Prior to the 2008 season, there was a brief sabermetric and internet fascination with Brian Bannister, who in 2007 had enjoyed a relative measure of success despite low strikeout totals and a general scouting consensus that his stuff was average to below average. So was Bannister lucky or was he using his intellect and guile to find success? To make the story more compelling, Bannister admitted that he didn't fit the evidence shown by numerous sabermetric studies, and showed an awareness of a number of advanced stats -- I guess he lives in his mom's basement too -- and what they meant for him going forward.
Well, 2008 was a victory for those who said he was lucky in 2007. Despite a hot start, Bannister quickly fizzled, and ended up with a 5.76 ERA, up from his 3.87 mark in 2007. Compounding the matter, the league average dropped nearly half a run in '08, which meant his ERA + skyrocketed, from 121 in '07 (that is, 21% better than league average) to a 74 mark in '08. And yes, the batting average on balls in play number increased, from .266 in '07, to .316 in '08.
The odd thing is, this increase was paired with some improvement in Bannister's major weakness, his strikeout rate.
So was he actually a little bit unlucky in 2008?
In a kind of Greek way, this may make sense: lucky in 2007, unlucky in 2008, same guy somewhere in the middle. The one problem with just writing off 2008 as bad luck is that Bannister was also hit much harder in '08. So the issue isn't just that he allowed a few more singles past a diving Gload. (The Royals defense may have taken half a step back in '08, but more or less seems mostly comparable. The team's defensive eff. rank, a decent ballpark figure, were fairly similar for '07 & '08.)
What never quite got talked about enough in the emerging Bannister mythology of 2008 is that he was actually also pretty damn lucky in 2006, which then made 2007 look like something that he just knew how to do. A .254 BABIP? Johan Santana has never allowed a number that low. (stats) I guess he's dumb.
So obviously, returning to the lovely chart above, there's the strange matter of Banny's increasing LD% (imperfect, but not useless as a measure of how hard he's being hit) and the big jump in the HR total in 2008, against the fact that he set a new career high for strikeouts.
Especially odd is the fact that as Bannister's season went down the drain, he started striking out more people. I don't feel like calculating his K/9 by hand, but this increase is easy enough to see just looking at his monthly splits. For example, in 35.2 IP in April, Bannister posted a 4.04 ERA with just 21 strikeouts, basically a typical Bannister effort. In both July and August, Banny was blown up to the tune of 7.20 & 7.76 ERAs, but had relatively high strikeout totals those months (24 & 20) in more limited time on the mound. Now, there is a chance that a certain number of these increased Ks in less innings are to be expected, as Bannister was facing more men per inning, as he was getting shelled. When you zoom out however, and look at the overall season numbers, I think there's enough there to safely say he was a) striking out more guys and b) getting hit harder.
So where does Bannister go from here? The projection systems are fairly disparate, suggesting Banny remains a unique player. James sees a 4.26 ERA, Marcel 4.66 and Chone 5.01.
In 2008, when things started to go bad, did he change his approach and begin to go for a higher risk-reward scenario (more strikeouts, more homers) by putting more pitches in the heart of the zone? Did he make a modest improvement stuff-wise, only to see it mitigated by bad BABIP luck and an increased homer rate? Or did really nothing change, other than what we finally saw on the field?