Before Rich Hill was widely accepted as baseball's version of a phoenix -- rising from the ashes of injury and ineffectiveness to become a strikeout wizard and a valuable commodity -- he was a pitcher doing something weird.
Breaking from baseball convention, Hill threw his curveball nearly as often as his fastball -- more than 40 percent of the time -- and ripped off a shockingly effective run of starts to close 2015. Now, with about half a season of new Rich Hill to vouch for the potential viability of the approach, it may be worth examining another starter giving an offspeed pitch an outsized role: Matt Shoemaker.
The Angels starter came out of nowhere to finish second in the voting for the 2014 AL Rookie of the Year Award, but he disappointed in '15. After a rocky start to start '16, Shoemaker made a change. He's using his splitter nearly half the time and de-emphasizing a battered four-seam fastball.
The early returns -- 22 2/3 innings pitched, 31 strikeouts and no walks over his past three starts -- are striking.
Prior to this season, Shoemaker threw his splitter about 22 percent of the time, emulating Japanese imports Hisashi Iwakuma and Masahiro Tanaka (whom Shoemaker happened to be slated against Monday in the Bronx). But in May, Shoemaker turned to a version of the tactic behind Hill's star turn, ramping up his splitter usage to a level that would be unprecedented for a starter.
According to FanGraphs, no starter in the PitchF/X era (going back to 2008) has thrown a splitter more than 35 percent of the time in a full season, much less 40 percent of the time. But norms are being challenged across the spectrum of offspeed offerings.
Hill, Padres starter Drew Pomeranz and Mariners starter Nathan Karns are on pace to throw their curveballs more often than any qualified starter of the past decade. Tyson Ross threw his slider more than 40 percent of the time in 2014-15, finding success despite a relatively high walk rate.
Shoemaker's move seems logical. Hitters have swung and missed at nearly 22 percent of the splitters he's thrown in his career. Over his past three starts, Shoemaker has shunned a four-seam fastball that opposing batters are slugging .544 against this season.

And it seems to be working. After giving up seven homers in his first seven starts, Shoemaker has avoided the long ball in each of his those past three outings -- reducing his 40.8 percent fly-ball percentage from 2014-15 (among the 20 highest for pitchers with 200-plus innings) to 34.1 percent in 2016.
More encouraging, Shoemaker's 13.8 swinging-strike percentage ranks sixth in baseball (minimum 50 innings), and all of his pitches are generating career-high whiff rates.
The question, then, becomes whether this formula is sustainable. The answer likely depends on whether Shoemaker can continue inducing swings on the splitters. Hill's success has been predicated on throwing his curveball for strikes, creating unfavorable counts and then capitalizing.
Splitters are typically thrown low, inducing weak grounders or swinging strikes by diving out of the zone. According to Statcast™, Shoemaker's splitter has the second-lowest average spin rate of any changeup thrown more than 100 times in 2016 -- an attribute that usually corresponds with downward movement.
So Shoemaker has focused on getting ahead in the count. His 70.4 first-pitch strike rate is tied for first among pitchers with more than 50 innings, and he's using the splitter in that endeavor -- throwing it twice as often on first pitches.
Sure, it's only a few starts, but for an Angels team struggling with injuries, Shoemaker's recent run is a breath of fresh air.