Keston Hiura Versus the Regression Monster

Keston Hiura hit 38 home runs last year. There are qualifications to that statement, sure — 19 of those home runs came in the homer-happy PCL, and the majors weren’t much better when it came to mass dingerization. But still — Keston Hiura, who hit 13 home runs in 535 plate appearances in 2018, hit 38 home runs in 2019. What did he do to become such a great power hitter, and should we expect to see it again in 2020?

The first place you’d generally look, when considering an outlandish result like this, is for something wildly unsustainable. Maybe he turned half of his fly balls into home runs, and we can just point at that and move on. Indeed, Hiura’s HR/FB% was a juicy 24.1% in the majors last season, and an even more preposterous 36.5% in Triple-A.

Let’s throw out the Triple-A numbers for now. The combination of a new stadium in San Antonio and a wildly changed offensive environment makes putting those home runs into context difficult, so we’ll simply focus on the major league numbers. Non-pitchers hit home runs on 15.4% of their fly balls in 2019. Lower Hiura’s 24.1% to average, and he would have lost out on seven home runs. Easy peasy, let’s get lunch… right?

Well, yeah, not so much. I prefer to look at a different denominator: balls hit with between 15 and 45 degrees of launch angle rather than “fly balls.” That adds some line drives, which are potential home runs, and removes balls hit at too high of an angle to get out. Hiura had 83 of those in 2019, and turned 22.9% of them into homers. The league turned roughly 15% of theirs into dingers. Still the same seven home runs.

But batters aren’t all average. They have control over their home run rates, far more so than pitchers. Regress Hiura’s results in 2019 back towards the mean, and they suggest a true talent home run rate around 20.5%. That would still give him 17 home runs in the majors, not too much worse than his actual production.

Of course, we have more data than just how many balls Hiura put in the air. We know how hard he hit each of those, and we can use that to refine our expectations. Was he Eric Sogard out there, cranking home runs that just barely squeaked out? Hardly. Plug in Hiura’s relevant batted ball data, and his air balls produced an expected 22 homers in 2019. Maybe we’re underestimating his power!

All told, Hiura’s expected home run rate in 2019 was in the top 20 in all of baseball. He was right next to Bryce Harper when it comes to home run rate, and he put the ball in the air roughly equally frequently when he made contact. So should we just pencil him in for 40 bombs and move on with our lives?

Not exactly. After all, we know other things about Hiura, like the fact that he hit 13 home runs on 205 air balls in 2018 in roughly average hitting environments. That’s why projections pencil him in somewhere a little below our regressed estimate; ZiPS and Steamer both see him with 28 or so home runs in 2020. If you instead used our regressed estimate and the projection systems’ estimates for balls in play, that would work out to 30. If you added his exit velocity data to the mix, it’d be more like 36. In other words, the systems mostly believe his major league results, with a dash of skepticism thrown in due to past performance — but they’re skeptical that he’ll continue to hit the ball as hard as he did this year.

They mostly believe his major league results when it comes to home runs, at least. As great as the power was, a huge part of Hiura’s value last year came from hits that didn’t land past the fence. He batted .333 on grounders last year, nearly 100 points higher than league average. He batted .737 on line drives that weren’t homers, again 100 points over league average. And he even batted .151 on flies and pop ups that stayed in play, as compared to .098 for the league as a whole. That’s how you reconcile his ghastly strikeout rate (30.7%) with his enviable batting average (.303).

Can he keep that up in 2020? I mean, no. He had a .402 BABIP in 2019, which should tell you all you need to know. That’s not to say that none of it is earned, however; he has some characteristics that lend themselves to a high batting average on grounders. He’s in the bottom 10% of the league when it comes to pull rate on grounders, hitting far more of them up the middle. He’s in the top 15% in terms of hard-hit rate on the ground, another good sign. He’s not particularly fast, but he’s not particularly slow either. It seems reasonable to expect another year of above-average BABIP, though not at the previous level.

More important than either BABIP or home run rate is the synthesis of the two — production on contact. Hiura has all the hallmarks of a power hitter. He produced barrels, the most valuable batted ball type, on 13.9% of his batted balls in 2019, a top 20 rate in baseball. Barrel rate is the best predictor of power on contact, and he’s got it. That shows up in the projections — Depth Charts expects a .439 wOBA on contact in 2020, which would be far lower than his 2019 output and also far higher than league average.

