Finding Baseball’s Most Improved Hitter

Now that we’re close to the end of the season, you’re going to see a lot of talk about the bests in certain categories. There’s a major award that goes to each league’s best pitcher. There’s a major award that goes to each league’s best rookie. There’s a major award that goes to each league’s best player (sort of). And there are further awards that are supposed to go to the best hitters. It’s seldom easy to identify a given season’s best whatever, but at least you can limit yourself to one year of data. That makes things simple, relatively speaking.

It’s half as simple, or twice as complicated, if you double the number of years. But that’s something you have to do if you want to measure improvement. And this is something I’d like to spend a few minutes on — who, among major league baseball players, is the most improved hitter? There’s no great reason to focus on this, I guess, but there’s no great reason to focus on anything. Major improvement is good. Makes for a compelling story. Why not find the most improved hitter? All we need to do is agree on a method.

That’s where it gets difficult. That’s where it always gets difficult. The challenge with these things always boils down to the following: What we have are results. What we don’t have are accurate measures of true, underlying ability. Some results are more indicative than others, but because results and true-talent level aren’t the same thing, there’s always going to be some amount of uncertainty. Can’t be fixed. So it goes. We make our attempts anyway.

For this exercise, let’s look at players who played at least semi-regularly in the majors in each of the last two seasons. There are 230 players who have batted at least 250 times each year. That’ll work as a player pool. I know that it leaves some players out, but it gets only more complicated when you try to fold in the minors.

When you think about finding the most improved hitter, what’s the first thing you might check? You got it — the largest improvement in results. That’s easy. Out of our player pool, the guy who’s improved the most in terms of wRC+ is Ryan Zimmerman. Therefore, Ryan Zimmerman might be the most improved hitter. He has a pretty healthy, double-digit lead over guys like Jake Marisnick and Aaron Hicks. The story checks out. Zimmerman has admittedly cooled off, but still, his overall numbers are impressive, and they’ve given the Nationals a significant boost.

There’s no reason to stop there, though. We already know that wRC+ measures results, and not necessarily talent. If wRC+ is noisy, then two years of wRC+ might have the potential to be extra noisy. What if we were to compare actual 2017 player performance to preseason projected player performance? Those projections could be taken as a better reflection of skill. Now, I don’t have access to wRC+ projections, but I do have access to wOBA projections, which is fine. Looking at the difference between actual wOBA and projected wOBA, the most improved hitter, by far, is Zack Cozart. He’s surpassed his projection by a clean 100 points; the guy in second place is at 85. Interesting!

Still, we can keep going. If projections better reflect actual ability, why not compare preseason projections and updated projections? Those updated projections take 2017 into account, so, there’s a strong case to be made for this method. The most improved hitter is…a tie. Cozart is there, having improved his wOBA projection by 38 points. But Tim Beckham, also, has improved his wOBA projection by 38 points. Maybe that does it, then! Or maybe it doesn’t.

Forget, for a moment, about all-inclusive results. What if we were to focus on process? What could be better than a hitter who’s gotten better at drawing walks, and avoiding strikeouts? Maybe you think this is flimsy, but I should at least make a point to include it. I looked at hitter K-BB%. It’s the same number that we use with pitchers, only, for hitters, you want a smaller K-BB%, instead of a big one. By this measure, the most improved hitter would be Justin Smoak. Blue Jays fans, certainly, could get behind that. Smoak has been one of the rare bright spots in an otherwise miserable season.

Still, there’s another area. If you *really* want to talk about process, why not increase the denominator? Instead of focusing on walks and strikeouts, why not examine swing tendencies? Generally speaking, it’s good to swing at pitches in the zone, and it’s bad to swing at pitches out of the zone. So, generally speaking, it’s good to have a wide difference between in-zone swing rate and out-of-zone swing rate. If you look at those differences in 2016 and 2017, then the most improved hitter, by process, would be Jed Lowrie. Lowrie’s difference has improved by 15 percentage points. Only one other player has improved by even more than eight percentage points. It’s been something awfully dramatic.

You might be wondering something. This is all well and good, but, why not make use of some data from Statcast? Couldn’t that, in theory, tell us even more than the sort of information we’ve had for several years? I’d agree with that, so, as always, I’ve pulled information from Baseball Savant. I first decided to look at average exit velocities. Comparing differences between the last two years, the most improved hitter would be Jed Lowrie. Right behind him is Ketel Marte. Lowrie, though, has now come up twice in a row.

At last, there’s the matter of expected wOBA. That’s also housed at Baseball Savant, and although it doesn’t yet have much of a track record, you can see how expected wOBA might be the best reflection of talent of all. It folds in all the new batted-ball information and, while it doesn’t account for, say, player speed, it at least gives us a useful result, telling us what an average player might’ve done with the same walks, strikeouts, and contact. Comparing differences between the last two years, one more time, the most improved hitter would be Jed Lowrie.

The top five, in terms of improvement in expected wOBA:

1. Jed Lowrie, +0.071
2. Justin Smoak, +0.065
3. Ryan Zimmerman, +0.058
4. Giancarlo Stanton, +0.057
5. Carlos Gomez, +0.056

I’m willing to get on board the Jed Lowrie bandwagon. He’s been a quality major-league hitter before, but he was mediocre in 2014 and 2015, and quite dreadful in 2016. He’s now drawing more walks, and making more contact. He’s hitting for more power, having generated dramatically more lift. It’s happened quietly, but, since the All-Star break, the A’s are tied for the American League lead in team wRC+. Lowrie has factored into that. So have Khris Davis, Matt Chapman, Matt Olson, and Matt Joyce. Whole bunch of Matts. The strength of this year’s A’s was supposed to be the young starting pitching. Instead, the pitching has been poor, and the lineup has become something potent. Okay. At least it hasn’t been dull.

Compared to last year, Lowrie has improved on his WAR by more than four wins. That’s not very common for a 33-year-old. And the story here is fairly important, even for its banality. What’s been behind Lowrie’s resurgence? He’s just been able to play as a healthy ballplayer. He doesn’t have 2014’s fractured finger. He doesn’t have 2015’s torn ligament. He doesn’t have 2016’s torn capsule. Lowrie has long had these skills — health has just allowed him to tap into them more consistently. It’s simple and obvious and crucial. Then there’s one more, more unusual twist: As Lowrie told Eno a few months ago, he’s getting far better sleep, after having had repair for a deviated septum. It all folds together. We can’t actually test the difference that better sleep might be making, but, in short, Jed Lowrie has gotten a lot better because, for the first time in years, his body is functioning the way it was designed to function.

So, there, you potentially have it. If we’re just comparing 2016 and 2017, then 2017’s most improved hitter would appear to be a 33-year-old Jed Lowrie. Or it’s just Aaron Judge, if we want to think about everyone. It’s probably actually Aaron Judge. Have you seen what he’s been doing? My goodness. But Lowrie is pretty good too. His has been a different and more subtle form of improvement.