Rafael Devers Is Still a Work In Progress
Generally speaking, doing more of one thing means doing less of another. A positive development can come at the expense of some other attribute and unintended consequences can make that positive thing decidedly less so. It’s one thing to strike out less often. Putting more balls in play provides the potential for positive outcomes. It’s another thing to strike out less often and walk more often. Doing those things in concert can have a great effect on getting on base and not making outs, which is almost always the point of every plate appearance from a hitter’s perspective. Sometimes, doing both doesn’t lead to positive results, and Rafael Devers found that out the first month of the season. Instead of going back, though, he’s moved forward and is hitting better than ever.
The 2018 season wasn’t a great one for Devers statistically. He wasn’t terrible, posting a slightly lower than average 8% walk rate and 24% strikeout rate, as well as better than average power with a .193 ISO, but a .281 BABIP meant a .240 batting average a sub-.300 on-base percentage and a below-average 90 wRC+. In his 2017 debut, Devers’ numbers were mostly the same, but a .342 BABIP meant a 110 wRC+. The Statcast data indicates that the difference between 2017 and 2018 was mostly luck, and Devers’ batted ball profile in terms of ground balls, line drives, and pulled balls were all pretty similar in 2017 and 2018, giving credence to Devers being a bit lucky in 2017, with 2018 his natural level if all else remained the same.
Devers wasn’t content with remaining the same, so he spent the offseason working on his weight and now focuses more on not striking out and incorporating video of pitchers in his preparation. While all those changes are admirable, they failed to make a difference the first month of the season. On April 25, Devers had completed 101 of his 201 plate appearances this season. True to his word, Devers struck out a lot less, lowering his strikeout rate down to 16% on the season and saw his walk rate rise to 12%. Devers also saw his BABIP rise to .338 so the lowered strikeouts and increased walks caused his batting average to stand at a healthy .276 with a very good .370 OBP. Unfortunately, Devers had yet to hit a home run and only had six doubles, leaving his ISO at a meager .069 and his wRC+ right at 90, the same as where it was the previous season.
Going for more walks and fewer strikeouts took away Devers’ power and any gains he made were neutralized. Statcast metrics indicated the same as Devers’ xwOBA remained almost the same. Devers was different, but he wasn’t better, as the table below shows.
| BB% | K% | BABIP | ISO | GB% | Pull% | wRC+ | |
|---|---|---|---|---|---|---|---|
| 2017-2018 | 7.7% | 24.4% | .301 | .195 | 47.2% | 36.3% | 97 |
| 3/28-4/25/19 | 11.9% | 15.8% | .338 | .069 | 57.7% | 38.9% | 96 |
Devers’ focus on putting the ball in play meant he was putting the ball in play with considerably less authority. His exit velocity and xwOBA were down on batted balls from previous seasons, and even if he were hitting the ball harder, so many more ground balls leads to fewer extra-base hits. Over the last month, Devers has changed again. He’s kept many of the gains relating to strikeouts, but he has begun putting the ball in the air a lot more often and started putting the ball over the fence.
| BB% | K% | BABIP | ISO | GB% | Pull% | wRC+ | |
|---|---|---|---|---|---|---|---|
| 2017-2018 | 7.7% | 24.4% | .301 | .195 | 47.2% | 36.3% | 97 |
| 3/28-4/25/19 | 11.9% | 15.8% | .338 | .069 | 57.7% | 38.9% | 96 |
| Since 4/26 | 7.5% | 14.2% | .364 | .247 | 37.3% | 47.0% | 155 |
It’s necessary to point out that much of Devers’ big month is due to a .364 BABIP, but even with a .300 BABIP, Dever would be hitting .299/.349/.546 and about 30% above league average, which is about where his season line of .315/.383/.478 and 126 wRC+ puts him. As the ground ball and pull numbers indicate, Devers is doing a better job of getting the ball in the air and pulling it with authority. Here’s an inside fastball to Devers from April.
Now, here’s one from the other night.
Devers moved to a more open stance on April 15, and instead of doing a toe tap like he did early in the season, he just keeps his foot in the air before planting it. While swinging his leg back and forth might seem more complicated, he gets his foot planted earlier so that his swing is more balanced and he gets more force behind it. Here’s another view on a different pitch and location, a slider away from early in the season.
And here’s a recent cutter in a similar spot.
Devers’ swing is stronger and slightly more balanced so that he was able to get under the pitch and poke it into right field. It’s a tough pitch to hit in both instances, but Devers’ ability to put it in the air caused a base hit instead of a grounder. It’s too early to say whether he can keep this going, but here are Devers’ plate discipline numbers.
| 1st Stk% | SwStrk% | O-Swing% | Z-Swing% | O-Contact% | Z-Contact% | Zone% | wRC+ | |
|---|---|---|---|---|---|---|---|---|
| 2017-2018 | 64.1% | 13.0% | 37.1% | 68.9% | 64.0% | 82.0% | 45.5% | 97 |
| 3/28-4/25/19 | 48.5% | 10.9% | 32.5% | 70.7% | 69.3% | 82.9% | 43.0% | 96 |
| Since 4/26 | 58.5% | 12.1% | 41.7% | 73.3% | 69.4% | 83.9% | 44.1% | 155 |
That first strike percentage jumps out a bit, as it went way down to start the year before going back up. At the beginning of the season, Devers didn’t swing at a whole lot of first pitches, but over the last month, that percentage has jumped up, causing more first-pitch strikes but also three of his six homers. Devers is swinging at a lot more pitches, particularly those pitches out of the strike zone, but he’s been able to keep making contact and do relatively well on those pitches in the last month. Not getting enough pitches in the zone to hit and making too much contact on hard-to-reach pitches out of the zone is a good way to get into a slump, so there are likely more adjustments pitchers can make against Devers.
