Cole Hamels' 2009: More than just bad luck?

In 2008, Cole Hamels was hailed as one of the best young pitchers in baseball, carrying his team through the postseason along the way.  All eyes were focused on his 2009 season, as everyone was waiting in anticipation of what he would do for an encore.  Needless to say, he ended up disappointing most, posting a 4.32 ERA and losing the faith of Phillies fans around the globe...

Or Did he?

Any sabermetric follower will tell you that Hamels was effectively the same pitcher in 2009 as he was in the year previous, posting near carbon copies of his K/9, BB/9, and HR/9, good enough for a 3.72 FIP, the exact same FIP he made in 2008.  They were quick to point out that Hamels was a receiver of bad luck on balls in play, as a .325 BABiP was well below his previous marks of .270 and .289; the latter numbers being somewhat sustainable for a flyball pitcher who generates a good amount of popups.  He also stranded less runners, posting a 72% strand rate as opposed to 76% and 78.7% in the previous years.  Not only that, but he actually saw improvement in his plate discipline metrics, as hitters made less contact with his pitches overall and in the zone.  In all aspects, Hamel's was a prime bounce-back candidate in 2010, as sabermetricians found nothing different in how he performed between the two years.

But what if there was?

From watching some of his games, if there was one thing that appeared to be common with many of his starts in 2009, it was that Hamels would have little trouble the first time through the order, and he would subsequently fall apart in later innings.  This would suggest that Hamels' was becoming predictable to hitters, as opposed to being unlucky or suffering from nagging injuries, but do the stats back this up?

According to Baseball reference, something is definitely up.

In 2008, his Split-OPS+, BABiP, and K/BB ratios each time through the batting order are excellent and consistent throughout, in fact he would end up becoming better as the game went on.

First time through the order: 76, .259, 2.95

Second time: 71, .289, 3.89

Third+ time: 67, .240, 4.57

But in 2009...

First time: 80, .270, 4.67

Second time: 100, .329, 4.00

Third time: 124, .413(!), 3.42

The fact that his BABiP increases in an ordered fashion is very intriguing, as a pitcher suffering from bad luck would likely expect their BABiP splits to be random or uniform.  There isn't much else to extract from this set of data, Cole Hamels definitely got worse as the game went on, even though his peripherals remained excellent.  If Hamels' LD% could be split in a similar fashion and it followed a similar trend, then this theory would suddenly become much more probable, as line drives went for a .700+ BABiP last year, and they can easily cause a discrepancy between a pitcher's ERA and FIP.

2010 is a new year, and there is no doubt that Hamels is working hard to return to baseball and prove the doubters wrong, and return to his position as one of the best young pitchers in baseball.  Is he still an excellent bounce-back candidate?  Yes, that never changed, but perhaps it will take more than just the return of lady luck to do so, because 2009 might not have been as big of a fluke as it is commonly made out to be.