J.J. Hardy’s Hidden Offensive Collapse
Last season was a banner year for the Baltimore Orioles. The team won 96 games in the regular season and advanced to the ALCS despite a whirlwind of adversity: significant injuries to Matt Wieters and Manny Machado, and huge power declines from Chris Davis and J.J. Hardy. Hardy’s 2014 was quite interesting. He was hampered by back issues throughout the season, hitting only nine homers after clubbing a total of 87 his first three years in Baltimore, but it ended well, as he quickly signed a three-year, $40 million extension after the season ended. The scary thing is, Hardy’s offensive performance last season was quite a bit worse than his .268/.309/.372 line indicated.
Hardy was the Brewers’ second-round draft pick in 2001 out of Sabino High School in Tucson, Ariz., and quickly advanced through the minor leagues. He was given a challenging assignment in his first full pro season in 2002, opening at High-A and receiving a promotion to Double-A by season’s end. Each year, I compile an ordered list of minor-league position player prospects based on their OBP and SLG, relative to their league and level, adjusted for age. Traditional scouting techniques are then used to tweak the order. Hardy’s minor-league performances resulted in three top 100 rankings, quite unusual for a shortstop. His peak ranking was No. 20 after the 2003 season, when he hit 12 homers as a 20-year-old in the pitcher-friendly Southern League. Such rankings and performances projected a long major league career, with some offensive upside.
Hardy indeed has had a long run as a regular MLB shortstop, arriving in 2005 with only a couple of injury interruptions since. After hitting all of 29 minor-league homers — with that single-season high of 12 — Hardy has gone on to hit 167 as a big leaguer, clearing the 20-homer mark five times. In many ways, Hardy has tracked the performance of Jimmy Rollins, who hit 36 minor league homers — also with a single-season high of 12 — before hitting 215 in the majors through his age-35 season. Rollins also was a second-round high school draftee, and earned a ranking on my minor league position player prospect list four times, including two top 100 slots, with a peak at No. 44 in 2000.
They’re very similar defenders, as well: solid but unspectacular range, great hands and extremely accurate throwing arms. Both like to swing the bat, causing their respective OBPs to suffer as a result; but they grew into some power, and have extremely respectable 97 (Rollins) and 96 (Hardy) career OPS+ marks for shortstops. The big difference between them, of course, is Rollins’ speed, which separates the two into different tiers as players.
Enough about Rollins. Let’s get back to J.J. It’s tempting to chalk up the entirety of Hardy’s 2014 offensive struggles to his back injury, which primarily affected him in the early stages of the season. After all, he didn’t hit his first homer until June 21, and then managed eight more the rest of the way. That’s way down from his 2013 total of 25, but one can easily make the case that a healthy Hardy would have hit 15 or so. Once you dig a little deeper into the numbers, however, other disturbing items emerge. To get to the bottom of Hardy’s 2014 campaign, let’s examine his plate appearance frequency and production by BIP type data. First, the frequency info:
| FREQ – 2014 | |||
|---|---|---|---|
| Hardy | % | REL | PCT |
| K | 18.3% | 98 | 50 |
| BB | 5.1% | 65 | 17 |
| POP | 8.1% | 101 | 52 |
| FLY | 27.0% | 94 | 42 |
| LD | 21.1% | 100 | 55 |
| GB | 43.7% | 104 | 60 |
Hardy’s 2014 strikeout rate was smack in the middle of the population of MLB regulars, with a percentile rank of 50. That doesn’t seem notable at first blush, but it’s way up from his first three years as an Oriole, when he posted K rate percentile ranks of 44, 30 and 11. His BB rate percentile rank is poor at 17, but that’s no surprise, as it has been floating in a narrow band between 17 and 23 since 2011.
His only BIP type frequency of any note is his liner rate. Though a 55-percentile rate might not catch your eye, it’s way up from the 21, 8 and 8 marks he posted in his first three years as an Oriole. While liner rates fluctuate more than those of other BIP types, some hitters do develop strong tendencies to post high or low marks. It would seem that Hardy’s inability to create line drive contact — he’s had liner rate percentile ranks below 10 in four of the past seven seasons — is part of his “true-talent” portfolio.
So we have a guy who doesn’t and won’t walk, whose K rate spiked considerably and who hit a bunch of liners that he doesn’t normally hit. Some orange, or even red flags there. The single most piece of damning information suggesting that Hardy over-performed his true talent level in 2014 sits in his production by BIP data, which gives us a better feel for his batted-ball authority. Let’s take a look:
| PROD – 2014 | |||||
|---|---|---|---|---|---|
| Hardy | AVG | OBP | SLG | REL PRD | ADJ PRD |
| FLY | .173 | .436 | 39 | 42 | |
| LD | .674 | .872 | 103 | 98 | |
| GB | .343 | .376 | 197 | 120 | |
| ALL BIP | .329 | .456 | 94 | 78 | |
| ALL PA | .265 | .303 | .368 | 92 | 77 |
Hardy’s actual production on each BIP type is indicated in the AVG and SLG columns, and it’s converted to run values and compared to MLB average in the REL PRD column. That figure then is adjusted for context, such as home park, luck, etc., in the ADJ PRD column. For the purposes of this exercise, SH and SF are included as outs and HBP are excluded from the OBP calculation.
