clock menu more-arrow no yes

Filed under:

Projecting (half of?) 2019: Dallas Keuchel

New, 56 comments

I’m guessing the Braves are guessing they have a pretty good sense of what they’re getting.

Atlanta Braves v Houston Astros Photo by Stacy Revere/Getty Images

In the offseason, I committed myself to doing a “Projecting 2019: [player name]” post for every notable acquisition the Braves made. That commitment made some (hopefully interesting) hay when the Braves quickly pounced on Josh Donaldson and Brian McCann, as well as a reprisal I wasn’t too keen on with the later signing of Nick Markakis. ...Aaaaand... that was it. Until now, anyway. It’s June 7, and last night, the Braves added Dallas Keuchel to their organization on a one-year deal.

The 31-year-old Keuchel has been an above-average pitcher since the start of the 2014 season, and if not for an elevated case of the dingers, perhaps since the start of 2013 as well. (His rookie season, in 2012, consisted of an awful 85 innings across 16 starts.) From 2013 through 2018, he was in the top 20 among all starting pitchers in innings pitched, fWAR, and RA9-WAR. His success wasn’t just a durability play either, as among the 200 or so “qualified” starters in that span, he finished 18th in xFIP-, 29th in FIP-, and 36th in ERA-. Basically, he’s been something like a #1 or #2 starter, on average, since 2013.

After a Cy Young campaign where he finished fifth in MLB in fWAR and fourth in RA9-WAR in 2015, he backslid over the following two seasons, but remained a solidly above-average run preventer, aside from some poor ball-in-play and sequencing fortune in 2016. His most recent season was mostly the norm for him at this point: 88 FIP- compared to 91 in his career; a somewhat-elevated-but-still-quite-grokkable 91 xFIP- owing to his lowest groundball rate since his terrible rookie half-season.

(There’s probably an interesting discussion to be had here about whether Keuchel’s pitch-to-contact tendencies make him especially prone to ERA vs. FIP deviation — in five of six full seasons, his fWAR and RA9-WAR have differed by at least one win, and have differed by closer to two full wins in three of those six seasons, generally balancing each other out as time goes by.)

Take all of this together and hurl it, churn it, make a finger painting of it all you want. In the end, it all comes out as Keuchel being an above-average starting pitching human, and that’s basically what the projection systems roll with as well.

The expectation is, as ever, above-average run prevention, leading to a 3ish wins per 200 innings pitched rate. At this point I’m not sure exactly what run environment inputs are governing all three systems (IWAG is just using its preseason assumptions for 2019, as opposed to any in-season adjustments; also remember that the FIP-s presented here are my own calculations since they’re not presented for projections on the Fangraphs player pages), so I wouldn’t worry about the minute differences and contradictions between IWAG forecasting a better FIP- but a lower WAR accumulation rate as a result.

The one weird thing is Steamer’s forecast, which somehow pro-rates to four-plus wins over a full season. Looking at current starters with FIP-s around 90, I don’t see any way in which that translates to a four-win season, even with really aggressive infield pop rates (which Keuchel doesn’t have anyway) or pitch framing detriments (which the Braves won’t deliver). This may be ironed out in the next day or so when someone catches it, but it’s what I’ve got to work with for now. Somewhere in the low three-to-four-wins-over-200-IP range seems the safest bet, I think. The IWAG distribution curve is pretty boring, mostly flooring Keuchel at average run prevention and limiting him to upside short of four wins, aside from a little reserved possibility that he is able to prevent runs at a Cy Young-type rate once again.

Of course, there are two elephants in the projection system server room that are very much worth noting:

  • The above chart is a full-season projection. Keuchel won’t pitch a full season. How much he will pitch is currently unknown. If he debuts in July, that’s neatly a half-season threshold, so you can simply cut the numbers above in half and go on your merry way. If he debuts later, a bigger proportional cut is warranted. The later he debuts, the fewer opportunities he’ll have to rack up wins with his above-average run prevention.
  • There isn’t a huge track record of players signing and debuting midseason, and it’s even shorter for pitchers. Will this affect Keuchel in some way, or will he simply be Dallas Keuchel: Late Edition? Not only do I not know, I don’t really even have a guess. If he were simply signed in the offseason, I’d have a fair bit of confidence in his projection, but that’s not the case. Unfortunately, this hasn’t even happened frequently enough to know what it does to the variance associated with any projection, so... as with everything else, I guess we’ll just find out. Eventually.

In any case, the Braves added some wins, and that’s pretty cool. I checked the Fangraphs Depth Charts for the remainder of the season before the signing was announced last night and they had the Braves’ rotation as 17th-best going forward (albeit with stuff like Mike Soroka only throwing 80 or so innings for the rest of the year; don’t throw stuff at me, I didn’t do it) and it’s now jumped up to 15th, so that’s something. However, that jump appears to use the Steamer assumption of +2 wins in 100 IP, so maybe it’s a something that’s a little specious? We’ll see.


Projections aside, it’ll be interesting to track which version of Keuchel the Braves end up getting. There’s a ton that goes on with Keuchel year-to-year in terms of pitch mix, movement, and how it affects hitters’ approaches and results against him, but 2018 featured less break and more contact for him across the board, though it didn’t really affect him that much because he simply managed to transform the prior whiffs on balls outside the zone into weak contact instead, and the walk rate dropped as well as balls were put in play rather than extending plate appearances. Whether the re-juiced ball puts a crimp in this strategy, or whether Keuchel simply shows up doing something different, will be worth watching. However, it’s likely that with his addition, the Braves will have two launch angle destroyers in their rotation: Keuchel had a negative average launch angle against him in 2017, which is mind-boggling. Still, if you wanted a reason to worry (what’s wrong with you), Keuchel’s continued ability to succeed while pitching to contact should be your huckleberry here, as everything but his cutter (which he threw a lot more in 2018 than ever before) got worse in either xwOBA-against or whiff rate (though not necessarily both) in his most recent season.