"I hope [my favorite player] doesn't get selected for the Home Run Derby," went much of baseball fandom last week before the participants were announced, "because it will screw up his swing and ruin his second half."
Remember when Joc Pederson hit 20 homers before the break in 2015, then 39 more in the Derby -- but then only six more for the remainder of the season, eventually being benched in September?
It's easy to blame the Derby for ruining Pederson's season. It's especially easy when you choose not to notice that after finishing off a streak of five homers in five days on June 3, he hit all of .175/.322/.308 in 36 more games before the Derby. In reality, Pederson's slump began weeks before the Derby ever arrived, not because of it. And despite the fact that there have been near-countless studies over the years showing that the Derby isn't the cause of (most) second-half downturns, plus many players themselves saying it's not real, the myth still persists.
Now that's not the same thing as saying that none of these eight hitters will have decreased production in the second half, because at least a few of them absolutely will. If you're judging these things merely on "number of home runs hit," then most of them will hit fewer simply because "first half" is a misnomer. Some teams will hit the break having played 93 games. Charlie Blackmon will have received about 400 plate appearances, and he's certainly not getting 400 more. They'll simply have fewer chances to hit home runs.
But far more importantly, what we're talking about is a great example of selection bias. That is, the participants in the Home Run Derby have been chosen for this contest specifically because they've spent their first halves crushing baseballs at an elite level, in most cases far above their career baseline or any reasonable expectations -- numbers they can maintain for a few weeks or months, but generally not for a full season. If they come back to earth somewhat in the second half, that was almost certainly going to happen anyway, Derby or not.
Take, for example, Yankees slugger Aaron Judge. He's having a sensational season by any definition one could possibly think of, but think about what it would mean if he simply continued this pace for the rest of the year, finishing with the same .331/.449/.697 line he has now. It would be, without hyperbole, one of the 30 greatest hitting seasons in the history of modern baseball. Judge's .427 BABIP would be the highest ever, topping Babe Ruth. Judge is good; he's probably not that good.
That's the entire point, really. You can't simply take a player's first half and double it to get a full-season number. If you could, Alex Wood would go 20-0. Chris Sale would strike out 332 hitters. Great as they are, it's not going to happen. So if Judge hits "only" 20 homers in the second half after 29 (through Wednesday) in the first, it's not because of the Derby. It's because he's been so unbelievably good in the first half that it's unrealistic to expect him to maintain it. (And if, by some miracle, he does? Well, then the Derby definitely didn't hurt him.)
Think about it this way: If a player in the Derby hits fewer homers in the second half, you'll notice, and you'll suggest it's because of his participation in the event. But if a player not in the Derby improves his second-half dinger total, will anyone say it's because he wasn't a part of the contest? Of course not.
But OK, enough with theories. How about some numbers? Let's try to keep this simple. It's not whether Derby participants do worse in the second half; we know many do. It's about whether they do worse than non-Derby All-Stars.
As we said, many studies have looked at this before, so rather than go back many years to re-prove a point that's pretty well proven, let's keep it simple and stick to the last three years, with a very simple question. If that's not enough of a sample, well, this is again just adding on to previous evidence.
Did Derby participants underperform relative to other All-Stars?
No. Not in the past three years, where we're looking.
In 2015, when Todd Frazier won, the eight Derby participants dropped from a .380 first-half wOBA to a .341 second-half mark, a 39-point decline. Again, all of the non-Derby All-Stars fell back too, from .364 to .339. As a group, the post-break All-Stars all performed identically (hence .341 vs .339), but the Derby entrants had overperformed before the break.
In 2016, when Giancarlo Stanton topped Frazier, it was much the same story. The eight Derby entrants saw their wOBA drop from .367 to .330, again fueling the story that the Derby was at fault. But that 37-point drop was almost exactly mirrored by the 39-point drop from the non-Derby All-Stars, who went from .379 to .340.
As a group, All-Stars are likely to do less damage post-break -- not just Derby hitters. It's the nature of having a great first half.
So, what might happen to our participants? No one can tell the future, of course, but there is a way to test the premise. The respected Steamer projection system takes into account a player's past performance to output an expected future line, and by looking at what they've actually done so far in 2017 plus what's projected for the remainder of the season, we can look at 2017 projections for our eight sluggers.
These are almost uniformly below what the players are currently doing, but that's the point. If you come back here in October to check what actually happened, you'll get a better idea of how accurate these were -- and if the Derby really had any impact.
(wOBA, or Weighted On-Base Average, is just like on-base percentage except it gives more weight to homers and extra-base hits than singles. The 2017 Major League average wOBA is .321.)
Remember, last year, when Mark Trumbo went from a monstrous .288/.341/.582 (.387 wOBA) first half to a more pedestrian .214/.284/.470 (.321 wOBA) second half, it's not because the Derby "broke" him. It's because that second half was far more in line with his career mark of .252/.305/.467 (.331 wOBA) than the out-of-character first half was.
You don't get to the Derby if you're not having a tremendous first half -- but that first half is often so good it's impossible to keep up. Remember that, when Judge or Bellinger or Moustakas doesn't have quite the same second half as they did in the first. It's not the Derby. It's expected.
On Tuesday at 7:30 p.m. ET, tune in to the 2017 All-Star Game presented by Mastercard live on FOX, and during the game visit MLB.com to submit your choice for the Ted Williams Most Valuable Player Award presented by Chevrolet via the 2017 MLB All-Star Game MVP Vote. The 88th All-Star Game, in Miami, will be televised nationally by FOX, in Canada by Rogers Sportsnet and RDS, and worldwide by partners in more than 160 countries via MLB International's independent feed. ESPN Radio and ESPN Radio Deportes will provide national radio coverage of the All-Star Game. MLB.com, MLB Network and SiriusXM will also provide comprehensive All-Star Week coverage. For more information, please visit allstargame.com.
Mike Petriello is an analyst for MLB.com and the host of the Statcast podcast. He has previously written for ESPN Insider and FanGraphs. This story was not subject to the approval of Major League Baseball or its clubs.