Mythbusters: Home Run Derby Edition
If you watched the Home Run Derby on ESPN, you saw Yoenis Cespedes and his raw, yet explosive swing, hit 17 home runs in the first round of the derby. You also saw Chris Davis staying true to his swing and swinging at any pitch that he thought he could handle, hitting the ball where it’s pitched, and even swinging at some pitches that were borderline balls. If there was anyone to be concerned about changing his swing to fit the Derby, it was Davis–the guy who has so much strength that all he needs to do is stay within himself and swing easy to hit a homer. One might worry that Davis would swing too hard or try to pull everything, thus regressing into the “quadruple-A” player as he was once labeled, swinging and missing at a such a rate that he became a liability.
Anyone who has played baseball at a high level knows that a successfully executed sacrifice bunt, or grounder to the right side of the field with a man on second and nobody out, is frequently celebrated as much as a hit. Quality “team baseball” seems to be more effective than a mere amalgamation of flashy superstars that doesn’t mesh (I’m looking at you, 2012 Red Sox or 2013 Blue Jays). The Home Run Derby is kind of counter-intuitive to many MLB managers. Old-schoolers like Mike Scioscia would rather his players did not participate, saying, “I haven’t seen somebody come away from that derby and be a better player for it.”¹ The Home Run Derby turns the team game into an individual competition. Players exhaust themselves and risk tweaking their swings, but has the derby really affected the second-half performance of its participants?
To answer this question I looked at what goes into a player’s stats. There is a lot of luck involved in baseball, so I took a look at the differences in the way players hit the ball before the derby compared to after the derby. Looking at the past five derbies, I calculated the average batted-ball flight for players that were healthy for both halves of the season (38 players, excluding only Rickie Weeks in 2011 and Jose Bautista in 2012).
| LD% | GB% | FB% | IFFB% | HR/FB | |
| Pre HR Derby | 19.1 | 40.8 | 40.1 | 8.5 | .204 |
| Post HR Derby | 19.5 | 41.0 | 39.2 | 9.3 | .166 |
| Difference | <1% | <1% | <1% | <1% | .038 |
The consistency in the way players hit the ball is incredible. Derby participants hit the ball almost the same before and after the derby as a group. The HR to FB ratio drops considerably, and could explain a decrease in batting average and slugging percentage, as well as on-base percentage. It seems that players hit the ball the same way, just with slightly less power. Here are some of their standard stats from the second half:
| K% | AVG | OBP | SLG | OPS | ISO | BABIP | |
| Pre HR Derby | 17.87% | 0.302 | 0.385 | 0.570 | 0.956 | 0.268 | 0.322 |
| Post HR Derby | 19.60% | 0.282 | 0.369 | 0.499 | 0.869 | 0.217 | 0.316 |
| Difference | 1.73% | 0.020 | 0.016 | 0.071 | 0.087 | 0.051 | 0.006 |
Isolated Power (ISO) measures a hitter’s power in extra bases per at-bat (2B+3Bx2+HRx3)/AB. The large drop is ISO shows that indeed power does decrease for derby participants in the second half, and the overall line shows that players do perform worse. It’s not merely a function of hitting the ball to the wrong place, as the .oo6 drop in Bating Average of Balls in Play (BABIP) is not really significant. Players strike out a little bit more, but the notion that players change their swings and have trouble hitting the ball the same way after participating in the derby seems misguided when considering the small change in K% along with the consistent batted-ball percentages outlined in the first table.
Data suggests that players do perform worse in the second half of the season after participating in the HR derby, but that their performance isn’t due to a change in their swings. There have, however, been some significant changes in performance for some individuals. Taking a closer look at some of them, the poor performances can be explained without blaming the Home Run Derby.
| 2008 | Total derby HR | pre/post | AVG | SLG | OPS | ISO | BABIP | HR/FB |
| Dan Uggla | 6 | pre | 0.286 | 0.605 | 0.978 | 0.319 | 0.341 | 21.30% |
| FLA | post | 0.226 | 0.396 | 0.739 | 0.17 | 0.295 | 13.60% |
Uggla has a reputation as a streaky player, but he went from an MVP candidate in the first half to a guy who didn’t belong in the starting lineup after the derby. Taking a closer look, however, Uggla began slowing down in late June, and suffered a leg injury that kept him out nearly two weeks just prior to the All-Star Game. He only lasted one round, anyways, so it’s hard to blame the derby for his drop off, although it was certainly a big one.
| 2008 | Total derby HR | pre/post | AVG | SLG | ISO | BABIP | IFFB% | HR/FB |
| Lance Berkman | 14 | pre | 0.347 | 0.653 | 0.305 | 0.37 | 2.80% | 20.60% |
| HOU | post | 0.259 | 0.436 | 0.177 | 0.298 | 13.20% | 10.30% |
By 2008 Berkman had been a good hitter for many years. His second half was hurt by the amount of pop-ups he hit. a 10.4% increase in infield fly balls mean close to a 10% increase in outs, and his average decrease supports that notion. His increase in pop-ups could have been a result of an uppercut swing that developed in the derby, but his average had dropped 20 points in 16 games prior to the derby, and his career IFFB% is 11.5%, not too far off from his second half percentage. Perhaps the derby hurt Berkman’s swing, but more likely he was finally coming back down to earth after his torrid start.
