Going Deep: 2019 HR/FB% Regression Candidates

Dan Richards utilizes Statcast to identify a few HR/FB% regression candidates.

I haven’t been shy in recent articles about my penchant for using Statcast data to evaluate hitters. Let me explain why:

Al Melchior did some great work for Rotographs in which he correlated Statcast metrics to HR/FB%, ISO, and Hard%, while also providing their year-to-year correlations for 2015-18. I’m most interested in the metrics that are repeatable and have strong correlations with HR/FB%. I like to focus on HR/FB% because it is a measure of how often a player muscles the ball out of the park when he puts it in the air. By determining which metrics influenced HR/FB% the most, Al found which ones best supported a player’s ability to hit his fly balls out of the park. Then it’s just a matter of picking from those the ones that are repeatable. If we know which Statcast metrics are both repeatable and strongly correlate to HR/FB%, then we can discern whose HR/FB rates were inflated by luck and whose were earned. Ever wonder why all of a sudden a hitter’s HR/FB% spiked and he hit 35 homers instead of 20 as he had in years prior?

What Al found was that Brls/BBE% shared a Pearson’s r coefficient of 0.85 with HR/FB% and had a 0.80 year-to-year correlation. Exit velocity on fly balls and line drives had a 0.79 correlation to HR/FB% and a 0.82 year-to-year correlation. Average fly ball distance had a 0.80 correlation to HR/FB% and a 0.65 year-to-year correlation. Those are all fairly strong. I should note that unlike exit velocity and barrels, you can’t just find average fly ball distance on Statcast’s leader boards; you have to search for it here. Anyway, these are the best available raw power metrics. How hard a player hits his fly balls and line drives, how far he hits his fly balls, and how frequently he barrels the ball are, not too shockingly, our guiding metrics.

By way of example, I’ll apply Brls/BBE%, exit velocity on FB/LD, and average FB distance to four hitters to show how I think their HR/FB rates will change in 2019. I also provide each hitter’s launch angle for context. For comparative purposes, here are the averages of those metrics for players with a minimum of 50 batted-ball events:


LA Avg FB Distance FB/LD Exit Velo Brls/BBE% HR/FB%
12.67° 319.53 feet 92.45 mph 7.16% 13.02%


Teoscar Hernandez (OF, Toronto Blue Jays)


Like many in the fantasy baseball community, I love to hate Teoscar Hernandez. That 31.2 strikeout rate? Ew. The guy was such a letdown after becoming a consensus sleeper candidate going into 2018. Now, after having been burned by him in two leagues, I hate to love him, but his numbers speak to me.


Player LA Avg FB Distance FB/LD Exit Velo Brls/BBE% HR/FB%
Teoscar Hernandez 17° 343 feet 97.7 mph 15.5% 15.9%


That is a surprisingly modest HR/FB% for such impressive power numbers. Compare those numbers with the averages above, and it’s obvious how well Hernandez should grade as a slugger. For context, of all players with 30 BBEs, Hernandez is:

  1. The only player to rank in the top 48 for Brls/BBE% who also had a HR/FB rate lower than 16%. Even more jarring is that he’s actually toward the top of the Brls/BBE% leaderboard at 11th overall. There were actually 101 players with a better HR/FB% and worse Brls/BBE% than Hernandez.
  2. The only player in the top 37 for average FB distance with a HR/FB% lower than 16%, and he’s 10th overall in that metric.
  3. The only player in the top 20 for FB/LD exit velocity to post a HR/FB rate lower than 16% and, you guessed it, he’s near the top of the leader board there, too, at 5th overall.

This all bespeaks an extreme outlier. A guy who hits fly balls extremely far, hits fly balls and line drives extremely hard, and barrels the ball extremely frequently but isn’t even in the top 100 for home runs per fly ball? Expect Hernandez’s HR/FB% to improve.

