A few weeks ago, the Mets were playing the Padres out in San Diego. Keith Hernandez usually doesn’t take long road trips with the Mets. Why is this important? It gave the other two members of one of the best booths in baseball – Gary Cohen and Ron Darling – a chance to talk about FIP, or Fielding Independent Percentage. If Keith were there, he would’ve plucked every hair out of his glorious mustache one by one until the discussion was over.

So, person reading this, I’m sure you’re wondering what FIP is. At this point, all the info I’ve given you is that it’s an acronym and that Keith Hernandez hates the modernization of baseball. Fielding Independent Pitching, as defined by Baseball Prospectus, “converts a pitcher’s three true outcomes into an earned run average-like number.” Those “three true outcomes” are strikeouts, walks, and home runs.

“Jack, if we wanted to see a basic explanation, we could’ve just googled FIP.” You’re right. Let’s go deeper.

For all intents and purposes, FIP is a more accurate measure of pitcher performance than ERA. FIP controls for differences in defense, hence the “fielding independent”, by using a formula to place values on those three true outcomes. The formula is:

(13 * # of HR + 3 * # of BB – 2 * #K) / IP plus a constant (generally 3.2) that puts it on the same scale as ERA

In other words, FIP can be interpreted similarly to ERA (3.00 is good, 5.00 is horrible), but only includes what the pitcher can actively control. Each of the outcomes is weighted, with homers and walks increasing the number, while strikeouts decrease it. Do I know the reasons for the specific weights? Of course not, but I can hope to explain them.

The heavy weight for homers makes sense, as that is the only play that directly result in a run (or runs) for a pitcher. A walk puts a runner on base, but is obviously less harmful than a homer. And a strikeout is weighted less because who knows, but strikeouts are important as batters putting balls in play immediately puts the outcome out of the pitcher’s hands.

At its core, all ERA tells you is the amount of runs scored while a pitcher pitches. FIP doesn’t tell you runs scored, which is why people have issues with it. It’s more of an abstract number that doesn’t directly tell you any information when you look at it, but looking at it on the ERA scale makes the meaning discernible.

FIP could (and in my opinion, should) replace ERA as the baseline stat for pitchers (though there are more complicated and weighted versions of it, but I’m too math averse to delve into that). But in this context, we can use it to validate – or invalidate – the current ERA of a pitcher.

First, let’s look at a “control”, or in this case, someone whose FIP lines up with their ERA. The best pitcher on the planet right now, Clayton Kershaw, has otherworldly numbers across the board, so he makes sense to look at.

Year | G | IP | ER | HR | BB | SO | ||
---|---|---|---|---|---|---|---|---|

2016 | 1.56 |
11 | 86.2 |
15 | 4 | 5 | 105 |
1.49 |

9 Yrs | 2.39 |
255 | 1697.2 | 450 | 101 | 471 | 1851 | 2.56 |

As you can see, Kershaw’s ERA and FIP are incredibly close, meaning that the numbers he is producing and his success are due to his own skill, not that of his defense. He doesn’t put runners on base, and doesn’t give up homers, while being absolutely incredible striking them out. Furthermore, it means that you can expect him to continue his dominance, as those numbers are in line. He’s not being buoyed by outside influences, just his own skill.

With our control in mind, let’s look at someone outperforming his FIP, Jimmy Nelson out in Milwaukee.

Year | G | IP | ER | HR | BB | SO | ||
---|---|---|---|---|---|---|---|---|

2016 | 2.88 | 11 | 72.0 | 23 | 9 | 28 | 59 | 4.51 |

4 Yrs | 3.92 | 59 | 328.2 | 143 | 33 | 117 | 272 | 4.09 |

Without watching Nelson and just looking at his ERA, you would think he’s having a very good season. Even by just looking at his 28 walks, you get a sense that there’s something up. And then you see a 4.51 FIP and – if you’ve been following along – you should know that his “success” is a sham. He’s most likely been getting lucky in tight spots, which to some degree is a skill, but not when you almost have half as many walks as strikeouts. Nelson should be hit by some hard regression in his ERA at some point, as his luck starts to fade towards the mean and the free passes he gives out come around to score.

Now let’s look at the other side – someone with a high ERA but low FIP: David Price.

Year | G | IP | ER | HR | BB | SO | ||
---|---|---|---|---|---|---|---|---|

2016 | 5.11 | 11 | 68.2 | 39 | 7 | 18 | 79 |
3.09 |

9 Yrs | 3.18 | 229 | 1510.1 | 534 | 136 | 389 | 1451 | 3.19 |

Price’s ERA has ballooned so far this year in his first season in Boston. But according to his FIP, this is due mostly to balls being scattered around the yard, not out of the park. While this could be misleading, as FIP isn’t totally comprehensive, it can be assumed that the 5.11 isn’t all Price’s fault, and his strikeout numbers belie hopeful improvement on his runs allowed. His home run total is high, which can sometimes regress as well, but the fact that he still has a low FIP despite the homers tells you that he’s been sterling in walks and strikeouts.

It’s also worth noting that over his career, Price’s ERA has been due to his success and not that of his defenses, with his career ERA and FIP just .01 apart.

So, now that I’ve thrown tables and some words at you, what is the value of FIP? FIP allows us to evaluate a pitcher solely on their merits, not those of their defense or, in more advanced versions, their ballpark. Baseball is a sport of player valuation and evaluation, and a stat like ERA can sometimes be misleading. By looking at more advanced (though just scratching the surface) stat like FIP, we get a better idea of whether their success is for real or something happening independent of their pitching. FIP is already frequently used by baseball writers like Jonah Keri, but it should hopefully soon break into the mainstream through broadcast teams. And when it does, you’ll appreciate their explanation that was much clearer than mine.

Did you like this explanation of FIP? Probably not. If there’s something you want clarified, have more questions, other stats you want me to (try to) explain, or have a complaint, let me know either in the comments or on Twitter @jfmclooney.

Pingback: Quantifying Luck: Explaining BABIP | JackMcLoone.com