Rafael Devers’ historically lucky 2017
Rafael Devers had one of the luckiest seasons ever against LHP
Featured image courtesy of Zimbio.com: (Oct. 4, 2017 – Source: Bob Levey/Getty Images North America)
So, I noticed this weird thing a while ago and planned on writing about it for some time. Do you remember in late-August when left-handed hitting Rafael Devers homered off of left-handed pitcher Aroldis Chapman to tie the game in the top 9th? Everyone watching was stunned and rightfully so. Not only did the rookie third basemen save the game in a critical AL East showdown, but he hit one of the most improbable home runs ever.
Chapman had not surrendered a home run to a left-handed batter since June 2011 against Luke Scott. Okay, so that, in itself, is pretty bonkers. Devers had also only been in the majors for less than a month and it was his first plate appearance against the dominant southpaw, augmenting the “unusual” (amazing) narrative. Further, it was the fastest pitch, clocked in at 102.8 mph, to be hit out of the park in the pitch-tracking era (2008). It was probably the highlight of the Red Sox’ 2017 season.
You get the picture and, honestly, probably know about this already, anyway. So, why am I rehashing this? Well, because this lefty versus lefty battle, taking place on August 13th, 2017, is the perfect representation of Devers’ historically dominant season against lefties.
Last season, he posted a 188 wRC+ (100 is league average) against LHP in 57 tries, ranking 12th in baseball among players with at least 50 plate appearance against southpaws. For those of you unaware, here’s the definition of WRC+ over at MLB.com.
wRC+ takes the statistic Runs Created and adjusts that number to account for important external factors — like ballpark or era. It’s adjusted, so a wRC+ of 100 is league average and 150 would be 50 percent above league average.
12th-place in anything in baseball is probably impressive but does not resemble anything close to historic. The uniqueness of this stat is shown when you put it in the context of lefty versus lefty. Among all left-handed hitters in 2017, Devers, with his 188 wRC+, was the best against left-handed pitchers, minimum of 50 plate appearances. The closest to him was Lonnie Chisenhall, sporting a comparatively less impressive 164 wRC+ with roughly the same amount of trips to the plate. Chisenhall ranked 26th in baseball in the same category that Devers was 12th.
The former Red Sox top prospect’s 188 mark is not just the best in 2017. In fact, you have to go all the way back to 2008 to find a better vs. L as L offensive season when Ian Stewart narrowly edged him with his 189 wRC+ in 60 plate appearances. In other words, in the last nine years Devers has led all left-handed hitters in wRC+ against their pitching counterparts.
The disclaimer, here, being this is using the somewhat arbitrary cutoff of 50 plate appearances. Obviously, it is much easier to perform at historical levels over a smaller sample size than it is to sustain over the course of a full season. Still, Devers being this good against lefties is worth mentioning and further investigation. Can he be one of the premier southpaw mashers going forward, both relative to the league and his fellow left-handed batters?
Unfortunately, we only have one partial season of major league data to examine. Fortunately, though, we have a couple of years of minor league data to look at! To retrieve a decent sample, I will reveal vs. LHP/RHP splits for levels where Devers accumulated at least 50 at-bats against each type of pitching. And, yes, I’m using at-bats because MILB.com does not list plate appearances and I am too lazy to figure that out. All statistics below are courtesy of, you guessed it, MILB.com:
vs. RHP (216 AB): .306 AVG/.368 OBP/.625 SLG/.993 OPS
vs. LHP (71 AB): .282 AVG/.370 OBP/.423 SLG/.793 OPS
Nope, not even close to the 1.074 OPS he posted against lefties in the majors in ’17. This is the most recent season we have of somewhat telling data, so this is not encouraging. Then again, maybe it’s aberration. Players have those all the time. Let’s look at more levels.
2016 Single-A (Advanced):
vs. RHP (352 AB): .284 AVG/.326 OBP/.474 SLG/.801 OPS
vs. LHP (151 AB): .278 AVG/.355 OBP/.371 SLG/.726 OPS
We have a larger sample and we get inferior results against inferior competition. Granted, someone has to take into account the development of Devers as a player and hitter since then, so these numbers aren’t gospel. What they are, however, is disconcerting regarding his ability to sustain his ’17 results.
An interetsing pattern that is beginning to emerge is that Devers seems to have a more patient approach against lefties. In both levels, his OBP were superior against lefties opposed to righties, in spite of the fact he hit for a better average against right-handers at both levels. Further evidence of this, was his 12.3 BB% against LHP compared to his 6.0 BB% against RHP in the majors last year. Here’s one more year of results:
vs. RHP (334 AB): .302 AVG/.343 OBP/.479 SLG/.822 OPS
vs. LHP (135 AB): .252 AVG/.297 OBP/.356 SLG/.652 OPS
His 2014 rookie ball numbers against lefties are even worse, if you needed that tidbit to convince you of this increasingly strong axiom: Rafael Devers has never demonstrated excellent results against left-handed pitching before his brief stint in the MLB last year.
I mean, his 188 wRC+ was always unsustainable but there’s a lot of information that’s screaming he may not even be an above-average hitter against lefties going forward. This is bad, because if you strip away his numbers against southpaws, or even regress them to normal levels, we would have had a below-average hitter last season. It is always wise to view the body of work as more predictive of future events, but there’s also the possibility Devers adjusted his approach at the major league level, putting him in a better position to succeed against lefties. Let’s see if we can find anything.
First of all, the fact he struck out 29.3% of the time against left-handed hitters does not help his case. As previously mentioned, he did walk quite a bit but striking out nearly a third of the time made him very reliable on quality, or lucky, contact to post his 188 wRC+. In terms of quality, his 36.4 Hard Contact% against lefties, per Fangraphs, was only marginally better than his 34.1 Hard Contact% against righties. Meanwhile, his 18.2 Soft Contact% was the same facing righties and lefties. Based on those, numbers he should have produced offensive numbers somewhat similar versus righties and lefties. Instead, his 87 wRC+ vs. RHP was a far-cry from his 188 wRC+ vs. LHP.
So, what was the difference? Well, my friends, the large discrepancy was achieved, quite simply, through luck. Devers’ BABIP (batting average on balls in play) was a confounding .581 against southpaws compared to a .282 BABIP against righties. The league average BABIP was .302 against left-handed pitching last year.
.581!? Sure, it was accumulated in a mere 57 plate appearances, but that number is historical, and I am not just italicizing that for hyperbole. Through Fangraphs, I was able to sort through every year since 2000, using 50 plate appearances as the minimum, to see the highest BABIPs put up against left-handed pitching in a given year. No one has posted a higher BABIP against LHP than Devers since 2000. Quite possibly, further than 2000. It was more difficult to find this information past that season.
That is startling and, relative to 2000 and probably further, historical. The crazy thing was no one else was really close, either. Melvin Upton Jr. 2004 season, where he managed a .533 BABIP in 66 PA, was the closest season to Devers since 2000.
Historical relevance — or irrelevance, if you really don’t care — aside, this does not bode well for Devers 2018 outlook against southpaws. He got historically lucky against lefties in ’17 and there is no evidence that this is remotely sustainable. Additionally, Devers has no track record of really even hitting LHP well. Sure, he seems to have an augmented sense of plate discipline opposing them, but that does not make up for the, at least, inevitable .200 point decrease in BABIP he will see against left-handers if he were to play a roughly full season in 2018.
Yeah, maybe you shouldn’t expect Devers to take an Aroldis Chapman fastball deep again next year. Would be pretty sweet if he did, though.