clock menu more-arrow no yes

Filed under:

What 3 new pitching metrics say about the Rockies’ rotation

New, 10 comments

There’s an especially promising picture for Jon Gray

The first thing to know about Baseball Prospectus’s new trio of pitcher metrics is that they are centered on process. The Power, Command, and Stamina scores—laid out by Kate Morrison, Harry Pavlidis, and Jeff Long—aren’t drawn from results like walks and strikeouts, but from methods like pitch selection and location. The second, and related, thing to know is that the scores aren’t shortcuts to answering the question “how good is this pitcher?” Instead, they help answer the question “by which mechanisms does the pitcher reach his results?” The new metrics contribute to typological understandings more than pitcher quality.

With those essentials in mind, let’s look at where the 2017 Rockies’ pitchers who threw more than 110 innings from the starting rotation rated, with particular focus on the returning players.

Power Score

Three components comprise the Power Score: peak fastball velocity, frequency of fastball usage, and the velocity of off-speed pitches. Fastball velocity is given the greatest weight in the calculations, and velocity of off-speed pitches the least. Like the Command and Stamina scores, Power is on a scale from 0 to 100. Among pitchers with at least 110 innings pitched, Pirates’ starter Chad Kuhl had the highest power score at 68. RA Dickey had the lowest possible Power score, 0.

Because off-speed pitch velocity is factored in to the Power Score, pitchers who favor sliders and sinkers over curveballs and changeups will have higher Power Scores. Unsurprisingly, righties also tend to have higher scores. As you might expect, Antonio Senzatela, Jon Gray Tyler Chatwood, and Germán Márquez all had fairly high Power Scores. Senzatela led the pack because he basically has two pitches, a fastball and a slider. His usage probably bumped him above Gray, who has higher fastball velocity. The lone lefty of the bunch, Kyle Freeland, has the lowest Power Score.

Chart by Eric Garcia McKinley, via Tableau


It’s always useful to remember the distinction between command and control. Control is the ability to throw strikes, while command is the ability to locate pitches where the pitcher wants. There’s rarely a reason to want to locate a pitch in the middle of the zone, so Command scores are drawn from the pitcher’s ability to consistently locate pitches at the edges of the strike zone. The Command Score is built on Baseball Prospectus’s Called Strike Above Average and Called Strike Probability metrics. They are used to identify ideal targets in each quadrant of the strike zone, and pitchers are penalized for missing those targeted areas. The penalties are harsher for missing outside of the zone than in it.

The 2017 leader in Command Score was Kyle Hendricks, with 76. That’s not surprising, and it helps this metric pass the smell test. Zack Godley had the lowest Command Score at 23.

The Command Score is less intuitive in terms of its foundations than the Power Score. Similarly, one natural results-based statistic for measuring command, walks per nine innings, doesn’t map on to Command Score. The smallest BB/9 differential among the five Rockies’ pitchers who threw at least 110 innings is between Gray’s 2.4 and Márquez’s 2.7; however, the largest Command Score differential is also between Gray and Márquez, 60 for the former and 33 for the latter. This suggests that Gray missed fewer targets near the edges of the zone than Márquez, and when he did miss Gray’s pitches were more likely to end up in the strike zone.

Chart by Eric Garcia McKinley, via Tableau


The Stamina Score attempts to capture a pitcher’s workload capability. Rather than innings pitched, the Stamina Score is built from pitches thrown per outing, batters faced, and days of rest. Specifically:

What we found most effective is a model that combined calculating the daily number of pitches thrown from a six-day moving average, with the straight average of batters faced per game against the square root of the mean of the days of rest between games.

It’s possible for workhorses and mop-up guys to have high Stamina Scores due to the de-emphasis on innings pitched, although truly bad pitchers probably won’t be around long enough to gain a high Stamina Score anyhow. In 2017, Justin Verlander and Rick Porcello led baseball with Stamina Scores of 86. The lowest stamina scores among pitchers with at least 110 innings pitched was also a tie: Madison Bumgarner and Jon Gray both had 53.

Despite the emphasis on pitches thrown and batters faced, it may be no accident that the two starters with the lowest Stamina Scores both pitched partial seasons. And in fact, of the 35 pitchers with a Stamina Score below 65, only one pitched more than 150 innings—Alex Wood. But at the same time, Dinelson Lamet had a score of 68 in just 11413 innings pitched.

Other members of the 2017 Rockies’ rotation had higher Stamina Scores. Márquez had the highest, while Freeland, Chatwood, and Senzatela clustered together in the mid 60s.

Chart by Eric Garcia McKinley, via Tableau

★ ★ ★

There are two big takeaways for 2018 here. The first has to do with the look of the rotation as a whole.

The 2017 Rockies were full of power pitchers. Getting full seasons from Tyler Anderson and Chad Bettis would disrupt the dominant, power pitcher, type. There may be a benefit there. Bettis throws a lot of curveballs and changeups, and his fastball tops out in the low 90s. He’s decidedly not a power pitcher. His Power Score in 2016 was 49, and it went along with Command and Stamina scores of 62 and 78. Bettis’s Power Score is not extremely low, but it is low relative to the rest of the rotation. Similarly, Anderson’s presence as a low-power lefty with good command could add yet another different look to the Rockies’ rotation. Anderson’s Power Score in 2016 was 37, while his Command and Stamina scores were 63 and 66. (This is a good time to remind everyone that a low Power Score does not mean bad pitcher; Clayton Kershaw’s Power Score in 2017 was 45.)

Second, there are mixed signals about Jon Gray here, but it’s an ultimately encouraging view. In 2017 Gray was a power pitcher with good command. There were only four starters with at least 110 innings pitched to have Power and Command scores in the 60s: Gray, Justin Verlander, Gerrit Cole, and Jacob deGrom. The type of pitcher Gray is intuitively fits with those pitchers. It’s good company for him to be in. The difference between Gray and those three pitchers, however, is that they also had high Stamina scores. Verlander was tied for the highest in baseball at 86, and Cole and deGrom weren’t far behind, as they each had 82. As noted, Gray’s Stamina Score was tied for the lowest in baseball.

While the authors make it very clear that the Stamina Score is not meant to be a shorthand to “good” and “bad” pitchers, stamina is a desirable trait for a starting pitcher. It’s easy to look at Gray’s Power and Command scores and wish he completed the trifecta with Stamina. It’s fair to say that this is where he can improve to elevate his game, and it’s also fair to say he may do that simply by not being injured and pitching 175 innings.

In fact, looking back to Gray’s 2016 makes his trajectory even more promising. In 168 innings pitched, Gray’s Power, Command, and Stamina scores were 56, 35, and 74. That tells us that the 2017 version upped his power game and vastly improved his command. While Gray doesn’t need to raise his Stamina Score (he had a better season than Cole in 2017) to elevate his game and the Rockies’ rotation along with it, it couldn’t hurt. The ingredients are all there. They just haven’t been combined in the right way yet.