James Shields stats look pretty good. He’s 8-3 with a 3.70 ERA. But, Shields’ season numbers have been skewed by his 2.03 ERA in April. In May and June, Shields posted ERA numbers of 4.69 and 4.81.
That’s hardly the performance of an ace.
So, what’s happened to James Shields the last two months?
My first thought was to look at his average velocity. A drop in velocity is almost always associated with a loss in effectiveness. It can also signal a pitcher trying to throw through arm problems. Yet, that doesn’t seem to apply to Shields. In fact, as you can see below, his average velocity has trended up on all of his pitches since the beginning of the season.
While Shields velocity seems fine, his pitches are far less effective than they were in April. In the season’s first month, Shields whiffed opposing hitters 25.5% of the time vs. 16.3% in May and 14.0% in June.
Throughout his career, Shields’ out pitch has been his outstanding change-up. As we can see below, Shields’ has suffered a slight down-tick on his swing and miss percentage on his out pitch, from 19.13% in April, to 17.99% in May, and 17.65% in June. Yet, that drop-off doesn’t seem to be enough to explain his slide in performance. But we also see a slide in whiff percentage on his curve, and a slight increase on his cutter. Overall, Shields’ isn’t getting as many swing-and-misses on his breaking stuff.
However, if we look at batted balls in play (BABIP) we see that in March/April Shields had a very low .240 mark, which has zoomed to .341 in May and .349 in June. That works out to a BABIP of .310 for the year. That’s significantly higher than his career .299 mark, which suggests Shields has been getting a little unlucky with regard to hits falling the last two months.
That his line drive percentage has held steady over the last 3 months suggests that the above analysis is correct—though opponents slugging percentage against Shields is increasing as the months pass: .318 in April, .478 in May, and .573 in June.
To check if players were squaring up Shields more the last two months, I used Baseballheatmaps.com to calculate the average fly ball distance vs. Shields in May and June and compared it to his number for April. Interestingly enough, his average fly ball distance declined to 271.33 in his last two bad months from 281.13 in his dominating April.
That fact, combined with his steady line drives against rate, suggests to me that the “bad luck” with BABIP might be what’s going on.
Let’s hope I’m right.