The Cleveland Indians have a true ace at the top of their rotation. This lefty has blown away opposing hitters this year, and he is inspiring hopes of bringing the Cy Young award to Cleveland for the second year in a row.

These two sentences do not seem very surprising, considering the Indians possess one of the best left-handed pitchers in the game, CC Sabathia. However, as Sabathia has struggled (side note: CC has been excellent in his previous two starts, and has begun throwing his slider once again), fellow lefty Cliff Lee has been fantastic.

Back in 2005, Cliff Lee was excellent. He pitched 202 innings and managed a fine 3.79 ERA. However, his xFIP (a measure of a pitcher’s ERA that adjusts for luck) of 4.56 and BABIP of .277 suggested that Lee had been the beneficiary of some good luck. Sure enough, in 2006 Lee’s ERA rose to 4.40. Last season Lee was even worse, posting a 6.29 ERA before getting himself demoted to Triple-A. This year, however, Lee has been legitimately fantastic: in addition to an incredible 0.96 ERA through 37 innings, Lee has a 32/2 K/BB ratio.

Lee has certainly benefited from some luck this season, as any pitcher posting an ERA under one almost always does. His BABIP is a measly .189 and he has given up homers on 2.8% of his fly balls. However, Lee’s xFIP is still only 2.88. While Lee has been lucky, he has also been one of the best pitchers in the American League so far. What is he doing differently?

The first thing that jumps out is the fact that Lee has walked two (!) hitters this season. This is impressive for any pitcher who has thrown 37 innings – it’s even more impressive for Lee, who has never been known for impeccable control. In his solid 2005 campaign, Lee walked 2.32 batters per nine – not bad, but nothing amazing. Lee’s walk rate rose to 2.60 per nine in 2006 and then up to 3.33 per nine last season. In eight starts last season with Triple-A Buffalo, Lee walked 25 in 41 innings.

Lee is throwing more strikes this year: 68% of his pitches are strikes this season (granted, in a small sample). His previous high was 66% back in his good 2005 season; he has been below that every other year. Interestingly, Lee is throwing far more pitches per at bat than in previous seasons: this year, he’s throwing 4.16 pitches to every hitter; as opposed to 3.72, 3.81, and 3.68 pitches per hitter over the last three seasons. He is getting more called strikes than in the past (30% this year, as compared to 28%, 27%, and 27%) and batters are making contract at a lesser rate this year (81% of all swings make contact) as compared to previous years (83%, 84%, 85%).

Additionally, a much higher percentage of his strikeouts have been called third strikes this year (38%, as compared to 32%, 28%, and 29% over the last three seasons). It appears that the quality of Lee’s “stuff” has not changed dramatically, but that his control (as evidenced by lack of walks) and command (as evidenced by called strikes) has improved. I do not know what to make of the fact that he’s throwing more pitches per at bat, except for perhaps that in previous seasons batters were given pitches to hit more often and thus at-bats ended sooner.

His pitch data shows that his fastball still averages around 90 MPH as it did last year. His curve is the same (75 MPH), and he is throwing both his slider and changeup slower this year (81 and 82 MPH) than last year (84 and 83 MPH). Interestingly, he is throwing his fastball far more often this season than he did last year: 85% of Lee’s pitches this year have been fastballs, while last year he threw his fastball 72% of the time. Last year he threw his changeup 16% of the time, while this year he’s used it only 4% of the time.

In addition to throwing it more often, Lee’s fastball appears to be different this year than it was last year. It has more horizontal movement this year (6.59 inches as compared to 3.83 inches last year) – what I guess we’d call “run” or “tail.” Lee is also getting more sink on his fastball this year; this would explain why he’s getting more grounders than fly balls for this first time in his career (in fact, his 45% ground ball rate is 10% higher than any other year in his career).

Perhaps most interesting of all is that Lee’s release point appears to be different this year than it was last year – especially when he throws his fastball. Open this and this in adjacent windows, scroll to the last chart (“release point”) in both windows, and flip back and forth. What do you see? Lee’s release point has clearly changed by at least several inches – especially when he throws his fastball.

To recap: although he has been lucky, Cliff Lee has also been legitimately fantastic this year. He is throwing his fastball more often and his changeup far less often; his fastball has more sink than in the past, and as a result his ground ball rate is higher than it has ever been. Lee’s release point – especially when he throws his fastball – is different than it was last year, perhaps resulting in the additional sink on his fastball and/or his increased control.

Although Lee has had an easy schedule so far (facing the Athletics twice, as well as the Mariners, Twins, and Royals once each), he only ranks 83^rd  (of 120 pitchers who have pitched at least 25 innings) in his opponents’ adjusted OPS – meaning that despite Lee’s easy schedule, 37 starting pitchers have faced worse hitters this season.

Cliff Lee won’t finish the year with an ERA under one – but it’s not a stretch to say that he can continue his dominant pitching.