Maybe concrete examples will prove my point. One that stands out to me is Shohei Ohtani. Ohtani’s Statcast metrics predictive of HR/FB% are nearly identical to Hernandez’s. He had an average FB distance of 343 feet, a FB/LD exit velocity of 97.8 mph, and a 16.0 Brls/BBE%. But Ohtani’s HR/FB rate? 29.7%. Other similar players include Aaron Judge, Tyler Austin, and Daniel Palka, each of whom nearly doubled Hernandez’s HR/FB%. I’m not saying Hernandez will hit 40 homers with a 30 HR/FB%, but there should be improvement. Combine that with the hitter-friendly confines of the Rogers Centre and Hernandez’s fly ball and pull tendencies (17-degree average launch angle, 43.7 FB%, 37.7 Pull%) and you’ve got yourself a breakout candidate.

He probably won’t be the best player in fantasy because he won’t hit .300 or steal 30 bases, but go grab him at his NFBC ADP of 359.91 (since Jan. 1), and maybe he’ll swat 35 homers. His unluckiness has dug him into a playing-time hole, so he might make me look stupid again but he’s basically free at that price.


Jackie Bradley Jr. (OF, Boston Red Sox)


After an excellent article by Mike Petriello that uncovered Jackie Bradley Jr.’s 2018 surge in Hard Hit%, JBJ has become a popular sleeper candidate going into 2019 drafts. And with good reason; just look at how unlucky he got in 2018:


Player LA Avg FB Distance FB/LD Exit Velo Brls/BBE% HR/FB%
Jackie Bradley Jr. 12° 332 feet 96.3 mph 10.3% 10.6%


Despite the 58th-best average FB distance, the 25th-best FB/LD exit velocity, and the 70th-best Brls/BBE%, JBJ maintained a below average HR/FB%. To make matters worse, his Brls/BBE% would have been higher had he hit more fly balls. Encouragingly, JBJ increased his FB% in the second half of 2018 from 34.1% to 39.6% with a corresponding decrease in his GB% from 44.9% to 40.3%. How much sillier would that 10.6 HR/FB% look if his Brls/BBE% were even higher?

Jorge Soler is a reasonably similar player to JBJ, what with his average fly ball distance of 341 feet, 96.3 mph FB/LD exit velocity, and 10.3 Brls/BBE%. Despite playing in a park that’s ostensibly worse for power than Fenway, Soler still managed a 17.0 HR/FB%. Likewise, Bryce Harper had a 23.1 HR/FB% as a result of an average fly ball distance of 338 feet, FB/LD exit velocity of 96.3 mph, and an 11.5 Brls/BBE%. Mark Trumbo managed a 20.0 HR/FB% from an average fly ball distance of 339 feet, FB/LD exit velocity of FB/LD of 95.9 mph, and 10.2 Brls/BBE%. There are countless other examples. More of JBJ’s fly balls should have left the yard this past season, more will next season as his HR/FB% regresses based on his raw power, and even more will as he continues to refine his approach at the plate for greater elevation.


Omar Narvaez (C, Seattle Mariners)


Omar Narvaez first became fantasy relevant this past year when Welington Castillo was slapped with an 80-game PED suspension, opening up the catcher role for the White Sox. Owners took notice when the catcher hit .275 with reasonable counting stats. Narvaez actually hit nine homers in 322 plate appearances, which prorated out to a full-length season for a catcher equates to about 15 total. Batting .275 with 15 homers really isn’t that bad when you consider that Buster Posey hit .284 with five homers (and three steals) and is generally ranked in the top seven catchers for 2019.


Player LA Avg FB Distance FB/LD Exit Velo Brls/BBE% HR/FB%
Omar Narvaez 13° 308 feet 89.4 mph 2.8% 14.5%


Yet, Narvaez’s HR/FB% stands out like a sore thumb relative to his light-hitting peers. In 2018, Narvaez had the 43rd-worst Brls/BBE%, 68th-worst average FB distance, and 40th-worst FB/LD exit velocity. In fact, he ranks well below average in all of those metrics, yet maintains an above-average HR/FB rate.