Hardy’s fly ball contact was not very authoritative in 2014. He batted just .173 AVG with a .436 SLG in the air, good for an actual 39 REL PRD, adjusted upward minimally for context to a 42 ADJ PRD. Last week, I wrote about Austin Jackson, who also had a 39 REL PRD on fly balls in 2014, tied for 11th worst among American League hitters with 215 or more batted balls. Among those 12, Jackson was the outlier, the only one with above average fly ball velocity. Hardy’s average fly ball was over a half standard deviation below the 2014 AL average fly ball velocity. That’s down from 2013, when it sat squarely in the average range. This is one area in which the back injury would seem to have a direct impact.
As also indicated in the Jackson article, a small decrease in fly ball velocity can make a huge difference. When hitters hit a fly ball at a speed of 92.5 mph or higher in 2014, they batted .560 and slugged 1.884. When they hit a fly ball between 75 mph and 90 mph, they batted .077 and slugged .148. Hardy has always lived in close proximity to that fly ball cliff. Most of his homers are dead pull, and many are of the just-enough variety. Regardless of whether Hardy’s back is healed, he has already about tapped out his ability to selectively pull the ball in the air (1.93 fly ball pull ratio). On balance, one would expect Hardy’s power to be in retreat at this stage of his career.
The one area in which Hardy most clearly, grossly overperformed in 2014 was on ground balls. Somehow, someway, Hardy batted .343 AVG with a .376 SLG on grounders last season, good for an outlandish 197 REL PRD. That was third in the AL behind 2015 crash-landing candidate Danny Santana (210) and Mike Trout (203), both of who can fly. Hardy is righthanded and quite slow. Though he does hit his grounders fairly hard (120 ADJ PRD) and doesn’t have an excessive pulling problem (3.27 grounder pull ratio), the singular explanation for Hardy’s 2014 grounder performance is dumb luck.
Once all BIP are combined, Hardy’s actual 94 REL PRD plunges to a 78 ADJ PRD. Add back his K and BB, and those marks retreat subtly further, to 92 and 77, respectively. In other words, based on true talent, J.J. Hardy wasn’t a .268/.309/.372 hitter in 2014, but more like a .240/.279/.335 guy instead. Reliable, in-demand dependable defensive shortstop or not, does he get a three-year, $40 million deal if he puts up those numbers? Pretty good question.
Let’s go back to his back injury for a second. Let’s take that true-talent level and apply his 2013 fly ball impacting ability to it, and see what happens. Despite hitting 25 homers in 2013, he only had a fly ball ADJ PRD of 74; that reflects the high number of dead-pull, just-enough homers, as well as the multiple friendly dead left-field environs in the AL East. With a 74 fly ball ADJ PRD, Hardy is a .250/.288/.370 true-talent hitter. Let’s now dock him for the unusually high number of liners (for him) he hit last season; putting him down into the 10th liner percentile where he generally resides makes him a .240/.279/.360 hitter overall.
Camden Yards might make him a little better than that, but not much. The bottom line is Hardy’s offensive game is in decline, and the extent of that decline was obscured by the 13 or so extra ground ball hits he received last season just by randomly hitting them where fielders weren’t standing. He may be able to squeak out enough value to justify his contract by simply remaining healthy and making the trains run on time defensively, but those expecting material offensive value from Hardy moving forward are likely to be disappointed.
This is worrying, indeed. But Hardy also adapted his approach to account for his back injury, and was more conscious of hitting the ball away to right field. If his back is healthy, he can take the pitch on the outside half that is up in the zone and drive it, as he did with Scherzer in game 1 of the ALDS (home run to the Orioles’ bullpen). To me, this is why merely plugging in batted ball rates without context gives an incomplete picture, because the batter’s approach and the way he’s pitched is very fluid. If Hardy’s back is healthy, his approach will be different, and he will hit for more power. It is unlikely to be a truly productive hitter with his hacktastic ways, but if he can provide 15-20 homeruns and continue to play great defense then he is still a + player. I hope Buck rests him a bit more, and that we get a viable utility player instead of Flaherty (probably E. Cabrera) to make sure his back is healthy.
What do you mean by approach? If you mean what pitches a hitter swings at, and the trajectory / velocity / angle of those batted balls, I strongly disagree.
Batter swing rates stabilize faster than almost any other stat. If their approach were really super fluid, that wouldn’t be the case.
Batted ball component stats seems to fluctuate quite a bit. I suppose this could indicate an approach constantly in flux, but in the absence of any specific evidence that is true, I assume it’s because of random fluctuation. (I am willing to change my mind pretty quickly in the face of other evidence, but I’m not even sure what form that could take.)