| 2009 | Total derby HR | pre/post | K% | AVG | SLG | ISO | BABIP | HR/FB |
| Brandon Inge | 0 | pre | 24.60% | 0.268 | 0.515 | 0.247 | 0.304 | .22 |
| DET | post | 29.10% | 0.186 | 0.281 | 0.095 | 0.247 | .08 |
Brandon Inge? Yeah, Brandon Inge was in a Home Run Derby. He only has a career HR/FB ratio of .10, and a career batting average of .233, so his second half was closer to what Inge’s career looked like. Plus Inge didn’t even hit one out of the park, so could ten swings really ruin his season?
| 2009 | Total derby HR | pre/post | AVG | SLG | ISO | BABIP | IFFB% | HR/FB |
| Ryan Howard | 15 | pre | 0.257 | 0.529 | 0.272 | 0.301 | 1.10% | .23 |
| PHI | post | 0.305 | 0.621 | 0.316 | 0.352 | 0.00% | .28 |
Wait a second…? Was Ryan Howard better after participating in the derby? Yes! After the slugger hit 15 big flies in the derby, he went on to hit more homers in less at-bats afterward. With zero infield flies in the second half of the season, his swing was just fine.
| 2011 | Total derby HR | pre/post | K% | AVG | SLG | ISO | BABIP | IFFB% | HR/FB |
| Jose Bautista | 4 | pre | 14.40% | 0.334 | 0.702 | 0.368 | 0.321 | 11.50% | 27.40% |
| TOR | post | 20.40% | 0.257 | 0.477 | 0.22 | 0.291 | 20.50% | 15.40% |
After a hot start in April and May, Bautista had his worst month of the season in June, before the HR Derby. While Bautista was better overall before the derby, he was better in the two months following the derby than he was before it.
| 2012 | Total derby HR | AVG | SLG | ISO | HR/FB | |
| Prince Fielder | 28 | pre | 0.299 | 0.505 | 0.206 | 16.10% |
| DET | post | 0.331 | 0.558 | 0.227 | 20.00% |
Prince puts a lot of power into his swings, and when he hits 28 balls out of the park, he exerts a lot of energy. Prince won the derby in 2012, and continued winning games for the Tigers after the All Star Break. Hitting for a better average, and with an improved HR to FB ratio, Prince shows that the derby can kick start a player’s second half.
____________________________________________________________________________________________
Conclusion: The notion that participating in the Home Run Derby leads to a drop off in performance is a myth. Although data suggests that Home Run Derby participants do indeed regress in the second half of the season, the derby is not to blame. As baseball is a game of superstitions, players are aware that the derby can have harmful effects if they aren’t careful. Even Chris Davis was wary, saying, ”I wanted to be conscious of not changing my swing at all… I tried to stay up the middle and let the ball travel and not try to get pull heavy. But it looks a lot easier on TV than it really is. Once you get out there and start swinging and your adrenaline wears off, you realize how tough the Derby really is. It’s exhausting.”² While the derby curse isn’t real, it’s hard to continue chasing a 60-home-run season with a popped blister. Get some treatment on that hand, Chris.
1 http://www.latimes.com/sports/sportsnow/la-sp-sn-angels-relieved-mike-trout-not-in-home-run-derby-20130709,0,7051643.story
2 http://mlb.mlb.com/news/article.jsp?ymd=20130715&content_id=53853822&vkey=news_bal&c_id=bal&utm_source=twitterfeed&utm_medium=twitter
All data from Fangraphs.com
Georgetown Sports Analysis, Business and Research Group http://georgetownsportsanalysis.wordpress.com follow @GtownSports
Outstanding! Good delivery, interesting analysis and timely subject matter. I wonder, how has the league overall done from one half of baseball to another? Do players in general lose power in the second half due to fatigue? Some kind of control group or comparison point would help out here.
Otherwise, way to go.
I did wonder that – I expected to see a table showing that the second-half decline was in fact true of the league as a whole as players grew more tired (though there are other factors, like scrubs arriving on rosters in September).
But overall, really loved this piece and feel quite confident that the Home Run Derby fears are superstition.
The main increases seem to come in K% and IFFB%, which could be indicative of a player that is trying too hard to hit homeruns. Perhaps that player is swinging harder and with more of an uppercut, causing more K’s and popups.
Not sure if power increases or decreases empirically as the season goes by, but the generally hotter weather of the summer months would seem to support a power increase. Balls tend to carry better in warmer weather.
Speaking of Brandon Inge …
I think some consideration has to be taken into the player sample here. We don’t see all the same sluggers in the Derby year after year. We get new blood to keep things interesting, and much of that new blood is made up of players who had dynamite first halves, out of character with the rest of their career. These players are likely to regress regardless. Post-derby power drops may not be a cause of the derby, rather the natural expectation of a player’s true talent level…
This… It is probably (somewhat) introduced as a sampling error. Guys going to the home-run derby are not randomly selected samples, even of what might be labeled “power hitters” in the game. Because of the focus on first-half success, you necessarily have people who are, more often than not, set to regress. Your careful research also reveals that the regression is not an outlier for BFAB (before all-star break) and AFAB (after all-star break) as your numbers indicated BFAB slides by participants.
Thanks for the feedback! Research seems to indicate that most baseball statistics remain the same in the second half throughout the league. For every guy that regresses there is an Adam LaRoche or a Mark Teixeira who tend to rebound in the second half. It all averages out. Baseball is funny that way… you can run but you can’t hide from luck. Good luck will find you eventually when things are going bad, and bad luck will find guys when things are going Brandon Inge-good.
-@GtownSports