Let’s build an optimal Paul DeJong

Quick preface: two hours ago, John Fleming let you know he was leaving Viva El Birdos. I want to say how sorry I am that he’s going, and how much I’ll miss his work. I’ve been around VEB forever but have only been able to see behind the curtain for about nine months, and in that time one of the main things I learned is that John loved this place and worked his ass off for it. If you want proof, look back over the posts written during the month-long gap between Craig leaving to go full-time at FanGraphs and SBN naming a new site manager: John wrote the lion’s share and then some. The rest of us tried to give what extra effort we could, but if it felt like John was writing every damn day it’s because he was. And when he couldn’t, he was the one making sure someone would. He did that because he wanted VEB to stay VEB, and to not drop off for even a week. Because he loved VEB and took it seriously.

John, thanks for everything, and you’ll be sorely missed. I’m glad I got to work with you. I hope everybody here recognizes what they’re losing. Be well.

***************************

Now, grudgingly, here’s some stuff about a ballplayer:

When he was in the low minors, you only knew his name if you were (like me) paying too much attention. Looking at his minor-league lines, he seems to have above-average power for sure, with questions about how it’d play in the majors because of iffy discipline. It was clear all along that he likes to swing the bat, first and foremost. And though he mostly played a corner early on, coaches in the high minors let him play an up-the-middle defensive position. There were some questions about whether he could stick there, but it did give a little more carry to his profile.

Overall, by the time he was in AAA it was clear he was a guy who would see time in the majors in some capacity, but the shaky discipline and not-exactly-airtight defensive reputation made him look like a probable bench guy. Which is fine. Might hit a couple clutch homers, get a funny nickname, and develop a cult following.

They brought him to MLB camp one spring, and yeah — he hit some dingers and caught some eyes. Nobody takes spring that seriously, but the ball jumped off his bat and that’s always fun. Coaches professed their impressed-ness. So more people learned his name.

Then, when faltering MLBers created an opportunity, he got promoted. And he ran with it. He hit a bunch of dingers. His defense actually looked pretty solid at that up the middle position. Yeah, he swung and missed a lot, but when he hit it he clobbered it, and he seemed to have a knack for finding the outfield grass even when it didn’t go over the wall. It tempted you to throw away what you know about the laws of BABIP and declare him your new god.

You know where this is going; everybody knows. These kinds of openings are frankly getting tired in baseball pieces, and I may have to start searching for another one when playing this cheap player-comp game. The player is Paul DeJong, but the player is also Randal Grichuk. Gotcha! (No, I didn’t.)

Randal Grichuk had red flags. Paul DeJong has the same ones. We can get that out of the way. However, even though real Grichuk wasn’t as good as pre-reality Grichuk, real Grichuk was a solid, two-winnish player. Because DeJong seems to eke more value out of a more valuable defensive position than Grichuk did, a real-Grichuk-ish (as a hitter) DeJong would be a slightly better player than Grichuk. Maybe a 2.5-3 win type. That’s a very useful player. Not a guy you’d kick out of the clubhouse for farting, you might say.

That could easily be the end of this story. DeJong could settle in as a guy who simply swings a lot, misses a lot, carries a low OBP but bops his homers, and makes a nice living as a #6 or 7 hitter who you’d really prefer not to see batting higher, but are fine with down there. Or, maybe, it won’t be the end of the story.

1:1 player comps like this are fundamentally lazy, because they omit the fact the fact that two different people may be very different in their capability to learn and adapt. Grichuk, somewhat famously (at least around here), never changed — never did learn to discriminate between things. DeJong, on the other hand, has already shown some capacity to adapt in positive ways, as Tyler Kinzy wrote a month ago. Cutting through all the nuts and bolts of Tyler’s analysis (for the purpose of brevity; it’s good analysis), this is the key finding:

DeJong was willing to take more called strikes as his swing percentage on pitches inside the zone decreased by 6.3% during hot streaks, but in doing so he began chasing fewer pitches outside the zone. This less aggressive, more selective approach at the plate resulted in higher contact rates and more favorable “hitter’s counts” as pitchers converted fewer first-pitch strikes. By swinging at fewer pitches, DeJong also became more efficient when the bat did leave his shoulders.

