Bryce Harper Is Zeroing In, Eliminating Few Remaining Holes
A frequently asked question this spring in FanGraphs chats, and presumably around water coolers inside and outside the Beltway, concerned which Bryce Harper we would see in 2017.
Would we see the 2015, Ted Williams-like, Griffey Jr.-in-his-prime, Hall-of-Fame-trajectory version? Or would we see something closer to the perplexing, if still productive, 2016 version. (Harper must have been restricted by nagging injuries last season, right?)
So far it seems like the answer is more likely the former, but perhaps it is neither. Instead of settling for somewhere between those outcomes, perhaps what Harper has really set out to do is to exceed the extremely high bar he set in 2015.
In a season of surprises, in a season of Aaron Judges and Ryan Zimmermans and Rockies and Twins, and talk about fly-ball and home-run trends, Harper has been the best hitter the game — and even better than his 2015 self with a quarter of the season having been played. That’s a surprise. I don’t think anyone believes Harper can sustain his .388/.503/.767 slash line or 220 wRC+, but then again this was player who produced a 10-win age-22 season.
What’s interesting to this author is that, at 24, Harper is presumably still learning the game, himself, and is apparently becoming a more skilled, more mature, and more selective hitter.
Consider where he most often took his swings in 2015:
Consider where he most often took his swings in 2016:
And consider where he is most often taking his swings in 2017:
Not unlike those heat maps of Miguel Sano, Eric Thames, those data-density charts have the look of a hitter narrowing his focus, becoming more selective, and perhaps zero-ing in on the regions where he can do the most damage. And that region is down in the zone.
And there is something else, too. Harper is also perhaps closing one of the few remaining holes in his swing. Even during his MVP season in 2015, Harper had some weaker areas in the zone — including a curious one for a left-handed batter: the down-and-in location versus all pitches. Consider his slugging percentage per swing by location in 2015:
And thus far this season:
If you examine those charts, you’ll see Harper has become more effective at hitting for power on the inside part of the plate — and particularly down and in — this season.
Consider his exit velocity on pitch locations in the lower- to middle-inner half of the zone and in off the plate, according to BaseballSavant’s detailed zone search:
2015: 85.8 mph mph (58 balls in play)
2016: 88.0 mph (52 balls in play)
2017: 91.8 mph (22 balls in play)
If you examine those charts, and those exit velocities, you will see Harper has become more effective at hitting for power on the inner third, and in and off the plate. This is interesting since Harper has always had that awkward, front-foot-in-the-bucket movement on certain swings on inside pitches. He still does it, and while it’s not textbook, it certainly does not sap him of power. There are even those who endorse the practice:
Consider his second home run of the season and the location of the pitch:
Consider the pitch location of the laser of a home run Harper hit Tuesday night in Pittsburgh:
He’s not only filling a void, but he’s better hammering mistakes. He’s doing roughly twice as much damage on pitches in the lower-middle of the strike zone compared to 2015. Consider this pitch in the ninth inning Saturday that was left out over the middle of the plate hours after it was announced Harper had agreed to a $21-million, one-year deal for 2018, his final year of club control:
The trade-off could be that he’s not covering the outside of the strike zone quite as well, but if it’s a trade-off for better damaging mistakes middle and in, then the Nationals will likely accept it.
I asked Harper on Thursday about any changes in approach he has made. And if he had made them, he didn’t divulge.
“I don’t look at heat maps. I don’t look at being patient or not being patient, or what the guy is going to throw me, or not going to throw me,” Harper said. “I don’t look at any video. I just go out there and try to play and let it all come together the best I can.”
Does he feel he’s improved on the inner third?
“I have no idea,” Harper said. “I am just trying to go out there and see the ball, hit the ball.”
Maybe Harper is the Wee Willie Keeler of the modern age. Maybe Harper really does engage in little analysis. He was the anti-Thames in terms of introspection Wednesday afternoon.
But maybe Harper isn’t even aware of gains he’s making. Perhaps he’s unwilling to consciously detail any growth or adjustments, or perhaps it is a subconscious development.
