Joe Sheehan, Parallel Universes, and HR/FB Ratio

A few days ago I was scrolling through my twitter feed, quietly enjoying my existence, when this one particular tweet caught my eye:

A.J. Burnett, failure who couldn't handle New York, had a 3.86 xFIP in 2011. Hiroki Kuroda, 2012's hero, has a 3.68 xFIP.

— Joe Sheehan (@joe_sheehan) September 16, 2012

As you might expect, Sports Illustrated's Joe Sheehan got quite a few strong reactions from Yankee fans over this statement, including one blogger's estimate that Sheehan's comment was the "worst saber tweet ever." While still others, intellectually inspired by all the recent debate on the value of ERA estimators, dared to ask the question, "Did you get dropped on your head as a child?"

Now, I don't want to speak for Joe Sheehan, certainly. He's a pretty big deal. But I am intrigued by the response Sheehan received because this is very easily something I myself could have tweeted. And if you know anything about Quantum Realities it's possible that somewhere, in some other universe, I actually did.

Imagine that for a minute. What if I was the one who had been casually scrolling along the leader boards at Fangraphs that day? I may have just as easily stumbled upon A.J. Burnett's 2011 xFIP and flippantly compared it to Hiroki Kuroda's 2012 xFIP in a smarmy 140-character quip and then bam! Just like that- James Gentile is now suddenly responsible for the worst saber tweet ever!

This is the worst nightmare possible! I'm now the laughing stock of Beyond the Boxscore. Everyone everywhere all at once unfollows me on twitter. Glenn DuPaul and Bryan Grosnick start throwing things at me-- tomatoes, spoiled lettuce, shards of glass, whatever. Managing editor Justin Bopp puts a "kick me" sign on my back and the whole world joins him as he starts laughing at me, chanting "Worst ever! Worst ever! Worst ever!"

How did this happen to me, mom? All I did was point out two similar stats from two Yankee pitchers, who've incidentally had very different narratives written about them.

The issue with Yankee fans, obviously, is that Burnett gave up many, many more runs that season, with a 5.15 ERA, than Kuroda and his 3.26 ERA so far in 2012.

There's absolutely no doubt that Kuroda in 2011 has pitched much better than Burnett in 2012. While the two had very similar strikeout rates in those seasons in question, Burnett walked almost twice as many batters and didn't go nearly as deep into games (5.9 IP/G) than Kuroda (6.7 IP/G). But none of that can really explain nearly two runs of separation between their ERA's until you begin to talk HR rate.

For A.J. Burnett in 2011, a highly unusual amount of his fly balls ended up in the seats. In fact, by my retrosheet estimate, his miserable HR/FB of 17.6% that year ranks as the 7th highest since retrosheet began recording batted ball data in 2003:

HIGHEST HR/FB SEASONS (min 150 IP)

His HR/FB ranks 6th highest by Fangraphs over that same time frame.

This is important to know because xFIP, for those who don't know, assumes only 11% (or so) of a pitcher's fly balls end up as home runs. It's designed as an ERA estimator, to estimate future performance, and not to assign value on current outcomes. xFIP is used because it was discovered that "pitchers don't seem to have a whole lot of impact on the number of home runs allowed, other than the extent to which they allow flyballs in general."

So, all Sheehan (or alternative-universe James Gentile) is really saying here is that, perhaps, Burnett wasn't as bad as you thought. Perhaps. Was Sheehan (alt-me) being unnecessarily provocative? Maybe. But you guys are Yankee fans. You've won something like 3,496 World Series trophies.

And as it turns out, Burnett's HR/FB has returned to a substantially less crazy 12.8% in 2012 and as a consequence he's managed to do well for himself with a 3.64 ERA. How much of that is owed to leaving the AL East? Not sure, but it's extremely rare that a pitcher posts this kind of miserable HR/FB rate in consecutive seasons in any division. Take our sample of miserables from earlier:

HIGHEST HR/FB SEASONS (min 150 IP year 1)

Nearly all of these pitchers saw their HR/FB's return to normal or normal-ish rates. Among the worst of the worst HR/FB rates of last decade, only Odalis Perez, Chris Volstad, Gregg Maddux, and Felix Hernandez (in 2007) saw HR/FB ratios above 13% for a second consecutive year.

Are there exceptions? Always. Braden Looper may very well have been knocked out of the majors for consistently high HR/FB rates that peaked in 2009, and it's a good bet there was more to it than simply bad luck. And we never got a chance to see how Cory Lidle would've rebounded, due to his tragic death in a plane crash during the 2006 postseason. But for proven major-league talent, regression is almost always on the way. Brandon Webb won a Cy Young award in the year after his +17% HR/FB, and I seem to remember Felix Hernandez having some pretty good seasons shortly thereafter as well.

Of course, A.J. Burnett made this list once before, with an even higher 18.7% HR/FB in his 2007 season with the Blue Jays, so you may be tempted to insist that A.J.'s struggle with the long ball in two separate seasons is much more than coincidence. I won't stop you from making that argument, but I'd like you to, at least, consult my friend Roy Sullivan about coincidences first.

Note: Retrosheet batted ball data will differ from BIS data at Fangraphs and Baseball-Reference.

If you thought the tweet featured in this article was bad, imagine that every day of your life @JDGentile.