Fielding independent pitching (FIP) is a recently developed statistic that is based on the DIPS (defense-independent pitching statistics) theory, which is built upon the assumption that pitchers have little, if any, control over the outcomes of balls put into play against them. The theory was first developed in the 1990's, when an increased number of the three true outcomes (home runs, strikeouts, and walks, plus hit batsmen) were permeating the game, a trend that has continued to this day. Since the three true outcomes are impossible to defend against, they were used, in the form of FIP, as a new method of measuring a pitcher's ability. The formula for FIP is (13*HR + 3*BB - 2*K)/IP + C, where C is a constant term that re-centers the league-average FIP to match the league-average ERA.
If you've been reading this blog long enough, you are probably aware that I am very skeptical about advanced stats, especially those based on assumptions (DIPS theory) or using seemingly arbitrary methods of justification (such as the constants). FIP is typically used as a predictor of future success rather than an indication of past success, and rightly so. Even though I have serious doubts about the credibility of this statistic, I sometimes glance at the number for a pitcher who I'm considering adding to my fantasy team, just in case there's a big difference between his FIP and his ERA (in which case I check the components of FIP, home run rate, strikeout rate, and walk rate, just to put everything on game level). Other than that, though, I don't use it often.
Since FIP is a modern stat, designed for modern pitchers, I don't think it should be used in Hall of Fame debates. Take Lon Warneke as an example. He's a pitcher I'd be fine seeing in the Hall of Fame. Warneke had solid game level statistics: a won-lost record of 192-121 (good for a .613 winning percentage, way above the Hall of Fame average) and a 3.18 ERA. Twice he led the NL in homers allowed, but it didn't really matter, as he won 18 and 17 games in those two seasons, respectively. Warneke's career FIP was 3.77, significantly higher than his 3.18 ERA. Should this be held against him? FIP was invented roughly sixty years after he retired, and back in the day Warneke was considered a top pitcher, making five all-star teams and finishing second in the NL MVP voting in 1932.
I don't think FIP should ever be considered in the Hall of Fame discussion because Hall of Famers should be the ones who help their teams win the most games, and runs are what ultimately win and lose games. That's what ERA is for. It doesn't really matter how the runs came across; what matters is whether they did or not. FIP punishes Warneke for giving up homers and not striking people out, but he also didn't walk many and was a master of getting soft contact, the very definition of a finesse pitcher. Warneke's run prevention skills helped him win games, and in the end that's all his team can ask of him.
I love FIP for its predictive value, and have no reservations over its general validity, but I don't think it should be used to make judgements on value - fundamentally, a pitcher's value is in not allowing runs. That's why I disagree with Fangraphs using FIP instead of ERA for its version of WAR.
ReplyDeleteI don't think there's anything wrong about using new stats for old players. It's just that FIP is really for prediction and we don't exactly need that for players long dead.
There are some exceptions to DIPS, but it's true for almost everyone.