What’s interesting is that much was made of the Chicago Cubs decision to pitch around Harper during a series last May, before Harper’s struggle began. There was a theory, a narrative formed, that it got Harper off his game. There was an idea that he began to expand his zone. (Of course, Harper’s swing-location frequency last season was similar in 2016 and 2015). Who knows. But Nationals manager Dusty Baker said something at the time that’s perhaps applicable to Harper’s success now.
Baker told reporters and Fox Sports something of interest last May:
“Before they stopped pitching to Bryce, about the last couple weeks in April, he was kind of struggling a little bit. You know what I mean?. So they’re actually doing Bryce a favor. The more pitches he sees he can zero in on what is good and what’s not.”
Another theory: maybe Harper learned to zero in last season, but he just needed full health to take advantage of the lesson, of the experience.
By refining his focus, he’s become more efficient when he swings. It’s similar to the Barry Bonds theory of hitting which Thames adopted in South Korea. It’s not unlike what Sano is doing in Minnesota, or to a lesser degree in Cleveland, where Francisco Lindor has increased his power.
Harper is zeroing in. What that means for a player who already has a 10-win season and MVP award on his resume is quite terrifying for opponents and a spectacle to enjoyr for the rest of us. We don’t where Harper will go from here in regard to performance, but there seems to be a good chance that we haven’t seen the best of Bryce.
A Cleveland native, FanGraphs writer Travis Sawchik is the author of the New York Times bestselling book, Big Data Baseball. He also contributes to The Athletic Cleveland, and has written for the Pittsburgh Tribune-Review, among other outlets. Follow him on Twitter @Travis_Sawchik.
I like the article, and as a Nats’s fan it annoys me that Harper (and Zim as well) don’t seem to want to care at all about advanced metrics. Harper’s obviously quite good and works hard, but the lack of understanding just rubs me the wrong way.
Meh. I wouldn’t put a whole lot of stock in what pro athletes give to interviewers. Harper speaks in cliche as much as anyone out there. It doesn’t mean he isn’t internalizing things behind the scenes.
I hear what you are saying, since this is fangraphs after all, but the actual process of hitting needs to be as simple as “see the ball, hit the ball.” Clearing the mind of the hitter is key. Let the down votes commence. (https://www.amazon.com/Mental-Keys-Hitting-Performance-Enhancement/dp/1888698381)
Eh. Whether that stuff is helpful or not really just depends on the player. He might be a guy who needs his mind totally blank going into a PA.
Harper recently talked about admiring how Jeter worked with the media (by never saying much). If anything, he has studied media relations with a sabremetric level granularity.
The player doesn’t need to know all the things the general manager knows.
Also, just one tiny issue: On what planet were Ted Williams and Ken Griffey Jr. similar hitters as you imply in the second paragraph? Griffey’s career best 164 wrc+ pales in comparison to Ted’s career *average* of 188.
I thought he was just trying to zero in, given that at this point in Harper’s (and everyone else’s) career Ted Williams seems like a bit of a reach
On offense? Yes. But in total WAR you could see Trout catching him. Potentially.
Not worth mentioning the .398 babip or the 32.5% hr/fb%?
“I don’t think anyone believes Harper can sustain his .388/.503/.767 slash line or 220 wRC+”
Could be more specific about why to show the degree to which he’s likely to regress.
Do you need everything spoon-fed?
Okay, I’ll spoon-feed you. His xBABIP for this year based on Mike Podhorzer’s latest formula, is .331. That means that given the expected level of luck, his batting average should be .316 (you can do the math yourself). That would be 22nd in the majors rather than 2nd. Think that degree of regression isn’t worth mentioning?
Only if you also perform the same xBABIP regression for everyone else on the BA leaderboard, then tell us where Harper ranks.
And under also. Presumably he’d be around 22nd, since the average xBABIP is going to be the same as the average BABIP. That’s pretty good, but not close to ‘the best hitter in the game’, as claimed in this article.
So his wRC+ would only be in the 180 range. What a terrible player!
