The formula takes out the luck of hits per balls in play. BAA doesn't do anything like that at all. The formula for hit rate is: (H-HR)/(BFP-HR-BB-K-HBP). BFP is batters facing pitcher. You really don't need the second part of the equation to do a projection, only the numerator. Multiply the numerator by the league average hit rate (around .300), add that to the pitchers HRA, and divide by (BFP-BB-HBP). That would you give the pitcher's true BAA, luck independant. I also have a formula for finding team-specific ERA & WHIP based on the team's park and defense, which basically takes the team's projected hit rate (based on historic hit rates) and has linear weights for BB/9, K/9, HR/9, and Team Hit Rate. Some pitchers do tend to have ERA's that stray from the expected ERA given by the formula, and those pitchers are usually extreme flyball pitchers or groundball pitchers (like Johan Santana, Brandon Webb, etc).
In Bonderman's case, his Hit Rate was .313 in 2003, and .278 in 2004. It's not a skill-related stat (RJ and Pedro have been near the worst in the league for hit rate) so it's not progression by Bonderman, it's just luck. Next year he could be way above .300, or way below .300, you don't know.
The formula said Glendon Rusch's ERA should have been 4.02 base don his K, BB, and HR in 2003, but in real life it was 6.42, quite a difference. In 2004 Rusch had a 3.75 ERA which doesn't seem all that shocking when you figure he was only 27 ERA points below his previous season's total. Bonderman's 2003 ERA should have been 4.75, not the 5.56 he actually had. Going back to previous years, Ryan Drese's aweful 6.55 ERA in 2002 should have been about 4.75.
Going the other way finding pitchers who had good luck, Kip Wells' 2003 looked great just by looking at ERA and BAA. 3.28 ERA, .233 BAA, how could you go wrong? But what SHOULD his ERA have been that year? About 4.36, more than enough information to tell you this guy's gonna be over-rated big time in 2004. So 2004 rolls around and he has a 4.55 ERA. People are shocked, thinking he was the next big thing. Sorry.
Another 2003 example is Ryan Franklin. His ERA was 3.57 in 2003. Again, people think he's a decent option as a 3rd starter maybe. Had a good ERA and BAA, so why not? Well his ERA should have been 4.60 according to my formula. 2004 rolls around and Franklin has a 4.90 ERA.
So yes, the formula does work. If you're looking for people who were lucky in 2004, Jake Peavy and Johan Santana top the list. Jake Westbrook and Joe Kennedy are also up there near the top. Unlucky oens were Sidney Ponson, Bronson Arroyo, Jeff Weaver, Derek Lowe, and Kris Benson.
"Jack, will you call me, if you're able?"
"I've got your phone number written, in the back of my Bible."