The Mystery of Fausto Carmona

A year and a half after he first cracked the starting rotation, Fausto Carmona is still, in many ways, a baffling specimen to both Tribe fans and opposing batters.  As a young(er) pitcher, he was known more for his command than his velocity, posting a 3.43 K/BB ratio during his years in the Indians' farm system.  In 2006, when called up to the show, he became something of a power arm in the bullpen.  Of course, by now we all know the story of his failed stint as a closer at the end of that season and his subsequent rise to dominance in 2007, during which he posted a Halladay-esque 5.73 K/9 with a 3.28 GB/FB, making him the best groundballer in the AL.  His K numbers improved throughout the season (6.37 K/9 after AS break) as did his ERA, which ultimately made him a dark horse Cy Young candidate.

As great as Carmona's 2007 was, his 2008 has been. . . well, weird.   Looking at Mona's peripherals, you'd think he was getting rocked pretty much every start, or at least you might conclude that he's gotten super, super lucky.  Yeah, he's got that 2.25 ERA, but as we all know, these things are wont to fluctuate, and that stats like K/9, BB/9, and BABIP offer better ways to predict future performance.  None of these numbers have great things to say about Fausto: he's got a 3.54 K/9 ratio, walks literally twice as many guys as he strikes out, and his BABIP is .247, suggesting that some correction will occur.   Moreover, his defense-independent ERA is 4.97 (!).

However, he also has by far the best GB/FB ratio in the league (3.88, next best in the AL is Halladay at 2.74) he also has posted a LD% of 14%, good for 6th in the AL (interestingly, he had the exact same LD% last year, then good for first in the league).  He's also had great success at limiting extra-base hits, holding opposing batters to a .302 slugging percentage. 

Maybe I am misinterpreting the numbers, but I am beginning to wonder if it possible that Carmona actually defies the conventional "sabermetric model" of pitching success.  Essentially, Mona's peripherals, other than his monstrous GB rate, say his ERA screams fluke--nobody can sustain the kind of success he's had with those K and BB numbers. However, when watching the guy, I can't help but feel his stuff is so unique, the movement and velocity on his pitches are so great, that there's no way he's going to regress as much as the numbers would suggest (although expecting him to hold onto that 2.25 ERA is probably unreasonable).   I mean, calling him a sinkerballer is the understatement of the century--I've never seen any pitcher induce as many broken bats and high choppers in front of the plate.  Hell, it seems like half the hits he gives up are infield singles on balls that take too long to get beyond the pitcher's mound.

My question to the statheads/people who are interested in this kind of stuff: Is there a precedent for somebody like Carmona?  Are the factors driving his success--an outlandish GB rate and a very low LD% and opposition SLG--enough to sustain his ace-level performance?  Can a traditional statistical evaluation really account for this guy's ability to induce weak/off-balance contact?  I'm interested to hear your thoughts.