The offense can be frustrating but the biggest question mark going forward is the pitching staff
On Monday I wrote about the conundrum the Astros face due to the combination of exceedingly thin starting pitching depth and potential work load limitations on certain pitchers. By Monday evening, we learned that Justin Verlander was placed on the 15 day injured list with a sore neck, which again gives us a stress test of the Astros’ pitching depth.
In the immediate future, the Astros must decide whether to move up Spencer Arrighetti as Thursday’s starter in Chicago or call up a AAA starting pitcher (such as Blair Henley, currently scheduled to pitch in AAA on that day). Since Arrighetti would be pitching on regular rest Thursday, that may seem like an easy solution. But it postpones the possible need to call up a AAA pitcher until the Orioles series which starts on Friday. So, would the Astros prefer a less experienced pitcher from AAA pitching against the weaker White Sox or the very strong Orioles? Conundrum indeed.
Understanding the future performance of the Astros’ rotation is a key to predicting whether the Astros have a feasible chance to come back in the AL West or AL wild card races by the trade deadline. We all know that the offense can be frustrating. But peaks and valleys are not unexpected for any offense. I would argue that the stability and performance of the Astros rotation in the future will be the most important factor for the Astros’ ability to contend this season.
I like looking for information in the numbers. A statistic which intrigues me is the difference between pitcher ERA and x-ERA. The expected ERA (x-ERA) is based on Statcast’s xwOBA allowed by the pitcher. Quite a few Astros’ pitcher have dramatic differences between their actual ERA and x-ERA. Some examples, with the number of runs that actual ERA is above expected ERA: Hader (+1 run), Dubin (-+2.3 runs), Brown (+1.4 runs), Pressly (+1.3 runs), and Arrighetti (+1.8 runs). (It’s also worth noting that some Astros pitchers, like Ronel Blanco and Taylor Scott, have a difference in the opposite direction, with ERA less than x-ERA.)
Comparison of Mariners and Astros x-ERA
The Mariners currently have a 9 or so game lead in the AL West, and they seem to be doing it by following the Astros’ 2022 playoff path: using pitching to suffocate the opponent. The injured Astros’ pitching staff has produced run prevention that is far from the 2022 version. How do the Astros’ and Mariners’ pitching staffs compare in terms of the difference between ERA and x-ERA?
I downloaded the Astros’ and Mariners’ pitchers individual x-ERA and ERA, and then computed a weighted average (based on inning pitched) difference at the team level. Here are the results:
Astros 4.08 ERA 3.87 x-ERA Difference: +0.21
Mariners 3.31 ERA 3.47 x-ERA Difference: -0.16
The Astros and Mariners are polar opposites in terms of the difference between x-ERA and ERA. The Astros’ expected ERA lies below its actual ERA and the Mariners’ expected ERA lies above its actual ERA. If results were just based on x-ERA, only four-tenths of a run would separate the Astros and Mariners.
For those wondering about statistical correlation, my cursory review indicates that Hard Hit Percent is moderately correlated with the ERA/x-ERA differential for the Mariners’ pitchers and correlated only in a minor way for the Astros’ pitchers. Barrel Rate is even less correlated.
Broadly speaking, I would attribute the difference between ERA and x-ERA partly to luck and partly to fielding run prevention. And there is reason to believe that defense is part of the explanation. Based on the Fielding Bible, Seattle’s team defense is ranked 6th with 30 runs saved so far, while the Astros’ team defense is ranked 16th with 7 runs saved. As I have written previously, if the Astros want to improve the pitching, the cheapest way is to improve the defense.
To the extent that luck (or more formally, random variation) is responsible for the difference between x-ERA and ERA, ordinarily we would expect some probability that the Mariners’ pitching staff ERA would regress upward and the Astros’ pitching staff ERA would regress downward in the future. The Fangraphs ROS runs allowed calculation expects some modest regression in exactly that direction for the Astros’ and Mariners’ run prevention over the rest of the season.
While the hoped-for regression may give some optimism about the Astros’ pitching situation, the injuries and almost non-existent rotation depth adds a considerable degree of uncertainty to our “expectations.”
How much uncertainty surrounds the potential return of Lance McCullers Jr. and Luis Garcia this season? As I suggested in Monday’s column, the Astros’ season may depend on their return to the rotation. How much should the Astros’ front office gamble their trade deadline on that risk? Is Verlander’s current injury a minor blip or does it foretell bigger problems?
Astros Playoff Odds
Not surprisingly, the Astros’ playoff probabilities are dwindling. The severity of the declining odds depends on whether you rely upon Fangraphs or Baseball Prospectus. According to FG, the Astros’ playoff odds currently stand at 29%. FG believes that the Rangers have almost fallen out of the playoff picture (8% odds).
PECOTA, on the other hand, places the Astros playoff odds at 44%, with the Rangers holding similar odds (42%),
Fangraphs and PECOTA have different takes on whether the Astros’ best probability lies in the division race or the Wild Card race. The FG odds break down as 19% division and 11% Wild Card, while the PECOTA odds break out as 28% division and 16% Wild Card.