The statement which I am questioning is that he’s been ‘the best hitter in the game’, which was made in this article.
It took you 4 comments in this thread to get to that point. The previous 3 comments do not in any way imply that it was that statement you were questioning.
Do you need to be spoon-fed? You seriously didn’t understand that this was implied when I wrote ‘Not worth mentioning the .398 babip or the 32.5% hr/fb%?’?
Does any of that contradict – or have anything to do with – the main idea of the article?
Do you even know what the main idea of the article is?
Did you take basic reading comprehension in middle school?
Being specific when what you’re responding to is not the main idea of the article isn’t called spoon-feeding, it’s called basic logical process.
The main point of the articles seems to be that harper is ‘filling the holes’ and has been ‘the best hitter in the game’. What i’ve written suggests that in fact he hasn’t improved but is simply a good hitter who sometimes gets lucky for a while and sometimes gets unlucky so i would say it does contradict the main point. Understand now?
Well, you have that in common with Harper; both of you are on a tear. He is on a tear crushing baseballs, while you are on a tear engendering downvotes.
Are you just getting lucky for a while or is this your normal level of performance?
This is pretty normal for me. I like to point out when articles are statistically naive, and readers here don’t like it because they think they understand statistics.
We like to point out when comments by certain users are consistently devoid of logical thought and reading comprehension, and JFTG doesn’t like it because he thinks he has normal human intelligence.
No, the main point is that Harper has eliminated some of his weaknesses, and that he’s getting better. That he also happened to have been one of the most effective hitters this season is merely an incidental outcome. And you just declared that you have less reading comprehension and logical thinking ability than a 5 year-old.
Would absolutely love to see Harper and Trout duke it out over the next 10 years. I know we have talked about this on numerous occasions before, and while Trout is still the better player, a healthy Harper could potentially make a run for the top spot.
And FWIW, since 1950, a 200+ wRC has been attained only 11 times. Bonds (4), Williams (2), Mantle (2), Bagwell (1), F. Thomas (1), McGuire (1). Mantle and Williams had a 200+ wRC in the same season (1957).
Currently, Harper, Trout, Judge, Zimmerman, and Freeman are at the 200+ plateau. There is a great chance none of them make it to that point. With that being said, if anyone can do it, Trout and Harper could make it happen. Looking forward to the rest of the season!
Harper is just not in the same class as Trout. Be serious.
And perhaps only 5 (Mantle 2, Williams 2, Thomas 1) without the benefit of PEDs.
I think Harper will end up being the better offensive player. Maybe not for his career, as Trout has a head start, but I’d venture his 5 best seasons are better than Trouts 5 best by WRC+
Well, Trout seems to put up a wRC+ around 170 and 9 WAR every year. I could see Harper having another 4 years like his insane 2015, but then another 5 years where he “only” has a wRC+ of about 150.
Sometimes the lack of perspective just cracks me up. After 2015 there was a spate of articles on how Harper had surpassed Trout. After 2016, most of those scribes wrote, mea culpa, what was I thinking? — one year is not enough proof!
So here we are in 2017, with Harper off to another monster start, and the drumbeat starts anew — yet Trout leads in wRC+, as usual. Wouldn’t it be smart to let Harper post at least 2 full great years before considering him in the same breath as Trout?
Speaking on offense alone, I would venture his peak is higher. Not a better player, just a higher peak offensively. Thus far his peak season is 12% higher (23Wrc+) and I would be shocked if Harper does not have a few more Ruthian seasons. With that said, who knows on Trout. He very well could still be improving. My only point was Trout might finish his career with a higher Wrc+ and almost certainly higher WAR, but I would venture Harper will have a better top 5 offensive seasons. One of those two players will assuredly have a 200+ season.
One year isn’t what most people mean by a peak. If you think Harper’s 5 best wRC+ years (not necessarily consecutive) will be higher than Trout’s, I won’t dispute; I don’t think that matters much. But if you think Harper’s best 5-year stretch of wRC+ will be better than Trout’s, I’ll just smile and wait. Only a handful have had 5-year runs better than what Trout’s already done.
