The most soul-crushing storyline of the Astros’ 2016 season has been Dallas Keuchel’s early devolution from Cy-Young Award Winner into Lucas Harrell. And if it hurts to read that, it should.
Keuchel entered play on Monday with a team-worst 5.32 ERA. The Crawfish Boxes has spilled no little amount of digital ink discussing the Astros’ ace’s ailing aura. (Also, check out reader rattler03’s excellent look at Keuchel’s season in the Fan Posts.)
Results are what matter most, and there’s no getting around the fact that Keuchel’s have been poor this season. He has allowed three or more runs in ten of his fifteen starts this season, something he did only eleven times all of last year.
Regardless of his performance, the Astros are still very much in a playoff hunt. It’s easy to see the impact that their best pitcher’s return to form could have during the second half of the season. Looking forward, the pertinent question is: “Will he continue being this awful?”
Luckily, we have math and stuff to answer that question.
The first thing to point out is that Keuchel has had the worst run support of any starter in the major leagues. Regardless of his ERA, the Astros aren’t scoring during his games. Of the fifty-nine pitchers who have thrown at least sixty innings so far during 2016, Keuchel ranks fifty-ninth in runs scored per game, at 2.53 runs. To put that in context, the Twins’ Ervin Santana also has little support, ranking 49th on the list. But Santana’s team has scored a full run per game more than the Astros have for Keuchel.
Astros Runs Scored per Game, by pitcher:
- Mike Fiers (5.38 runs)
- Collin McHugh (4.86 runs)
- Doug Fister (4.62 runs)
- Dallas Keuchel (2.53 runs)
The Astros have scored fewer than half as many runs for Keuchel as they have for Fiers. That is not a trend that can continue, because there is no reason for it to. The same guys typically bat every day, regardless of who is pitching.
Also, when it comes to the act of throwing the baseball with the intention of screwing up batters, Keuchel hasn’t been too bad this season.
Smart people have used regression analysis (comparisons to historical seasons) to determine what a player’s ERA should be based only on things he can control: namely, strikeouts, walks, batted ball data, and home runs. These metrics eliminate the vagaries of fielding, weather, park effects, and batter luck to help us judge two important things:
- Is something wrong with the pitcher?
- Can we expect him to get better or worse?
FIP, xFIP, and SIERA are three stats that use differing methodologies to create a “hypothetical ERA” that can be used to answer those questions.
Dallas Keuchel, 2016:
So while Keuchel has certainly not pitched as well as he did in 2014 or 2015, his actual execution should have put him a lot closer to those levels than his ERA indicates. That bodes well for the immediate future.
The last two stats are relevant in answering the two questions as well. Over decades of baseball history, BABIP and Home Run rate have been proven to be largely (not entirely, but mostly) outside of a pitcher’s control. BABIP likes to find it’s level around .300, and HR/FB around 10%. Since both of Keuchel’s values are so high, much of his runs-allowed results can be attributed (generally speaking) to bad luck.
None of this erases the fifteen games of overall stinkitude that Keuchel has graced the fans with this season. Those don’t go away, and he doesn’t get a do-over. Nor should anybody excuse that performance or pretend like it didn’t happen by pointing at other metrics besides ERA and Losses.
However, fans should now be more interested in what they can expect from Keuchel moving forward. A likely correction in run support, plus better luck on balls in play and fly balls, should herald the return of the Keuchel who carried the Astros on the strength of his left arm for the past two seasons, and greatly increase the club’s chance at reaching the playoffs again.