Red-hot Harper looking like a superstar again

Quick: Think about what you know about Bryce Harper's 2018 so far. You know that his batting average has been low, dropping down to .211 at one point in July, because it's all anyone has talked about all year. You know that Harper put on a show at the Home Run Derby in front of his home fans, but you also know his free-agent-platform year has hardly gone as planned for the disappointing Nationals, to the point that it was actually a serious question as to whether they'd trade him before last week's Deadline.
They didn't, of course, and now there's something new you need to know. Harper looks like himself again. He's been crushing the ball. Since the All-Star Game, Harper has been one of the dozen best hitters in baseball, thanks to a massive .359/.480/.692 line.
It's tempting, really tempting, to refute the myth of the nonexistent "Home Run Derby curse" by saying that blasting baseballs into the humid Washington night on July 16 somehow unlocked the inner talent Harper has always had. It's also not really true, because this didn't start then. It started weeks before.
Let's consider the bottoming-out point of Harper's year to have come on June 20, when he went 0-for-4 in a 3-0 loss to the Orioles. His season line after that game sat at .209/.347/.462. That's an .810 OPS, which to be fair, is still good -- the 2018 Major League OPS is .729 -- but it's hardly what you expect from Harper.
The next day, Harper doubled, tripled and walked twice. In 23 games between June 21 and the All-Star break, an even 100 plate appearances, he hit .230/.420/.486. Ignore the low batting average, because it's by far the least important number here. That's a .906 OPS, which is stellar, and it all happened before the All-Star break. Harper has been even better than that since the break, as we noted above, and so the real takeaway you want comes down to this, among hitters in each time frame with 100 plate appearances:
Through June 20, Harper batted .209/.347/.462, which ranked 116th in a field of 332 hitters. Since June 21, he is batting .274/.440/.558, which ranks 15th among 205 hitters.
That's an enormous difference. June 20 is an arbitrary endpoint, but it's also when things seemed to stop being bad and started to get better for Harper, so it's what we're going with. That leads us to the obvious question: Why? What's happening here?

It's not as simple as you'd think. We have some ideas, though. First, let's compare some important metrics over the two parts of Harper's season.
Harper is walking more: 17 percent before, 23 percent after
But he's striking out more, too: 23 percent before, 31 percent after
That's... not what you'd expect. Who strikes out more and does better? It's important, though. Let's get back to that in a second.
Harper is not hitting more or fewer grounders: 39 percent before, 38 percent after
He's not really hitting it harder: 44 percent hard-hit rate before, 48 percent after.
A jump of four points isn't nothing, of course, so that's nice, but it's nothing dramatic, either. Nor is Harper suddenly putting the ball in the air, which we've often seen to fuel offensive production like this.
But Harper is using all fields more: 24 percent opposite field before, 33 percent after
That's interesting, because a huge part of Harper's low-batting-average issue seemed to be about hitting grounders into the shift. Somewhat unexpectedly, he has crushed the ball to his opposite field (at a 90.6 mph exit velocity for a .780 slugging percentage), just as well as he does to his pull field (at 90.6 mph for a .705 slugging percentage). That wasn't true in 2017, however, when Harper slugged 1.221 to his pull side against only .598 opposite field, so that might not be more than small-sample noise.
(This hasn't actually forced defenses to change anything, for what it's worth. Harper has seen a shift on an identical 54 percent of pitches both before and after June 20.)

Harper is seeing more fastballs that are being thrown harder: 51 percent fastballs at 93 mph before, 57 percent fastballs at 94.2 mph after
This may or may not be meaningful, but it's certainly worth noticing. Were teams thinking that Harper could no longer handle fastballs? He hit .222 against four-seam, two-seam and sinking fastballs before June 20. Then again, Harper hit only .164 against all other pitches, so this might not have anything to do with anything.
Harper is swinging less, but more aggressively: 47 percent swing rate before, 43 percent swing rate after; 26 percent swings on first-pitch strikes before, 32 percent after
Harper simply isn't swinging as often, and that might go a long way toward explaining that large increase in walk rate. He's swinging less at pitches in the zone (77 percent before, 71 percent after), but also less at pitches outside the zone (29 percent before, 25 percent after). Harper is, however, going after first-pitch strikes more often. It seems like he's got more of a plan now, and you can see that in his swing maps. Instead of swinging at strikes, he's swinging at his pitch.

That's especially true when we go back to just fastballs, which Harper is seeing more of since he's turned his year around. He's focusing on a zone up and near the middle. 

So it seems like Harper is trying to use all fields more often and doing it well, and that might be a good approach for a hitter who was being eaten up by the shift. But hang on a second: what about that increase in strikeout rate? He's actually struck out more every month of the season, from 16 percent to 24 (May) to 31 (June) to 34 (July), not including August. That doesn't correspond to Harper's season success path at all.
Striking out more isn't what you want, obviously. But there's a difference between good contact and bad contact, and over the past few months, there's been a pretty interesting correlation between Harper's rate of contact outside the zone and his overall production.

While it might not make sense that contact could be bad, it is when it's contact you don't want. On pitches outside the strike zone this year, Harper is hitting .115 with a .156 slugging percentage -- which is, of course, horrendous. Inside the zone, that's .287 with a .660 slugging. Looking at those numbers, at any time before a two-strike count, why in the world would you want Harper to make contact outside the zone? You wouldn't.
That's not the same thing as saying Harper is missing on purpose outside the zone in an attempt to live to fight another day. We have no idea if that's true or not. But what we do know is that the trend here is pretty clear. He's swinging at fewer pitches outside the zone and making less contact on them, and both are good things
"He's going to hit, and I know he's going to carry us for a month or two," Nationals manager Dave Martinez said to MLB.com's Jamal Collier on June 20, on what we didn't yet know was the beginning of Harper's turnaround. It hasn't all been Harper, because Juan Soto and Daniel Murphy have been as good or better over that time. But this looks like the Harper we expected, and it hasn't just been over a few games. It's been six weeks and counting.
The Indians didn't trade for Harper, instead adding Leonys Martin. The Braves went for Adam Duvall. The A's didn't add a bat at all. Nor did the D-backs, Rockies or Mariners. They might regret that now, and even more so in a few weeks. Then again, it's not over, because Harper could still be traded in August, however unlikely that might be. It helps a lot that he's looking like himself, once again.