BTW, the kindest comparison of Harper to Trout — 2015 to now — finds Trout ahead 176-161 in wRC+.
This is what i’ve been saying, john, and you see the reaction. It’s like these people only read articles about harper when his babip goes up and so his stats go up. Then they skulk away. The interesting question is many people react this way with some people but not others. Obviously writers like sawchik are leading the way by jumping the gun and not bothering to check babip and hr/fb?, but why harper? this happens with some players but not others.
Meanwhile, the Blue Jays are eliminating his competition for the NL MVP. Goddamn, I hate Toronto so much right now.
Interesting that on a day when Fangraphs publishes an article called ‘Mariners Prospect Gareth Morgan Is Riding BABIP to a Resurgence’, it also publishes an article on Bryce Harper which doesn’t mention that he has a BABIP of .398 as opposed to an xBABIP of .331 (according to Mike Podhorzer’s latest formula), which would make his expected batting average .316 (22nd in the majors) as opposed to the actual .388.
And .316 is still an amazing expected batting average. I would bet everyone else in the majors currently hitting over .316 is also outperforming their xBABIP. What happens if you regress everyone else to their xBABIP and then compute what his MLB rank would be?
JFTG do something that doesn’t further make things fit his narrative? HA.
The narrative I’m giving was generally accepted by Fangraphs readers a month and a half ago before Harper’s recent run of .398 BABIP: I’ll refresh your memory. Take a look at what people wrote:
http://www.fangraphs.com/blogs/bryce-harper-or-manny-machado/
Remember back that far?
Obviously if you calculate everyone’s xBABIP (not just those currently hitting over .316 but those hitting under .316 as well), the average will be the same as the average BABIP. So since Harper’s xBABIP would give him the 22nd best batting average the majors, the normal expectation would be that his expected batting average would also be around the 22nd best. There are more or less 255 everyday players in the majors (8 on every NL team, 9 on every AL team), so that would mean Harper is better than about 91% of them. That’s pretty good but I wouldn’t say it’s amazing.
You have to go down 9 spots in the AVG leaderboard to find someone with a lower BABIP, and those 2 are Mike Trout and Freddie Freeman. So have you applied the same xBABIP regression to the rest of the AVG’s and THEN he ends up 22nd?
You would have to do that for EVERYONE’S xBABIP, not just the one’s on the list. And because the average xBABIP and the average BABIP are by definition going to be the same, of course he would end up around 22nd.
That’s not how it works. Without adjusting for xBABIP, Harper is first in the majors in average (or real close, I didn’t look but .388 is damn high). So after adjusting for xBABIP, we should also expect him to be first in the majors. You adjusted him first and then assumed he would keep that position after adjusting everyone else.
You obviously don’t understand how xBABIP is calculated. Here’s Podhorzer’s most recent formula:
.1770 — .3085*(True IFFB%) — .1285*(True FB%) + .3684*LD% + .0798*Oppo% + .0045*Spd + .2287*Hard% + Year Constant
There is no reason to assume that ranking in BABIP should correspond exactly to ranking in xBABIP for particular players. That’s the whole point in having a formula based on several variables.
For example, among the 14 Giants with the most plate appearances in 2016, Conor Gillaspie was #12 in BABIP but tied for #4 in xBABIP, and Eduardo Nunez was #14 (worst) in BABIP but #10 in xBABIP (they were unlucky), while Hunter Pence was #3 in BABIP but tied for #12 in xBABIP, and Travis Brown was #8 in xBABIP but #14 (worst) in xBABIP (they were lucky).
Understand now?
I understood before too, but thanks. If you calculate everyone’s new average using xBABIP instead of BABIP and Harper is 22nd I will happily eat my words.
Couldn’t find the year coefficient so I just calculated with the formula you gave minus that term. Harper is essentially tied for 5th with Posey and Murphy.
Did you calculate everyone in all of baseball? There are going to be quite a few people who aren’t on of this list who are getting unlucky so that their xbabips are higher than their babips. I agree that harper may be the 5th best hitter on this list so far this year, but that doesn’t mean he’s the 5th best in baseball.
