What’s Eating Dallas Keuchel?

If you’re reading this, the disappointment in Dallas Keuchel’s follow-up to his stellar opening act with the White Sox probably doesn’t need to be articulated to you in great depth.

A year after finishing fifth in AL Cy Young balloting, Keuchel finds himself with an unimpressive 4.44 ERA (and even less impressive 5.10 FIP) more than two-thirds through the 2021 campaign, with less-than-inspiring peripherals to boot. The team has lost six of his last eight starts, and as of late, even Keuchel’s successful outings have a slight discomfort to them.

It’s hard to be dissatisfied with the six innings of two-run ball that Keuchel delivered against Kansas City in his most recent start — the ire for that loss should be rightfully held for the slumping offense — but it was far from a dominant six innings, and it felt like a better opponent might have walked away with a few more scratches.

Keuchel has acknowledged that to some extent, there’s been a tradeoff of performance for reliability this season. As the rest of the starting rotation racks up innings in a season unlike nearly any other, his role as someone able to start every fifth day is an important one even if the results aren’t there.

That doesn’t make the losses and blown leads any more satisfying to watch, unfortunately. Ironically, in spite of the stability he provides in some senses, the second of Keuchel’s three guaranteed seasons with the Sox has been a pure roller coaster, and pitching every fifth day probably doesn’t explain all of it. Until just now, the track was in a deep valley: Over the seven games prior to Thursday’s slide-stopper, dating back to his “relief” appearance in the first half of June 20’s doubleheader against Seattle, Keuchel had been hammered for a ghastly 6.17 ERA, allowing 42 hits and 13 walks compared to just 25 strikeouts over 35 innings. That followed up an 11-start stretch from mid-April to June in which he was good for a 3.20 ERA and less than a hit per inning. And that, of course, followed up a bumpy three-start opener to the season that had many of us asking what happened to the Cy Young candidate who almost got them to the ALDS last season.

In a literal sense, the short answer is that nothing happened. On the level of “process” over “results,” Keuchel is more or less the same pitcher as he was last season. This is who he is, and the 3.20 ERA pitcher and the 6.17 ERA pitcher aren’t some kind of Jekyll and Hyde. They’re just two ends of a spectrum that happens to be much wider than the one most other pitchers occupy.

The most frequently-cited axis that the spectrum lies on is the steep decline of Keuchel’s effectiveness as he faces hitters for the second or third time. The numbers are quite ugly, as Keuchel acknowledged in the Athletic article linked above.

Baseball-Reference

There’s a somewhat-nuanced story to be told here about game management and getting the most out of an inherently-limited pitcher.

Keuchel has actually fared quite well in the few instances he’s worked into the seventh inning or later, allowing just four of 22 hitters faced to reach base. That’s an excellent example of survivor bias, as the only times that Keuchel works that deep into the game are the times when he’s completely locked in and in a groove. Those aren’t the games that we’re worried about, of course. But it’s nice to have some indicators that Keuchel is indeed still capable of carrying a team to a win now and then.

Those 22 batters faced in the seventh and eighth innings have come over just five appearances in 21 starts, after doing so in three of 11 tries last year. That passes the smell test. It stands to reason that Keuchel is still good for a solid seven-inning effort about a quarter of the time. With an $18 million salary this year and next, it might not be what fans want to hear, but the reality of having an 88 mph fastball without Kyle Hendricks-level command and feel is that there needs to be a heightened sense for when Keuchel’s effectiveness starts to slip. More often than not, it’s just not going to be wise to try to push him through six or seven innings unless he’s in the zone. Keuchel is just not that kind of pitcher any longer.

The key difference between 2020 and 2021 is that Keuchel’s effectiveness seems to be slipping with more and more rapidity, but managerial strategy has not shifted as a result. Tony La Russa isn’t pushing Keuchel into the sixth and seventh any more than Ricky Renteria was — he’s doing so at pretty much the same rate. But it’s also clear that Keuchel isn’t working at the same level of effectiveness as last season. Keuchel didn’t give up a run in the fifth inning in any single start last season. In fact, he only gave up four earned runs in total in innings one through five. Now that is earning an opportunity to pitch deep into the game. This year? Not so much the case.

