Andrew McCutchen was named to the home run derby tonight. Immediately after the announcement, Pirate fans began reacting. Some were extremely excited to watch their favorite Pirate get a national spotlight, and others less than thrilled about the idea because of the belief that participants in the derby have worse second halves than first halves. So for the last couple hours I've been collecting data to try and figure out if that rumor is true.
I used the last five home run derbies. I took all of the contestants and compared their first half numbers (in average, slugging percentage, and home runs per at-bat) and compared them to their second half numbers. Not only that, but I compared them to their numbers from 10 games after the derby and 30 games after the derby. This way we'll be able to tell if the players suffer a short term or long term drop off in numbers; if there is a drop off at all.
What I found was that the average home run derby participant over the last five years experiences the following changes in their stats.
10 games after the derby
BA -.019
SLG -.072
AB/HR +6.49
30 games after the derby
BA -.013
SLG -.045
AB/HR +8.02
2nd half of the season
BA -.008
SLG -.037
AB/HR +4.26
You can see that the derby has definitely had an effect on its participants. In terms of batting average and slugging percentage, hitters experience a temporary decline in performance. They hit 19 points lower in the first ten games after the derby, 13 points lower in the first 30, and only eight points lower in the second half as a whole. Only 14 of 40 hitters had a better batting average in the second half than the first half. In slugging percentage, 15 hitters had better averages in the second half. That means that more than half the derby participants had worse second halves in terms of batting average and slugging percentage.
In terms of at-bats per home run, the effect isn't on the same curve. However, it does take batters significantly more at bats to get home runs after they participate in the derby. Only 10 of the 40 participants in the study hit hit home runs at a better rate in the second half than in the first half. On average, it took batters 6.5 more at bats to hit a home run in the first ten games after the derby, 8 more at bats in the first 30 games after the derby, and 4.3 more at bats in the second half as a whole. Only 8 of 40 hitters hit home runs at a faster pace in the second half than the first half.
My next question was if hitters that took more swings in the derby suffered a worse drop off. To figure this out, I took the numbers from just the top two finishers in each derby. Here are those results:
10 games after the derby
BA -.032
SLG -.098
AB/HR +4.48
30 games after the derby
BA -.021
SLG -.039
AB/HR +13.55
2nd half of the season
BA -.011
SLG -.035
AB/HR +2.58
In average and slugging percentage, there is a noticeably bigger difference for the top two participants (the two guys that go an extra round or extra two rounds than everyone else) compared to the rest of the participants. For the second half, it took the top two participants 2.6 more at bats to hit a home run than it did in the first half. They struggle a bunch in the first 30 games, taking an extra 13.6 at bats to hit home runs. Granted, this is only looking at 10 batters over the last five years.
Now a possible explanation to some of these number declines is that players that make the home run derby have had huge first halves to the season, and they are due for natural regression anyway. You could compare these same trends to all-star players that aren't in the home run derby and see if their numbers changed in the same way. I'd rather not spend all night in front of an Excel spreadsheet though, so this is going to have to go unanswered for right now. The one thing that does kind of argue against these points is that the players seem to be the worst in the two weeks right after the derby. The players are well worse than their first halves right after the derby, and then they start to get better as they get further and further away from it. Is it because they haven't been in a game situation for a few days over the break? Could be, again, there's no way real way to know for sure. However, the information is interesting nonetheless.
So, in conclusion, there is a significant drop-off in performance for players that participate in the home run derby. On average, their average and slugging percentage both drop in the second half, and they also hit less home runs per at bat. It might not be fully because of the derby, but the data shows that there is definitely something going on.
Let's hope that Andrew McCutchen is an exception to what we've found here, because the home run derby has not been kind to its participants in the last five years. If McCutchen doesn't have as good a second half as he had first half, it's probably not going to be because he was in the home run derby – it's probably just because it's really, really hard to do what he's done so far this year.