Another look at the Home Run Derby hangover effect

Those who have been hanging around these parts for a while might remember my article from this time last year about the “Home Run Derby Hangover Effect”. That article has received a lot of attention recently from our friends Rob Neyer at ESPN, Tom Tango at The Book Blog, and Aaron Gleeman at NBC (thanks guys). Last year, I couldn’t find any evidence that a “Home Run Derby hangover effect” existed. Today, I thought I’d take another stab at finding it.

The Home Run Derby hangover effect

As a quick introduction, allow me to quote myself from last year:

For years now, we’ve heard how players who participate in the Home Run Derby screw up their swing or tire more easily in the second half of the year. It’s gotten to the point where players are declining invitations to the Home Run Derby in droves.

This year, we witnessed a huge uproar from New York fans and sportswriters when Robinson Cano announced he would be participating in the Derby. The problem, of course, is that we’ve yet to see a single piece of credible evidence to support such backlash.

A second look

For our second look at this effect, I decided to take all the Home Run Derby participants since 2001 and put them into one bucket. Then, I’d fill another bucket with similar players who could have participated in the Derby but, for whatever reason, did not.

I matched each player up with a “similar” player individually (and manually), although my criteria weren’t anything super-rigorous. I tried to define “similar” as players who had a similar first half, were of a similar age, played the same position, and had the same type of skills, where possible. The only players I removed from the study were Barry Bonds (2001 and 2002) and Albert Pujols (2009), because comparable players didn’t seem to exist. Of course this is all subjective and somewhat arbitrary, but I thought it would make for an interesting article.

+-------+----------+----------+------+----------+----------+------+
|       |     Derby Participants     |        Control Group       | 
+-------+----------+----------+------+----------+----------+------+
| Year  | 1H AB/HR | 2H AB/HR | Diff | 1H AB/HR | 2H AB/HR | Diff |
+-------+----------+----------+------+----------+----------+------+
| 2009  |     14.0 |     15.6 |  1.6 |     13.9 |     15.8 |  1.9 |
| 2008  |     16.7 |     23.4 |  6.7 |     16.8 |     17.1 |  0.3 |
| 2007  |     17.1 |     17.2 |  0.1 |     17.1 |     15.3 | -1.8 |
| 2006  |     13.9 |     15.2 |  1.3 |     13.9 |     16.7 |  2.9 |
| 2005  |     17.0 |     17.7 |  0.8 |     17.0 |     16.4 | -0.6 |
| 2004  |     14.7 |     16.0 |  1.3 |     14.8 |     14.7 | -0.1 |
| 2003  |     13.5 |     16.7 |  3.2 |     13.5 |     15.1 |  1.6 |
| 2002  |     13.6 |     16.3 |  2.7 |     13.3 |     16.2 |  2.8 |
| 2001  |     12.5 |     11.9 | -0.6 |     12.6 |     17.7 |  5.1 |
+-------+----------+----------+------+----------+----------+------+
| Total |     14.7 |     16.2 |  1.5 |     14.7 |     16.0 |  1.3 |
+-------+----------+----------+------+----------+----------+------+

What we see is that the Home Run Derby participants and our control group have identical first-half home run rates and nearly identical second-half home run rates. They differed significantly in 2008 (Lance Berkman, Dan Uggla, Chase Utley, and Grady Sizemore all had steep declines), but that’s the only real outlier here. On the whole, we again find that the Home Run Derby has no effect on a player’s second half. And with more than 20,000 at-bats in each bucket, our sample size is pretty large.

Study caveats

Of course, there are some caveats to this:

Generalizing to all players: This study looks at the participants on the whole. We are dealing with human beings, though, each having their own unique swings and physiologies. It’s entirely possible some players are affected by the Derby, even if the overall effect is small (or non-existent).
Derby participants: There might be some additional selection bias in who participates in the Derby. If a player is legitimately affected by the Derby, he is less likely to participate in future years and thus will be included in the study only once.
Steroids: A study like this necessitates using many years, since we have only eight sample points per year, but in doing so we look at years when guys like Sammy Sosa, Jason Giambi and Rafael Palmeiro were playing. Can we really say that the effects in these years will be the same as those in 2009?

So where has this theory come from?

While the theory doesn’t appear to be true, we’re still likely to hear about it from the mainstream media over the next few hours and days. Why do some in the media seem to believe this? Here are a few possible reasons:

Selection bias!: Those selected to participated in the Derby likely overperformed in the first half, so second-half regression to the mean is viewed by the uninformed as a decline rather than what it actually is—mere normalization.
2008: As noted earlier, 2008 seemed to “prove” the theory in a big way, and it was just two years ago, so people are going to remember.
Raw totals: Because the true 50 percent mark often occurs a couple of weeks before the All-Star break, “first half” totals can look inflated if compared directly to “second half” totals.
The media: Many love to use anecdotal evidence. For every 2005 Bobby Abreu there are 10 2009 Nelson Cruzes, but those pushing the “Derby effect” mention only Abreu.
Outspoken players: Reporters are a lot more likely to listen to players than look at numbers, and when players start blaming the Derby for second-half struggles, it’s an easy story to run with. Also, the players that do speak out have incentive to do so. If they have a poor half-year, that could hurt when it comes time to sign a new contract. Whether the Derby actually affected their swing or not, it makes sense to use it as an excuse for poor numbers.
Snowball effect: Once players start talking and complaining, it makes other players less likely to want to participate and draws more attention to the situation, creating a snowball effect.

2010 participants

So what does this mean for the participants in tonight’s 2010 Home Run Derby?

Miguel Cabrera
David Ortiz
Vernon Wells
Nick Swisher
Corey Hart
Matt Holliday
Hanley Ramirez
Chris Young

Absolutely nothing. I wouldn’t worry at all if you own one of these guys, and you might find it easier to acquire one if his owner is worried.

A Hardball Times Update
Goodbye for now.

5 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
Nick Steiner
14 years ago

Great job Derek.

John K
14 years ago

Wow – I think I’ve been reading espn fantasy analysis too much, b/c I just crapped myself when you appropriately used the term “selection bias.”

Andrew
14 years ago

Unrelated to this post, but any recommended Buys for the second half, Derek?

You seem to have a lot of standard Buy Low types on your squads like Quentin, Bruce, Stewart, C Pena, Nolasco, and Scherzer. That should bode well for your teams going forward.

Derek Carty
14 years ago

Thanks Nick and John K.

Andrew, it does seem that my team is loaded with these kinds of guys.  It’s probably too late to buy low on Quentin, but I do like him if he stays healthy.  Same deal with Scherzer.  If you can buy low on these guys, I’d absolutely do it.  I think Bruce should increase his power output, as should Stewart.  Nolasco has started to turn it around of late, and I’m definitely buying into him.  Haren, if his owner doesn’t realize he has fantastic peripherals, is a terrific option.  Baker’s in this boat too, though obviously not to the extent of Haren.  Chris Davis, if he hasn’t been grabbed yet, needs to be.  Nelson Cruz hasn’t done much lately, but I think he’s a top hitter, when healthy, if he is commanding that price tag.  Aaron Hill is a pretty decent bounceback candidate, and it’s not like his power has been bad.

Andrew
14 years ago

Thanks for the comment, Derek. Good luck in the second half.