Friday, October 8, 2010

Breaking down FIP

So, if you've been living in a cave, you might not have noticed the recent explosion of baseball metric stats. These stats claim to measure everything from how many runs a player is worth to the size of Bruce Bochy's head has attained on any given day. In my next few posts I'll take some time to break down a few of the metrics that have caught on and are becoming more common. Today's topic is FIP or fielding independent pitching.

FIP was designed by Tom Tango to be a metric that measures things that pitchers can control, and to give us an idea of the contribution the pitcher himself is actually making to his team.
The FIP formula is: (HR*13+(BB+HBP-IBB)*3-K*2) plus a league specific factor to account for league average ERA for that season. Tom Tango and Dave Cameron have both asserted that the stat isnt perfect, and doesnt take into account batted ball data (line drives vs ground balls) or situational pitching (runners on/ close and late etc), even so, I think that a few simple adjustments need to be made.

In my opinion, a BB isnt 1.5 times as bad as a K is good. It takes 4BB to create a run but only 3K's to end the inning. If a pitcher had a 1/1 K/BB ratio and struck out and walked 27 per 9IP he'd throw a shutout every time (that assumes an unrealistic no contact rate). If we assume that the pitcher walks one guy an inning has a BABIP against of .300, and a K/9 rate of 6 he would probably have an ERA right around 3. (One walk an inning, needs 2 singles or 1 double to score, pitcher K's 2 out of every 9 batters and gets 70% of hitters that make contact out)

Also not included in the stat are line drive or groundball metrics. I think a pitcher is very much in control of how hard the ball is hit (in most cases) and a line drive vs grounders and IF popups would also be beneficial. Yes, groundballs for hits would actually benefit the pitcher which may be a problem, but I think these two minor changes would improve the reliability of the metric and not give us numbers like this: Javier Vasquez went 11-12 with a 4.84 ERA in 2006, but had a 4.04 FIP metric. Meaning he was good from a FIP standpoint (or at least above average) but terrible from an ERA standpoint. Was his team that bad? Did or did he just give up a lot of hard hit balls and inopportune times? Was he bad or unlucky or both? How do we figure all of this out? We start by adjusting the metric to include a couple more things, and make BB less harmful for pitchers.. (Jonathan Sanchez and Clayton Kershaw seem to be able to survive walking people).





No comments:

Post a Comment