Whether Devers can continue to adjust will determine if the last month can be more representative of his talents going forward. He’s still just 22 years old and he made changes to make himself better this season, but when those changes weren’t working, he adjusted instead of going back to his old ways. Those adjustments have made a big difference over the last month, and his willingness to change for the better is a good sign going forward.
Craig Edwards can be found on twitter @craigjedwards.
Devers had ten multi hit games in May and has homered in three straight. As this stretch began, the (dumb) question was whether the lack of balls over the fence was a problem.
Along the lines of what I said in the Chavis thread: strength of schedule is a factor here. Devers had a brief 5-game cooling period: 2-22 against COL and HOU. The damage this month has largely come against CHW, SEA, BAL and TOR. You can’t just throw the numbers out, but there are more awful pitchers on these teams than not (article on BAL’s HR allowed pace please?).
It’s just especially noticeable this year because of what the mid April narratives were. You even had players commenting that they might have to trade guys at the deadline. I’m skeptical when the narratives paint such a stark contrast just 4-6 weeks later. Maybe some crappy pitching is just what the doctor ordered. In any case, people including myself often forget Devers is only 22. He’s almost a year and a half younger than Chavis, with a year and a half more MLB experience.
Devers is a promising young player, but Chavis could be anything! He could even be Devers!
FWIW, I think you’re starting from a false premise. The opening implies that reducing one’s K% correlates with reducing BB%. For this year’s K% reducers, at least, the opposite is true — their collective BB% is up, not down. Bellinger, DeJong, H.Dozier — most of the major K% improvers have higher BB%.
I believe the “more of one thing and less of another” concept is referring to Devers’ early season lack of power – on the year his BB% is up as well, which is discussed in the article.
I responded to this passage: “It’s one thing to strike out less often…. It’s another thing to strike out less often and walk more often.” The latter pairing is actually the norm.
I don’t know that we can talk about BABIP, power, and the Red Sox without discussing Fenway Park and its bizarre dimensions. He is batting average is 55 points higher at home than on the road, but he’s got twice as many homers on the road as at home. Road-Devers, over 600 PAs, hits .290 with 24 home runs over 600 PAs. Home-Devers is more like .345 with 13 homers. He’s basically hitting like “good” Joe Mauer at home, and “good” Kendrys Morales on the road.
I don’t think this negates anything said in the article, but there are some good reasons why he’s got such a weird profile right now.
it’s simple: don’t hit groundballs, hit homers. pull the ball in air.
It seems like if he can manage to get the ball in the air more he has the potential to be consistently productive. His exit velocity is really good, but his average launch angle is terrible.
What a great read. Love watching this kid hit and will enjoy following his continued development over the coming years. Thanks for writing this.
I’m just not sure the K% isn’t due for regression. The SwgStk% is in line with career average and he isn’t making more contact. Sure, he is swinging more often, but unless he had some serious problems with called Ks, I don’t see a sign that anything has really changed.
I see him as more of a Nicholas Castellanos upside with better defense.
He has improved his SwStr%, though.
2018: 13.1%
2019: 11.3%
That’s almost a 14% improvement over 2018. It’s true that his career rate doesn’t make the improvement look as significant, but that’s because his career rate includes his improvements this year.
His Contact% is up a few points as well.
It doesn’t look like a swing overhaul or anything, just some small changes.
Not to be pedantic, but ok I’m being pedantic. That’s not how percentages work. It’s a 1.8% improvement, not a 14% improvement. A change from 13.1% to 11.3% is not the same thing as a change from 100% to 86%.
Re: defense–Castellanos must be really, really bad out there. Though the fielding issues with Devers appear to be mental. He seems to have the athleticism and arm for third.
There’s star potential here. Not that average third baseman who hits like Castellanos would be a bad outcome.
Players on THE BOARD top 50, who are older than Devers (22y 7m), followed by either peak MILB level or how many fewer PA and WAR they have accumulated in MLB so far:
(7) Nick Senzel – 15 months older – 852 fewer PA, 3.2 fewer WAR
(13) Keston Hiura – 2 months older – 902 fewer PA, 3.1 fewer WAR
(14) Brendan McKay – 9 months older, peak level AA
(25) Brent Honeywell – 18 months older, peak level AAA
(28) Brendan Rodgers – 2 months older, 921 fewer PA, 3.3 fewer WAR
(29) Michael Kopech – 6 months older, 14.1 MLB IP, 3.4 fewer WAR
(34) Chris Paddack – 9 months older, 51.1 MLB IP, 1.9 fewer WAR
(35) Sean Murphy – 24 months older, peak level AAA
(37) Mitch Keller – 6 months older, peak level AAA
(39) Kyle Wright – 12 months older, 20 MLB IP, 3.8 fewer WAR
(40) AJ Puk – 18 months older, peak level AA
(45) Alex Reyes – 26 months older, 53 MLB IP, 2.3 fewer WAR
(46) Brandon Lowe – 27 months older, 617 fewer PA, .9 fewer WAR
(47) Danny Jansen – 18 months older, 716 fewer PA, 2.3 fewer WAR
(48) Peter Alonso – 22 months older, 737 fewer PA, 1.7 fewer WAR
(50) Corbin Martin – 10 months older, 12.2 MLB IP, 3.3 fewer WAR
It’s easy to overlook in this pre-arb, prospect-centric age, just how hard it is to even be in the bigs as a competent regular at age 20. Virtually the entire current board will not be both in MLB and keeping heads above water as they do it.
Oh, and he hit the first LvL off Aroldis Chapman in six years on a 102.8 MPH pitch, what, like three weeks into his career?
https://m.youtube.com/watch?v=ZzwWoIhe11U