DeJong likes to swing — he likes to hunt pitches. But when he was at his best last year, the most obvious observable difference is that he was simply swinging a little less. Interestingly, he wasn’t just swinging a little less at balls. He was swinging less at everything, including strikes. So in addition to passing on some balls, he was passing on, presumably, some pitches he could drive. His discipline didn’t get better — not in the sense that we mean “ability to distinguish balls from strikes” when we say discipline — but his patience did. Overall, it resulted in (seemingly) better counts, better pitches to hit, and better quality on contact. And it appears to be a deliberate adaptation: both his swing rates both inside and outside the zone trended down as the year went on:

fangraphs.com

So, that’s a very interesting thing in its own right. It’s especially in light of this graph, taken from this Craig Edwards piece re DeJong:

fangraphs.com - Craig Edwards

As Craig put it:

No power-hitting player who swung outside the zone as often as DeJong last year saw nearly as many pitches in the strike zone. This next season, DeJong is likely to see fewer pitches to hit. How he responds will shape his season — and potentially future — as a hitter.

So this can go one of two ways, now (really more than two, but shut up). One, DeJong can continue or at least stabilize his trend of increased patience. With more pitches outside the zone than he saw last year, more patience will mean more good counts, likely corresponding to more walks or (because pitchers who fall behind have to come back into the zone eventually) more good pitches to hit. That would likely be a very good outcome. Or, DeJong can get tired of standing there, and go back to loving to swing the bat. That would lead to worse counts, fewer pitches to hit, and likely less quality contact. That would be bad. The former could be a very good player. The latter... “Grichuk but a shortstop” is a fine player, but there’s downside below that. It’s still possible for DeJong to swing his way out of relevance, probably.

What if we let ourselves dream on him a little, though? What’s the optimal Paul DeJong evolution look like? He likes to hunt pitches, and he’s good at it. Last year, he started (apparently) learning to not swing at so many balls by simply not swinging as much at anything. That was Paul DeJong’s first change. Maybe he can make another one. He’s got a good model playing on the team right now.

Marcell Ozuna used to look like he was cut from the Grichuk/DeJong mold: in his first full year (2014) he hit .269/.317/.455, with 27% strikeouts and 7% walks. His discipline profile from his debut through 2016 actually looked a lot like DeJong, too: 33% o-swing, 67% z-swing, 74% contact (compare to DeJong’s 2017: 34%, 72%, 74%).

Then Ozuna broke out in 2017. The only significant change he made in his breakout year was becoming more aggressive inside the strike zone. He didn’t start swinging more overall — his o-swing rate remained at his career level of 33% — but he had a notable jump on swings inside the zone: to 73% from a career rate of 67%. Ozuna started swinging more, but all his extra swings were at strikes. And it would be foolish to attribute his entire breakout to that change, but they did happen at the same time. It’s an example of a guy becoming (to crib a phrase from Cardinals hitting coaches in years past) selectively aggressive.

So, if we’re building a truly optimal DeJong, that’s probably what we’re looking for. Last year, as the year went on, DeJong learned to swing less. At everything. And the less he swung, the better he was. If he simply maintains that approach (no guarantee there), he’s probably a fine player.

But he’s already shown the ability to adapt — by swinging less. This year, pitchers will challenge him to maintain the adaptation. That’s step one. Step two, if he can get there, might be something like what Ozuna did last year: to keep spitting on the balls (taking balls is good, because it makes better counts), but stop spitting on so many strikes (taking strikes is bad, because it makes bad counts). Step two would be to, essentially, take the gains he’s realized by not swinging at balls, and take even more gains by hammering strikes.

I don’t know if DeJong can do that. It seems hard. MLB pitchers are good! But he’s already shown aptitude to learn in one way — by swinging less overall. Now, if he can continue to refine his approach by maintaining discipline against balls while regaining it against strikes, the Cardinals might really have something here.