trend since May 1…
… and his nearly identical four-seam fastball spin rate, per Statcast™:
We know that high spin for a fastball correlates with swinging strikes, so that spin rate decline is just as alarming as the velo drop. As you'd expect, the decline we're seeing has led to fewer missed bats. Through Aug. 14, which is the last peak on those graphs, Holland's strikeout percentage was 27.3. Since then, it's down to 20.0 percent.
What makes this more worrisome is that after pitching on back-to-back days on Aug. 27 and 28, Holland received a full 10 days off to rest what manager Ned Yost referred to as a "cranky arm," hoping the time off would help resolve the problem. Not only did his velocity not return, it continued to decrease.
Holland's four-seamer -- which had averaged around 96 mph in each of the previous three seasons -- was measured at 93.15 and 93.35 mph on Aug. 27 and 28, respectively, and was down to 91.25 and 90.31 mph in his two appearances after the layoff. Since his 2010 debut, Holland has appeared in 306 games, and those are the two lowest average game velocities he has ever registered.
Perhaps aware that his fastball isn't what it used to be, Holland has been using it less, in favor of his slider:
Through Aug. 14: 48.5 percent fastballs, 44.9 percent sliders
Since Aug. 15: 38.2 percent fastballs, 57.3 percent sliders
In and of itself, that's not necessarily a bad thing, because sliders often get more strikeouts. However, Holland's slider has seen a velocity drop as well -- down to 83.18 mph in September after being in the 85 mph range all year -- and has been less effective as well. He has allowed a .536 OPS against his slider, which is fine, except that it produced a .379 mark last year and .292 the year before. That's a problem considering Holland's fastball is more hittable than ever -- 92.1 percent contact rate on pitches in the zone after 84.9 and 78.3 the past two years -- and since he has a career-worst 12.9 percent walk rate.
Though Holland has only been the full-time closer for just more than three years, he is older than you think. He turns 30 this fall, and relievers tend to be so volatile that very few remain elite for a long period. As the Royals try to figure out just what's wrong with Johnny Cueto, they have an issue in the bullpen, too. Davis is a more than capable alternative, but the domino effect of moving him to the ninth would weaken the bullpen down the line. For a team that's been all but assured of a postseason trip for weeks, Kansas City suddenly has some questions to address.