Everyone (187 players) on the qualified leaderboard. I don’t care if someone with only 30 PA has a .400 plus xBABIP.
There is no published year coefficient yet for this year. I used the one for last year. It doesn’t matter if you left it out since this is relative ranking anyway and everyone gets the same one.
It would be nice if something with a full-time job doing these things, for example the writer of this article, would do these things. I don’t have time. When your batting average is 72 points higher than your expected batting average, there shouldn’t be too much doubt that your going to rank way higher in the first than the second. You’re either completely ignorant or clutching at straws.
“Don’t have time” is such a convenient excuse when somebody takes your “argument”, actually bothers to do the research, and whacks you in the face with it.
You’ve mentioned it twice in comments on this article. Do you feel better now?
Someone in a subthread to my original post challenged me to provide statistics, so I made another post with the statistics.
He whacks 13 home runs in 163 PAs. He draws walks on over 18% of his at-bats, higher than his K’s. His OBP is at .450. Saying that he’s overperforming his xBABIP is missing the forest for one really odd tree.
This amazing start by Harper puts Mike Trout’s greatness into even more perspective. Harper has gotten out of the gates with an unreal HR/FB% and a 398 BABIP – as is listed in the article. He’s been remarkable, and his 213 wRC+ indicates that. What’s truly remarkable though is Mike Trout has equaled him despite Trout’s numbers being slightly more sustainable. 352 BABIP, 33% HR/FB rate, and Trout is actually running a better wRC+ at 217. Even when Harper is at his absolute best, he’s equal to Trout. The debate on who one would rather have going forward has always been a non contest to me. Trout is durable, and as consistent as it gets.
Even if Harper’s down year was due to injury, he gets injured a lot so that isn’t something you can throw out. While Harper has been dominating the game, Mike Trout has just been Mike Trout… the same guy he’s been for 5 years. It’s fun to watch Harper, and it’s a treat to watch these two at the same time, but I do think there’s a good bit of separation between the two… and I think there’s a chance we’re watching the greatest player to ever live in Mike Trout which is not something that gets thrown around lightly.
Harper walks more and strikes out less (in 2015/2017 mode), so perhaps he still has a slightly higher offensive ceiling than Trout? Otherwise, yeah, Trout is pretty much to GOAT waiting to happen.
I kind of feel like Harper’s best seasons will be better than Trout’s, and his worst seasons will also be worse than Trout’s.
That feels hard to believe given that Harper’s season two years ago didn’t top Trout’s WAR’s in his rookie or sophomore campaign. It’s possible, but I wouldn’t bet on it; Trout seems to be improving his game, and his game was already better than peak Harper (so far) to begin with.
I should clarify–I meant on offense. I don’t realistically expect Harper to top 10.5 WAR in five seasons.
But going forward and factoring defense in (which admittedly changes the goalposts in a couple of ways), I still don’t think my statement is that crazy. Trout’s defense seems to have regressed, or he’s just a less-than-stellar defender in center.
Also, don’t get me wrong, if I get to pick only one player in the game to start my team I still pick Trout. I think he’s probably a lock for a wRC+ around 170 and 9 WAR every year. His track record is incredibly consistent. I can’t think of a better guy to have on my team.
sadtrombone — Why do you expect that? I am sincerely curious. I mean, yes, we’ve seen Harper have a season wRC+ better than Trout’s best (so far), and obviously several seasons worse than Trout’s metronomic norm; and we’ve seen Harper get hurt more often. Are there other reasons to expect Harper’s output to be more mercurial?
You’ve identified most of the key points. The variation on Harper’s offensive output is pretty high, for Trout it’s extremely consistent. Nothing more complicated than looking at their histories and thinking that Bryce Harper can have a hot streak that lasts the whole year and be only above-average the next, while Trout does the same thing every year.