Baseball-Reference

Between that and the per-inning numbers earlier, it’s easy to see how things are starting to come undone in the fourth and fifth innings much more often than they used to. The third time through the order has always been a problem for Keuchel, but his second time through has been rougher than it’s even been this year. Yet in 2021 as in 2020, Keuchel has been given the opportunity to pitch the sixth inning in about 72% of his starts. Having given up 23 earned runs in the fourth and fifth innings this year, that number probably needs to come down if Keuchel’s ERA is going to be salvaged.

Getting him out of the game at the right time wasn’t the only reason Keuchel was effective last year. He had a fair share of batted-ball luck go his way. Over a full season, his .255 BABIP almost certainly would have risen; in terms of contact quality, Statcast didn’t see a drastic departure from his also-abbreviated 2019 season. Keuchel also almost certainly would have given up a home run on more than 4.7% of his fly balls, an unsustainably low rate, and at more than a one-every-18-innings pace, setting a career-low with just 0.5 HR/9. It was the first time since Keuchel’s Cy Young campaign that he was able to keep such a low BABIP and a low home run rate simultaneously. His luck was probably bound to turn the other way, even if everything else went right.

This isn’t to say Keuchel is completely at the mercy of lady luck, but it does highlight the importance of good management when dealing with a pitcher like Keuchel. He’s still got the talent to put up very good numbers, but he won’t do that if he’s being hung out to dry in the middle innings. What’s the point of a deep bullpen if you’re not going to try to use it to win games?

This doesn’t mean to diminish Keuchel by rattling off his flaws. This entire premise takes for granted that there’s a better-quality pitcher here than the one that’s been seen on aggregate. The reality is just that pitchers who work like Keuchel — low velocity, lots of contact — have a thin enough margin for error that it’s really hard to put up a full season of results like the ones he delivered for two months of 2020. Sox fans might be tempted to point at Mark Buehrle’s almost-supernatural consistency in disagreement, but as a general rule, it’s pretty concrete.

It’s definitely possible to succeed without high velocity. Not every pitcher has to be Michael Kopech, and that’s one of the things that makes baseball great. But even by the standards of soft-tossing lefties, though, Keuchel has entered dangerous territory. Looking at his stats since 2017 (his last All-Star season), Keuchel’s skill at stealing strikeouts and avoiding walks has diminished to a place where it’s hard to see a consistent six-inning starter.

FanGraphs

Again, contact-oriented pitchers shouldn’t be legislated out of the game, nor should the only pitchers worth watching or acquiring have to throw in the high-90s and rack up strikeouts. It’s just hard to emphasize how much of a tightrope this is.

Since 2015, there have been 20 individual seasons in which a pitcher started 20 games and struck out less than 15% of batters faced while walking more than 7%. It’s not a particularly impressive list.

It’s brutal out here. Pitchers with this kind of skill set usually aren’t expected to hold down a roster spot for a full season, much less make starts in a playoff series. Nobody wants to relive the Mike Pelfrey Experience this coming October.

There’s also no limit to the ways in which stats can be cherry-picked to support any chosen narrative, so let’s be a little more fair to Dallas. He might strike fewer hitters out than most big-name free agent signings, but he’s also one of the most accomplished ground ball pitchers of the past decade. Pelfrey certainly wasn’t that.

Keuchel isn’t rolling wormburners at a 60% clip like he was earlier in the decade, but even if his true talent looks more like a 4 ERA than a 2 ERA at this point in his career, there’s no reason he can’t put up better numbers than he is now if he’s not put in a position to fail. Keuchel may have gotten out of the sixth inning unscathed on Thursday, but not without bated breath, as Carlos Santana squared him up for a 110 mph single. Against the Astros, Yankees, Red Sox, or Rays, Keuchel will have better bats to deal with after that than Hunter Dozier and Edward Olivares.

Both Keuchel and White Sox fans deserve better than to have him thrust into the role of James Shields in 2018. Being a stalwart in the rotation doesn’t mean he needs to just wear it into the seventh inning on a day when he doesn’t have it. Again, what’s the point of bullpen depth if it’s not to be used?

If, as Keuchel himself was quoted, he usually “has it” for the first 85 pitches or so, why push him beyond that? If he’s having one of those days where he’s painting corners like van Eyck, then let him ride. If not, thank him for keeping the team in the game through four or five innings and turn it over to someone who throws 98 mph. That will do plenty to win them games down the stretch in the regular season — games they’ve been losing quite a bit more than we’d like — and perhaps even beyond.