:format(webp)/cdn.vox-cdn.com/uploads/chorus_image/image/54701643/672812866.0.jpg)
So far, I have covered the basic and intermediate hitting stats/metrics. The links below:
Sabermetrics Hitting 101: Basic Hitting Stats
Sabermetrics Hitting 202: Intermediate Hitting Stats
Today, we’ll take the same approach for Hitting 101 for the basic pitching stats.
Earned Run Average (ERA)
Formula:
What does it actually mean: The amount of earned runs that a pitcher allows on average per nine innings pitched.
Is this useful? If so, how? Yes. ERA is essentially the go-to stat for measuring a pitcher’s performance. The goal of pitching is to prevent runs and the less runs you give up, the lower your ERA will be. There is an important distinction in using “earned runs” here - “unearned” runs (such as runs scored via an error) are not counted against the pitcher in ERA. Generally, however, the amount of unearned runs is fairly low so ERA will generally give a pretty good picture of how a pitcher performed (in terms of strictly run prevention) over a given timeframe.
How can I use it? ERA should be used strictly as a means of measuring past performance over a specific timeframe. ERA is not very predictive and we use other stats (strikeout/walk rates, FIP, etc.) to better predict if the pitcher was pitching to his “talent” or more on the “lucky” side. RA9 (which is earned + unearned runs per nine innings) and FIP generally give more information for pitching performance than ERA.
Deeper Dive: ERA can be heavily influenced by sample sizes, team defensive skill, and luck. We can look at them further:
Sample sizes: relief pitchers have notoriously volatile ERAs from year-to-year because one or two bad outings can really skew an ERA higher over a small number of innings (e.g. less than 70) despite stellar performance the rest of the year. Consider a reliever that has a 1.00 ERA over 72 IP that then gives up 5 runs in the next two games (and still gets 1 IP each outing). His ERA will jump from 1.00 to 2.19, but which is closer to the true talent level?
Team defense: Keep in mind that ERA is only measuring “earned” runs. However, a run is only “unearned” if there is an error assigned to the play. Having a bad defense will leads to a lot more balls in play that are not converted into out while still counting as “earned” runs - are these the pitcher’s or fielders’ fault? The quality of a defense can end up having a pretty significant impact on a pitcher’s ERA and this can be seen by ERA-FIP.
Luck: The ever-dreaded “L” word but it can have a pretty large impact on pitching. The main factors here are sequencing and errors. A crude example for sequencing can be seen in two pitchers with the following lines: 9 IP, 9 H, 0 BB, 0 K. Pitcher A gave up all 9 hits in one innings which led to probably 5+ runs scoring. Pitcher B gave up exactly one hit in all 9 innings and threw a complete game shutout. So while they have identical pitching lines, Pitcher A would have an ERA of 5+ for the game while Pitcher B would have an ERA of 0. This is an extreme example but it shows the impact that sequencing can have.
In general, RA9 ends up being a better overall performance indicator than ERA simply because the better pitchers will naturally prevent more runs from happening but controlling his K%, BB%, hits and homers. If a pitcher is getting a lot of unearned runs, he’s still allowing the offense to put the ball into play for these opportunities to happen, but ERA may not properly capture it.
Strikeouts Per Nine Innings (K/9) and Walks Per Nine Innings (BB/9)
Formula:
What does it actually mean: The amount of strikeouts and walks that a pitcher averages per nine innings pitched.
Is this useful? If so, how? Yes. Strikeouts and walks are both very important measures of a pitcher’s overall talent as they are both factors that the pitcher has control over. And they are extremely meaningful in conjunction with Fielding Independent theory, which we’ll discuss more with FIP down below. Specifically, strikeout and walk rates are the only stats that are mostly attributable to the pitcher rather than the hitter and/or team defense.
The “per nine innings” scale is used to match other pitching metrics, especially ERA and FIP.
How can I use it? Using these metrics is very straight forward: better pitchers will generally strike out more batters and walk less batters than average. The average K/9 in 2016 was 8.10 and the average BB/9 was 3.14. In general, the “elite” levels are about 10+ for K/9 and sub-2.5 for BB/9.
One thing to note: while BB/9 numbers have remained fairly steady over the years, strikeouts (and therefore K/9) have been steadily increasing. In 2000, the average K/9 was 6.53 and in 1990, the average was 5.72. So if you’re looking to analyze pitchers, especially from other years, be sure to look at the league environment and making comparisons based off that year’s baseline.
Deeper Dive: Since K/9 and BB/9 use Innings Pitched in the denominator, this means that they can be skewed on the high side for pitchers that give up a lot of hits. To better approximate a pitcher’s true talent, K% and BB% (below) are better indicators.
Strikeout Rate (K%) and Walk Rate (BB%)
Formula:
What does it actually mean: The rate at which a pitcher strikesout or walks batters.
Is this useful? If so, how? Yes. K% and BB% are very very similar to K/9 and BB/9 in regards to importance - however, since they are purely a rate stat on a per-better basis, they are slightly better indicators of true talent than K/9 and BB/9.
How can I use it? Similar to above: the average values in 2016 were 21.1% (K%) and 8.2% (BB%) and the “elite” values would be around 27% or higher (K%) and 4.5% or lower (BB%). As mentioned above, be sure to make comparisons to the league year for pitchers outside the current timeframe.
Walks and Hits Per Inning Pitched (WHIP)
Formula:
What does it actually mean: The amount of baserunners a pitcher allows per inning pitched.
Is this useful? If so, how? WHIP is a fairly good indicator of a pitcher’s ability to prevent baserunners. After all, the goal here is run prevention and few base runners means fewer opportunities to score runs. In many ways, WHIP is very similar to batting average - it gives an overall idea for the amount of baserunners that a pitcher gives up but it treats all baserunners equally. WHIP is good as a quick-glance approximation, but looking at things like OBP or wOBA are better rate stats that are used for more in-depth analysis.
How can I use it? WHIP is pretty straight forward - lower is better. An average WHIP is around 1.30 in 2016 and the elite pitchers generally have a WHIP lower than 1.00.
Deeper Dive: As mentioned with ERA above, WHIP is suspectible to the whims of team defense. A pitcher can control the walks portion of WHIP, but the hits/BABIP portion of WHIP are going to be part-pitcher and part-defense. A good example of an inflated WHIP is Robbie Ray in 2016 thanks to his .352 BABIP. An inflated WHIP will also lead to inflated K/9 and BB/9 numbers as a pitcher will be facing more batters in each inning pitched and therefore have more opportunities for strikeouts/walks in a 9 inning timeframe.
Unfortunately, I don’t have a lot more time to go any further so we’ll stop it here for today. There is a lot to cover with pitching (more than hitting) so the pitching articles will be a bit more spaced out.
I will not be available next Tuesday due to a work trip so don’t expect an article from me next week. As always, feel free to leave any comments/questions/criticisms in the comments below and I’ll see you all on May 23rd! Unless something big comes up, expect the next entry for pitching stats, which will focus on FIP, xFIP, and other pitching indicators.