First of all, I’ve done a couple updates to the Library over the course of the past week. We now have a section there on Contract Details, which can be found under the “Sabermetric Principles” heading. It’s filled with articles discussing the details behind such confusing things like waivers, player options, service time, and Super Two status. The beginning of the year is always filled with lots of questions about how many options players have and when rookies can be promoted yet still delay their arbitration clock, so hopefully these articles are helpful for everyone.
Also, I’ve finished adding pages on Shutdowns and Meltdowns and the minus stats (ERA-, FIP-, xFIP-). If there are any other pages you’d like added to the Library, feel free to reach out to me on Twitter or through the contact us form on the right-hand side of all Library pages.
And now, since the minus statistics are all still rather new around these parts, below the jump I’m going to include some brief thoughts on ERA-, FIP-, and xFIP-, using number from the Aughts (2000-2010) as an example. Since these statistics are park and league adjusted, they are perfect for comparing performances across different years and leagues.
As a quick summary, I like to think of FIP- and xFIP- as showing how good a pitcher is (compared to league average) in areas directly under a pitcher’s control: strikeouts, walks, homeruns, and groundballs. ERA- is different: it shows how good a pitcher is (compared to league average, again) at limiting run scoring. That’s an important distinction and can lead to very different scores in ERA-/FIP- over the course of a single season, but the difference between these statistics over large samples is normally small. In the long run, results almost always end up coming close to matching the underlying process.
Consider the above two charts. Pedro Martinez posted a 46 FIP- and 35 ERA- in 2000; Randy Johnson posted a 48 FIP- (only 2% worse than Pedro’s) in 2001, but his ERA- was 57…a whole 22 percentage points higher than Pedro’s ERA-. When we look at their overall results over the course of the Aughts, though, both statistics come closer in line: Pedro posted a 67 ERA-/66 FIP- and Johnson posted a 75 ERA-/69 FIP-.
Pedro Martinez is obviously amazing, and these numbers don’t even include his otherwordly performances from the 1990s (30 FIP- in 1999!). As it was, Pedro was still 54% better than league average in 2000, and 34% better than league average throughout the Aughts. Randy Johnson was equally impressive, especially when you consider that he was 36 years old in 2000; that’s right, Randy Johnson’s performance from ages 36-45 was only slightly worse than Tim Lincecum’s performance has been these past four year (ages 23-26). Talk about mind boggling. Oh, and yes, Zack Greinke’s 2009 season was every bit as spectacular as it seemed at the time.
Since I set the innings pitched parameter at only 300 IP through this time period, there are a couple names on the overall leaderboard that were dominant for a short period before suffering from injuries. Can you imagine what Mark Prior would have been like if he hadn’t gotten injured? The same can be said for Brandon Webb, but to a lesser extent as Webb did still pitch five seasons with over 200 IP.
For kicks and giggles, here’s a look at the opposite end of the spectrum: the pitchers that were the worst over this time period at allowing runs.
As good as Pedro Martinez was in 2000, Jose Lima was even further from the league average in 2005…just in the opposite direction. Similarly, Pedro led the majors by being 33% better than the league average in ERA- over this entire time frame, but Scott Erickson managed the impressive feat of being 47% worse than league average while still racking up over 300 innings pitched. Phew, talk about rough.
So again, the Minus statistics are great for comparing pitchers between teams, leagues, and years, and they are an easy way to discuss sabermetric ideas without coming across as confusing or nerdy. For more information on them, check out their page at the Library.