People will always be skeptical of the idea that pitchers don't have a lot of control over the outcome of balls put into play. It's beyond counterintuitive. So when I explain the concept to folks, I ask them to tell me who they think was the hardest pitcher to get a hit off in the last two decades. Sometimes they say Roger Clemens, and sometimes it's Pedro Martinez. I like to suggest Randy Johnson.
Then I ask them to think of the pitcher it would be easiest to make solid contact against. Because I'm in the Bay Area, sometimes they say Kirk Rueter. Other times they say Kirk Rueter. Then I give them the career batting averages of those pitchers on balls put in play.
Randy Johnson: .295
Pedro Martinez: .282
Roger Clemens: .286
Kirk Rueter: .289
I don't get it, either. But the arguments for BABiP and FIP are pretty convincing, and you can pick whatever pitchers you want from the same era. It's almost always the same thing. There are outliers -- Matt Cain likes to hang out with guys from the '70s and '80s on this list, whereas Ricky Nolasco always seems to get hit harder than the average pitcher -- but for the most part, when hitters get a bat on the ball, the hits fall in at the same rate regardless of who is pitching. It's been just under 15 years since this became widely known, and no one really disputes it anymore.
Tim Lincecum is having an awful season, with an ERA that's about twice as high as you'd expect. His batting averages allowed on balls in play since coming to the majors:
2007 - .283
2008 - .304
2009 - .282
2010 - .310
2011 - .281
2012 - .353
Case closed. He's been unlucky. Glad we can solve that, folks. Thanks for coming by. His FIP is 14th-best in the National League, just ahead of Clayton Kershaw. He's fine. Stats say so. Move along.
Here's the thing, though: I've watched every pitch Lincecum has thrown this year. And it sure as hell doesn't feel like he's been unlucky. When he gets runners on base, it seems like he's a different pitcher. You're expecting him to melt down. He looks great for the first two or three innings, and then everything falls apart. You start to pretend that you're a body-language expert, and you read too much into everything he does. And the hits keep coming and coming and coming.
So when I use FIP or BABIP in a Lincecum-related article over on McCovey Chronicles, I feel dirty. I can't shake the feeling that I'm hiding behind a magic talisman, or blaming everything on the rain gods. You can't get mad at Tim Lincecum! He was just minding his own business in the exercise yard when luck came up and shanked him with a sharpened toothbrush handle! Not his fault! La la la la la la la la la la. At the same time, it's almost impossible to believe that he's unlucky, so I feel dishonest.
FIP and BABiP are too counterintuitive to believe right away. You have to be hit in the face with them over and over again until you give in and figure there's something to them. That's why they'll never be on a scoreboard or a TV broadcast. Too far removed from what you think reality is. When a pitcher is allowing a lot of hits, he is being bad at his job. Not allowing hits is his job. It seems like there's a problem with the stats that disagree with that.
That's what makes them important stats, though. That's exactly why they're useful. It's too hard for the brain to process and calculate hundreds and hundreds of snippets of anecdotal evidence. FIP and BABiP question what you've observed like no other stat can. A guy with a high OBP? He probably looks like someone who spends a lot of time on the bases. A hitter with a gaudy oWAR or RC/27? You probably figured he was valuable before you saw those stats. But a guy with a FIP that's half of his ERA seems completely wrong. It goes against everything you've observed. I'm not saying that FIP is the perfect stat. But it might be the most important. Without fielding-independent stats, we'd be quick to make assumptions that probably aren't true.
Also, fielding-independent stats are telling me exactly what I want to hear at exactly this moment about Tim Lincecum. That also makes them great stats. That's the main point I was trying to get across here. But the other stuff too!