Perhaps balls on the outside part of the plate that he had been driving to left in previous years he instead hit for line drives to right and up the middle, accounting for the fluctuations in the LD rate.
But I am basing this more on impressions from watching him play, in noticing his approach, rather than on batted ball numbers. My overall opinion is that JJ has always been a so-so offensive performer.
I think an injury provides the evidence needed.
A batter’s approach (or pretty much anything in life) changes when there’s a need. I know most players don’t do it as much now, but an 3-1 hack is different than an 0-2 or 1-2 ‘choke up and protect the plate’ swing. That’s essentially what Hardy was doing for the larger part of the year last season.
Seems that deviations from career norms including decline in contact rate, increase in Sw Str and K rate, combined with lowest HR/FB ever are pretty clear indicators that something has changed for the worse.
As someone who watched ~110 Orioles games last season, I can say this gibes pretty well with what I observed, and Hardy really did seem to be struggling at the plate most of the season.
That said, it’s one season and 569 PAs. I’m not sure I’d declare him inexorably in freefall yet, and would expect his 2015 numbers to land somewhere between 2014 and 2013 barring further injuries. And I think he’s in a better place to avoid those injuries since the addition of Everth Cabrera gives them an at least average defensive shortstop on the roster (which Ryan Flaherty was not in 2014) and will make it easier to give Hardy rest if he needs it.
Give the cost of a win in 2015, the Orioles’ other options at SS and the likelihood that their pitching staff is very reliant on having an elite infield defense (in a way that defensive metrics may not fully capture the value of) I’d say Hardy is a bargain at 4/40 barring a further collapse in his offensive numbers, even if you assume the last year of the contract is essentially dead money.
It’s 3/$40 million. Less of a bargain but fewer years. Given our lack of alternatives and JJ’s role in the clubhouse I think it was a good signing.
It was interesting that he signed the deal without testing the waters. Consensus was that he left a few million on the table by signing the deal, but perhaps if his back was an on-going concern he thought he would avoid the risk.
You’re right.
But as I said, not a huge difference between 3/40 and 4/40 considering he’s unlikely to be a very productive player in 2018. Your hope if you’re the Orioles is that the contract pays for itself in 2015 and 2016.
Everth has always rated as a slightly below average defensive s.s. Minor nitpick, but you might want to change the tint a little on your rose colored glasses.
Perhaps (although defensive numbers…), but he’s better than Flaherty, is what I am saying.
Considering the disasters the Orioles had a this position before Hardy, I’m ok with pretty much any contract. Cintron, Andino, Fahey, Luis Hernandez – These names fill me with dread.
Izturis, Castro, Deivi Cruz…
…Julio Lugo, Freddie Bynum…
I would like to see this exact same analysis excluding all data prior to 6/21. I recognize that this is a rather arbitrary way of saying “he was healthy from here on in,” but I also think it would provide additional proof either way as to whether the author’s primary point remained accurate regardless of health.
Do you have any thoughts about Yasiel Puig based on this kind of analysis? There has been a lot of talk about his BABIP which some justify by the fact that he seems to hit very hard ground balls. Was wondering if this theory would be backed up by your analysis.
‘They’re very similar defenders, as well: solid but unspectacular range’
Doesn’t RZR, OOZ, and RngR measure a defender’s range?
While playing through the back injury sited in the article, Hardy was still 4th in RZR, 6th in OOZ, and 2nd in RngR in 2014. @014 is not the only season where he rated him in many if not all of those areas (like leading the majors in OOZ in 2012). I don’t know, seems he has very good range.
They’re very similar defenders, as well: solid but unspectacular range’
Doesn’t RZR, OOZ, and RngR measure a defender’s range?
While playing through the back injury sited in the article, Hardy was still 4th in RZR, 6th in OOZ, and 2nd in RngR in 2014. 2014 is not the only season where he rated high in many if not all of those areas (like leading the majors in OOZ in 2012). I don’t know, seems he has very good range.
You really have to look at the 2nd half alone and exclude the 1st half numbers where it was known he was hampered by the back
His back flared up in the 2nd half as well, still got a cortisone shot after the All-Star break I think. It was a chronic problem for him in 2014. Which worries me, since backs can be nasty. However, when he was asked about his worst ever injury this January he mentioned a separated shoulder he got many years ago and not the back, so maybe he isn’t worried about the back? I hope so.
I think, though, that the numbers mentioned in the article suggest that his back isn’t his only problem. He’s living on a pretty razor-thin margin when it comes to his batted ball profile, where losing just a little bit of power hurts his overall productivity a whole lot. Unless, of course, as I suggest above, he compensated by hitting line drives up the middle and to right knowing he couldn’t drive the ball. Stats people say that line drive % isn’t controllable, so what I am seeing with my eyes in a change of approach they think is not real or sustainable. We’ll see this coming season.