Indeed, with a similar average launch angle to JBJ, Narvaez’s fly balls traveled, on average, about 7 mph slower, 24 feet shorter, and were barreled at less than a third of the rate of JBJ’s. However, Narvaez bested JBJ’s HR/FB rate by nearly four points. Consider Wilmer Flores as a more reasonable comparison, whose average fly ball distance, exit velocity on FB/LD, and Brls/BBE% were all similar to or better than Narvaez’s but had a 7% HR/FB rate. In fact, Narvaez is the only hitter in baseball with a Brls/BBE% lower than 4% and a HR/FB rate higher than 12%.

With the starting job locked down in Seattle, Narvaez may appear to be an enormous value. And admittedly, Narvaez is really being drafted as a second catcher rather than a starter given his NFBC ADP of 288.89 since Jan. 1, so even with HR/FB% regression, he may return value at that price. Just don’t expect 10 homers with those terrible Statcast metrics and his low 29% fly-ball rate, particularly as he moves to T-Mobile Park.


Jonathan Schoop (2B, Minnesota Twins)


Jonathan Schoop is a really interesting player. One season he hits .293 with 32 homers, the next he hits .233 with 21. Most people probably just assume the real Schoop is somewhere between his 2017 and 2018 selves, and he’s being drafted as such by going 179.79 in NFBC drafts since Jan. 1. In fact, he probably looks like a value there, what with a 32 home run ceiling. But I’m not so sold on a bounce back.


Player LA Avg FB Distance FB/LD Exit Velo Brls/BBE% HR/FB%
Jonathan Schoop 11.1° 313 feet 90 mph 5.0% 15.9%


I won’t recapitulate what I wrote for Narvaez. Suffice it to say, Schoop’s on the wrong side of the Statcast leaderboard for all of the same reasons as Narvaez. He’s below average across the board in Statcast raw power metrics. It’s actually kind of funny to compare Schoop and Hernandez, both of whom maintained a 15.9% HR/FB rate this past season:


Player LA Avg FB Distance FB/LD Exit Velo Brls/BBE% HR/FB%
Teoscar Hernandez 17° 343 feet 97.7 mph 15.5% 15.9%
Jonathan Schoop 11.1° 313 feet 90 mph 5.0% 15.9%


Of course, Hernandez’s HR/FB% should regress too, making this an unfair comparison, but it certainly paints the picture that Schoop overperformed on his fly balls. In fact, a more equitable comparison to teammate Hernan Perez is revealing too. Perez had a lower 10% HR/FB rate, but his fly balls traveled 320 feet on average, his FB/LD exit velocity was 91.2 mph, and his Brls/BBE rate of 4.9% was nearly identical to Schoop’s. They hit in the same ballpark, so you can’t say Schoop was just taking advantage of tiny Miller Park. Schoop was just 59% luckier than Perez.

True, Schoop experienced an oblique injury in 2018 that may have sapped some of his power. But his 92.7 mph exit velocity on FB/LD in 2017 never really supported that 32-homer power either. Moving from Miller Park to Target Field is as uninspiring as it gets, even if he regains a little strength after recovering in the offseason. If Schoop repeats his relatively low launch angle but with a more reasonable 10% or 12% HR/FB rate, is he even worth a draft pick?

Photo by Gerry Angus/Icon Sportswire

Dan Richards

Dan is a lifelong New York Yankees and Giants fan. A practicing attorney, Dan is better known for aggressively bothering his leaguemates about trades. You can follow him on Twitter @Fantasy_Esquire or by clicking the button above.

One response to “Going Deep: 2019 HR/FB% Regression Candidates”

  1. theKraken says:

    Schoop certainly suffered an oblique injury that sapped his power. He also found himself in a bench role after the trade to MIL. He missed games too… He is quickly becoming my guy for 2019 because people are only looking at the numbers, which is not necessarily insightful – in cases like this it is the opposite. Its easy to see why a reasonable rebound seems plausible as 2018 was just a nightmare in every way. I don’t want the other guys for non-statcast reasons haha. JBJ and Teoscar are swing hard just in case you hit it types, which current statcast-based metrics overvalue – when it is going good its good, but they are even more prone to slumps than someone with a reasonable approach.

Leave a Reply

Your email address will not be published. Required fields are marked *

Account / Login