It’s happening again.  The baseball media, in search of a story, are attempting to form one out of thin air.  This isn’t a new story, it’s one that’s been floated out every year for almost two decades.  Beat writers, blog entryists, and commentators, worried about their star player all seem to bring it up at some point; “Will Player X’s performance be hurt in the second-half by their participation in the Home Run Derby?”

Horror stories of player after player are retold.  “Oh, his swing was just never the same afterwards.”  “Oh, it messed up his timing.”  “Oh, you swing differently when you’re trying to hit a homerun.”

The skeptical among us have always called bullhonkey on these notions.  I, for one, struggle to think of even one player who has had a career, or even the second-half of a season, distinctly change after their participation in the event.  Some smarter commentators and people close to the game bring up the fact that most players have daily home run derbies in batting practice.

Well, curiosity got the best of me and I did a quick study to find out whether or not there was a drastic difference in performance after the Home Run Derby.

Like I say, it was quick and by no means as comprehensive as it could be, but I decided to use OPS as the variable stat. It’s simple and to the point and although it’s by no means perfect, it does a decent job of estimating how good a player is at both getting on base and hitting for power; for sluggers especially, I think it works for this mini-study.

Secondly, I only looked at players that have participated in the event since 1995.  That was the first year MLB made it more than one round and consistently brought the same number of players every year.  Before ’95, there were as many as eight players and as few as four with only one round; whoever hit the most won.  In 2000, the number of players was brought down from ten to eight, where it has remained since.

Finally, before I started the study, I went to Baseball Reference and looked up the league-wide OPS between 1995 and 2010.  The total league average before the All-Star break is .743, while after the break, OPS falls to .722.  As the season wears on, players get tired and run-scoring invariably goes down.  It has gone down every year in the second half since 1939, with one exception, 2001, when the league average OPS went from .758 in the first half to .759 in the second.

So, it can be expected that the average OPS of players participating in the Home Run Derby should go down in the second half, since that is the tendency league-wide.

And, wouldn’t you know it, that’s exactly what happens.  The average first-half OPS of players participating in the Home Run Derby since 1995 is .983, while those same players after the break have had an OPS of .938.

The gap is slightly more pronounced than the league average OPS-drop, and I believe there are two explanations for this.  First, since the Derby participants are among the best sluggers in the game, their OPS is obviously further from the mean and therefore any differences will naturally be more pronounced.  Second, and more importantly, players in the Derby are being rewarded for good first-halves and given the small sample size that is a half-season, most of those players are bound to regress to their own averages as the season winds down.

On the individual level, out of 138 total players, only 11 players since 1995 have seen a drastic drop-off in OPS after they hit in the Derby.  Of those 11 there are a few that were having first-half aberrations and simply regressed to their normal selves after the break, (Joe Carter in 1996, Javy Lopez in 1998, B.J. Surhoff in 1999, Hank Blalock in 2004, and Brandon Inge in 2009) while a couple others got injured in the second half and clearly were affected by that since they have otherwise proven themselves to be consistent players (Ivan Rodriguez in 2000, Paul Konerko in 2002, and Justin Morneau in 2007).  The other four players (Raul Mondesi in 1995, Sammy Sosa in 2004, Bobby Abreu in 2005, and Dan Uggla in 2008) should be placed in the file labelled ‘coincidence’ and forgotten about.

Then there were the 15 players whose OPS actually increased significantly after the Derby and the vast majority that stayed roughly the same.  Those are the examples we simply never hear about.

It’s clear even looking through the data quickly that the myth of the Home Run Derby curse is complete nonsense.  Myth: Busted.

We can now go back to not worrying in the least about how tonight will affect Jose Bautista long-term.

You’re welcome.

Comments (6)

  1. Fun Fact from the research: Hee-Seop Choi has the lowest WAR and VORP of any player to ever participate in the Derby. That’s right…Hee-Seop Choi was in the Home Run Derby.

  2. Nice… I heard Tabby,(I think) basically debunk this quite logically as well by stating – rather obviously – that players have a HR derby pretty much ever day in their 3rd round of batting practice…. Where players just try to see how far they can hit it and seemingly, they’re all fine…

    Great article
    Curt

  3. Thanks for doing this, Travis. It’s also fun to point out that Ortiz has been doing this for what seems like several years now and never has a noticeable drop in the second half. In fact, at this point in the season, he’s just getting going!

  4. I was actually going to mention this and forgot: Of those 15 significant increases, Ortiz has three of them. He also has the largest and third largest single increase of anyone since ’95

  5. In looking at the 40 finalists and semi-finalists over the past 10 years, statistics show 60 percent of players saw a decrease in slugging percentage. Of those 24 players, nine suffered what can be considered a major loss of power (-0.100 in slugging percentage). In comparison only two players gained that much in slugging percentage.

    Slugging Percentage
    Year Name 1st Half 2nd Half Difference
    2002 Sammy Sosa 0.641 0.536 -0.105
    2006 David Wright(notes) 0.575 0.469 -0.106
    2005 Bobby Abreu 0.526 0.411 -0.115
    2001 Luis Gonzalez 0.745 0.62 -0.125
    2003 Garret Anderson(notes) 0.597 0.463 -0.134
    2009 Albert Pujols(notes) 0.723 0.582 -0.141
    2003 Jim Edmonds(notes) 0.668 0.507 -0.161
    2002 Paul Konerko(notes) 0.571 0.402 -0.169
    2008 Lance Berkman(notes) 0.653 0.436 -0.217
    Average .562 .432 -.130

    Players chosen for the Home Run Derby are top hitters from the first half of the season (this season captains were named to select the AL and NL sides). Some players have been invited on the strength of a great first half of the season. These are the players who fit the profile for a substantial decline in the second half.

    If you are going to take this kind of look at the Home Run Derby participants, it is also necessary to look at the sluggers that did not compete. AccuScore looked at the top five home run hitters who did not participate in the derby for each of the past 10 seasons. A whopping 70 percent of those players actually improved their slugging percentage after the All-Star break.

    Slugging Percentage
    Year Name 1st Half 2nd Half Difference
    2004 Albert Pujols 0.599 0.721 0.122
    2003 Alex Rodriguez(notes) 0.544 0.679 0.135
    2007 David Ortiz(notes) 0.556 0.695 0.139
    2001 Richie Sexson 0.476 0.625 0.149
    2008 Carlos Delgado(notes) 0.455 0.606 0.151
    2001 Shawn Green 0.527 0.682 0.155
    2010 Jose Bautista(notes) 0.543 0.702 0.159
    2002 Jim Thome(notes) 0.604 0.773 0.169
    2008 Manny Ramirez(notes) 0.518 0.723 0.205
    Average 0.535778 0.689556 0.153778
    The Home Run Derby curse is real. Since 2001, derby participants have averaged a .025 decline in slugging percentage. The top home run hitters who did not participate averaged a .036 increase in slugging.

  6. That’s exactly what I meant by the regression talk though; the players invited to the derby are the guys who’ve had the craziest first halves power-wise, they’re likely to regress as the season goes on. Looking at the other top homerun hitters who didn’t make it just means that they had their hot streaks in the second half.

    This is why comparing it to the league average makes way more sense. The league average has a much larger sample size than just the few top homerun hitters who weren’t on the derby card.

Leave a Reply

Your email address will not be published. Required fields are marked *