This whole idea becomes moot if Trout really has taken the next step and keeps slashing something like .340/.450/.740. Or that Harper’s last year was due to an injury that is unlikely to recur. I might be too slow to update my priors, which is that Harper can go off like Ted Williams for a year and slump the next but Trout is year-in-year-out consistent. Either way, I think it’s pretty clear that the “what sort of career will Harper or Trout have” question is pretty much impossible to answer since this level of performance is so rare.
I get all the Trout-love, he’s amazing and clearly the best player in the game. But until he puts up multiple seasons over 1.000 OPS you cannot plausibly call him the GOAT. For comparison Mantle had 2 years over 1.000 OPS in his first 6 years (and 3 in his first 7).
Mickie Mantle had the short RF fence of old Yankee Stadium, though, while Mike Trout plays in one of the toughest parks for hitters.
Mike Trout has stolen 2 fewer bases in his career than Mantle did. There are other aspects of the game that Trout is superior to Mantle in; not only that, but I think we’re talking semantics here in the difference between a .991 OPS and a 1000 OPS.
You can’t call any 25-year-old the greatest of all time. But your standard for Mantle’s superiority is conveniently arbitrary, since Trout’s had 2 years of .991 OPS.
Mantle had an .009 edge in OPS through 6 years. That’s not even enough to settle that one comparison, let alone a total value question.
I think you’re right that we might be seeing a legitimate Top 10 player, and a guy who could make a run as GOAT. But realistically, I don’t think we’re ever going to top Ruth’s peak and total career contributions. A quick thought experiment…
If Trout has 9 WAR seasons every year (including this one) for the next 7 seasons (which assumes no injury or down years) then he’s going to wind up with 113.5 WAR. So under the best case scenario he has a career WAR like Mantle and a little above Frank Robinson. Then maybe he puts up 5 WAR seasons for the next 5 years after that, that puts him at 129.5. So under these assumptions he winds up with a career kind of like Ted Williams, except Ted Williams lost prime years to WWII. Since I think Ted Williams is easily a Top 10 player I can definitely imagine a situation in which he gets to Top 10 status.
And yet, he’s still 50 (!) WAR short of Babe Ruth, not to mention that Ruth has the top four (!) single seasons in fWAR all time, plus #6 and #10 for good measure. Trout is very likely the only active player who could possibly reach those heights but I think his realistic ceiling is the Ted Williams/Hank Aaron/Willie Mays/Ty Cobb tier of greatness. And things still have to go better for him than the optimistic scenario I sketched out to even reach Willie Mays. Which is still stupidly good. Maybe he could be the greatest player since integration, if everything breaks right? Mays still has 149 fWAR…
Could Trout catch Ruth or Mays? Just for fun…
Since 1901, there are 39 hitters who:
(a) totaled 20+ bWAR age 20-24; and
(b) have completed at least age 34.
For age 25-34, that group nearly doubled its total WAR from the prior period, from an average of 28 to an average of 54.
But it’s not fair to compare raw totals by age brackets, since they didn’t all play full-time across age 20-24; a few debuted at 21. So for an adjusted 20-24 average, I projected their WAR/game to an expectation of 725 games in that span (5 years times 145 games/year).
Now their age 25-34 totaled 1.73 times the adjusted 20-24 total.
(And that’s with no adjustment for Wartime service that took prime years in the 25-34 bracket away from DiMaggio (age 28-30) and Williams (25-26), nor for Arky Vaughan sitting out age 32-34 in a feud with Durocher.)
Applying that multiplier to Trout would give him about 130 bWAR through age 34 — more than Ruth (123), Mays (120), or anyone else.
Now, a 39-man sample means little. I wouldn’t dare to *project* 83 WAR for Trout’s 25-34, or for anyone’s 10-year stretch. And any attempt to project his latter 30s, from here, would be just plain silly. At the same time, I think Trout’s consistency can make us forget that most of the biggest stars at age 20-24 still had their peak years ahead of them.
It would be good to add Harper’s 2016 SLG/P chart. The trend you note for *this* year vs. 2015 was already evident last year — improvement on down-and-in, decline on the outer third. Since the outer third was discussed as a big problem last year, I find it interesting that he’s hitting so much better this year without much improvement in that zone.