This analysis is all well and good, but it boils down to the fact that our projection systems are pretty good. Want to know how well Hiura will do this year? Look at how he’s done in the past, fold in aging and how similar players have evolved, and go from there. ZiPS and Steamer are great!

While that’s true, it leaves out just how rare an improvement like Hiura’s is. To come up with a cohort, I made a few broad characterizations. I looked for players who had produced a wOBA on contact below .400 in the minors in one year before checking in at .440 or higher in the majors the following season. Hiura had less power than that in 2018, and more in 2019, but we’re painting with a broad brush here. I further limited it to seasons of at least 300 PA in both the minors and the majors. It’s a short list:

It might look, at first, like a disappointing list. Every one of those players did worse on contact in the subsequent year. But it’s almost a lock that Hiura will do worse on contact in 2020 — he just had the 16th-best wOBACON of any player with at least 300 plate appearances since 2007 (the first year of our minor league statistics is 2006, which is why I started the major league ones in 2007). Unless you think he’s one of the best power hitters of all time, 2020 will likely be worse.

If you accept that Hiura won’t be quite as good in 2020 as he was in 2019, then the list starts to look more interesting. Every single one of these hitters was above average in terms of production on contact over their careers. There were no flukes, no slap hitters who managed to spike one great year of contact. Even if you widen the net by accepting lower power numbers in the majors or higher power numbers in the minors, everyone who shows up has at least average pop.

In other words, you shouldn’t worry that the power will fade away, leaving only the poor strikeout and walk totals. It could happen, of course, but there’s no precedent for it. Instead, you can dream on what can happen if the power remains and the strikeouts come down. Hiura struck out a lot in 2019, and he swung and missed a lot as well. He had the ninth-highest swinging strike rate in the majors — he didn’t get all those strikeouts by bad luck alone.

But there were encouraging signs. His first strike rate, which is generally an excellent predictor of strikeouts, was bad but far less extreme. He didn’t have a disastrous out-of-zone swing rate; he whiffed quite often on pitches outside the zone when he swung, but that’s not even a clearly bad thing, because there’s little gain to be found in putting those balls into play. And despite being swing-happy overall, he wasn’t reckless; he swung at a league-average number of pitches in the “chase” zone, with most of his swings at balls coming in the shadow area just off the edge of the plate.

That’s not to say it’s all roses; Hiura might have a good eye, but he’s always had a lot of swing-and-miss in his game. Even in 2018, when he wasn’t hitting for power, he made contact on only 75% of his swings, and that fell to 70% in Triple-A in 2019. A swing with so much violence and power, combined with his aggressive approach at the plate, suggests that strikeouts are always going to be a part of the profile.

But if he can lower the strikeouts even a little, it will give his game a much higher floor. And he’ll likely draw his fair share of walks; he’s not a particularly patient hitter, but pitchers will surely start avoiding him more than they did in 2019, when they threw in the strike zone a roughly average amount despite his prodigious power.

Put it all together, and Hiura can be worse than his 2019 in a lot of ways while still being an excellent hitter. He’s got power to spare; knock 100 points off of his ISO, BABIP, or wOBACON, and he’d still have reasonable numbers there. He doesn’t strike out enough so that it derails his game, and he doesn’t show the warning signs of someone who might not make enough contact to be playable. And it’s not as though these skills came out of nowhere in 2019, even if they weren’t there in 2018; his college career was basically an extended montage of doubles off the base of the wall with home runs sprinkled in for variety.

Keston Hiura will almost certainly show less power in 2020 than he did in 2019. The regression monster is indefatigable, the pull towards the mean tremendously strong. But he doesn’t need to beat it, merely fend it off. He can improve in other skills to offset it, and he produced at an All-Star level last year despite ghastly defensive numbers. And hey, maybe he’ll beat back the regression monster for a while after all. He does have the requisite heavy artillery: he just put one over the batter’s eye:

Keston OVER the batter's eye. ? pic.twitter.com/b9prSCVCxK

— MLB (@MLB) March 4, 2020