Let's start with the obvious: James Shields is performing way way better than he did the last year. First consider the raw numbers:
James Shields hasn't just been greatly improved. He's also been one of the best pitchers in baseball this year. He ranks in the top 10 in the AL in K/9 (6th), BB/9 (7th), ERA (4th), FIP (8th), xFIP (1st), fWAR (5th), rWAR (5th), and IP/G (1st). Shields has truly been a force to be reckoned with this season. The obvious pair of things to consider after any meteoric rise are the sustainability and apparent causes for change.
Any mention of James Shields's performance, particularly his 2011 season, is remiss without mention of the "luck" stats (BABIP, LOB%, HR/FB%). These three stats are generally accepted as not entirely under the control of pitchers, and prone to wild fluctuation. Although Shields's career LOB% is roughly what would be expected, his career BABIP (.304 vs average of .300) and career HR/FB% (11.6 vs 10.5) are both slightly worse than league average, suggesting that defense-independent pitching stats might not be optimal for evaluating his true talent level. Still, Shields's 2010 was arguably one of the unluckiest seasons of all time. Although many still believed that Shields could return to the ace-like form that he demonstrated in 2008, an equal number thought that he was done as a pitcher.
Still, not even the greatest Shields supporter could have possibly seen this year coming. While Shields's BABIP (.259) and LOB% (86.4%) suggest favorable luck, he's striking out hitters at a far greater clip while managing to marginally improve in avoiding walks and inducing ground balls. There's no question that Shields is pitching better. The reasons for this are less than obvious, but we shall look at a few of them below.
- Pitch Selection: Shields's actual pitch f/x makeup and results have barely shifted. His three main pitches, the fastball, curveball, and changeup, all have extremely similar velocity and movements. He has, however, greatly increased the usage of his off-speed pitches (using them a combined 48.1% of the time versus last year's 38.5%), which, in increased usage, are still highly effective. The curveball has an identical whiff rate to last year's 11.3%, and the changeup is actually showing a slightly higher whiff rate, from last year's 21.3% to this year's 22.6%. As a result, his swinging strike rate (which is the biggest determinant of strikeout rate) has spiked from 9.5% to 11.4%, explaining the drastic increase in strikeouts. These two pitches also get by far the most ground balls of his arsenal, which is probably a factor in the very slight uptick in ground balls (although given the minuteness of the difference it could well be random variation).
- Leaguewide Decrease in Offense: I won't spend too much time on this because it's pretty obvious but offense is down leaguewide yet again. In the AL the average team has scored roughly 4.3 runs so far, versus 4.5 last year and 4.8 the year before that. The old adage is that the ball carries better in the summer and the numbers back it up, so as league average offense improves it'll be interesting to see how that impacts Shields.
- The Pickoffs: James Shields to date has picked off six runners this season. This may not seem like a huge impact, but essentially this turns a baserunner into an out, not only making him closer to finishing the inning but also removing a potential scoring threat. A runner picked off saves on average roughly .62 runs. The six runners Shields has picked off would then save roughly 3.7 runs total, which again may not seem like a lot, but actually accounts for a difference of an expected whopping .40 runs of ERA in his season so far. This is pretty significant.
- Regression: Simply put, Shields's HR/FB, LOB, and BABIP were not only way out of norm with league average pitchers, but also way out of norm with his established career marks. There's no way a pitcher pitching in Tropicana Field in front of the Rays defense could possibly have a true talent BABIP of .341, no matter how "hittable".
These four factors above appear to be the main ones lying behind Shields's new found success. Based on his peripherals, going forward it seems pretty fair to expect Shields to pitch like an ace. Despite his history of underperforming his peripherals, given his xFIP of 2.74 and the environment he's pitching in, an ERA in the high 2s going forward seems to be his true talent level.