The HR Derby Curse is real!

Zach the intern — well, he’s not really an intern anymore but I still like to call him that — posted recently on the blog that “it seems that the “Home Run Derby Curse” is just a myth”. Well, as an agnostic I’m never one to just settle for blind faith, so I decided to do a little research on this subject and see if there’s any truth to that. Call it the first installment of Fantasy 411 Mythbusters.


First, I got the stats for every player who competed in the Home Run Derby from 1999 to 2007, broken out into pre-All-Star break and post-All-Star break. I probably could’ve gone back before that since the Derby dates to 1985, but this is the data I had on hand… plus, the Derby only switched to the current three-round format (or something like it) in 2000, so even the 1999 data might not be completely relevant. But I had it so I used it.


First I compared the stats for all players combined both before and after the All-Star break, and pro-rated them out to 162 games for easy comparison. The results make it clear: there IS a decline in second half stats for those players who have competed in the Derby since 1999. A few facts and figures:


* 43 of the 74 players who have competed saw their home run production decline in the second half, measured by plate appearances per home run;


* The average player hit one homer per 16.22 PA before the break, with a .995 OPS, but homered only once per 17.85 PA after the break with a .970 OPS.


* 14 players out of 74 – nearly 20 percent! – hit fewer than 10 homers after the All-Star break, although to be fair, one had only 104 at-bats and another only 38, as injuries curtailed their production.


Here are the overall numbers (pro-rated to 162 games):


SPLIT   AB     HR     RBI    BB     SO    AVG  OBP  SLG  OPS  PA/HR

Pre       593    43      130    88      113    .307   .399   .595   .995   16.22

Post     586    39      118    95      118    .300   .402   .569   .970   17.85


This makes it clear that, at least since 1999, the Curse is real, but then again, is anyone really going to complain about the difference between 43 homers and 39? That’s just greedy!

Hamilton.jpgSo now we know there is indeed a Curse, although it’s not a major one. But there’s a twist… while 43 of the 74 players saw a decrease in their production after the Derby, that means 31 of them — almost 42 percent — saw an increase, or no change, in their production!

So the real question becomes, how do we figure out who will be the second-half gainers and who will be the losers? If you could trade Justin Morneau ’07 (26 before the Derby, only 7 after) for David Ortiz ’05 (21 before vs. 27 after, in 43 fewer at-bats), you’d do that in a second, wouldn’t you?


To try and figure this out, I broke out the stats into three groups, based on how far each player advanced in the Derby: one round and out (including all ’99 players since they only each got one turn in an AL vs. NL format), semifinalists, and finalists. Here’s where the results get interesting, and perhaps telling:


First round only (42 players):

SPLIT   AB     HR     RBI    BB     SO    AVG  OBP  SLG  OPS  PA/HR

Pre       599    44      133    84      116    .307   .395   .599   .994   15.73

Post     586    39      119    91      123    .296   .394   .560   .954   17.66


Semifinalists (16 players):

SPLIT   AB     HR     RBI    BB     SO    AVG  OBP  SLG  OPS  PA/HR

Pre       574    42      126    104    105    .308   .417   .598   1.015 16.51

Post     562    39      115    110    104    .302   .421   .588   1.009 17.79


Finalists (16 players):

SPLIT   AB     HR     RBI    BB     SO    AVG  OBP  SLG  OPS  PA/HR

Pre       599    40      127    82      113    .308   .393   .583   .977   17.28

Post     607    39      118    88      119    .311   .401   .573   .975   18.39


Whoa, now THAT is interesting, isn’t it? Players who went out after the first round saw a considerable dip in their numbers, and while those who made it to the second round or the finals saw their HR per PA decline, their combined OPS was virtually unchanged, and those who reached the finals produced almost identical numbers in the second half. So in reality, the Derby Curse really only exists for those players who don’t get out of the first round… advancing to the semifinals, if not further, seems to excise the curse.


To be fair, we are dealing with extremely limited sample sizes, but when combined, the numbers do seem pretty clear: if you’ve got Josh Hamilton, Justin Morneau, Lance Berkman or Ryan Braun, sit tight… the second half should treat these players pretty well.


On the other hand, if you’ve got Dan Uggla, Grady Sizemore, Chase Utley or Evan Longoria, you might have some cause for concern… beware the Curse!





P.S. – The complete spreadsheet I used is attached here (HR Derby splits 1999-2007.xls) in case anyone wants to tinker further and provide more detail on these numbers, or even a rebuke. I didn’t consider a tremendous range of other factors that could come into play — such as the player’s physical size, career track record, home/away splits before and after the break, etc. etc. – but this is at least a jumping off point for further investigation.




I think the confusion here is between correlation and causation. While the data may show a slight dip, it can not be inferred that the HR Derby caused it. There are too many variables in play.

1. How does the data compare to players not in the home run derby? I would gamble that if you measured the player pool as a whole there would be a slight dip in the second half numbers. Players often tire as the season wears on and have a dip in stats in the second half.

2. Players chosen for the HR Derby are often hitters having a very hot start to the season. If they cool off in the 2nd half, its attributed to the Home Run Derby rather than natural stat correction. For example, if Marcus Thames went to the HR Derby after his 17 first half HR & then hit 5 more in the 2nd half, it would be a big story. But, if he now hits 5 more in the 2nd half without the HR Derby it would be viewed as his stats evening out and totally normal.

As a whole, I think its silly to attribute any drop in performance to the Home Run Derby unless similar analysis is done on the player pool as a whole and other players with similar numbers before the break who DO NOT go to the HR Derby.

Since according to the Derby Curse theory players second halves should mirror the first half, if the following players don’t end the season with over 35 homers, blame the curse of NOT being in the Home Run Derby:

Pat Burrell
Carlos Quentin
Ryan Ludwick
Nate McLouth
Jack Cust
Aubrey Huff
Rick Ankiel
Marcus Thames

Ankiel will approach if not top 40, as I predicted on yesterday’s show, which will debunk the non-HR Derby Curse.🙂


Touche’ – But then shouldn’t David Ortiz in ’05 debunk the HR Derby Curse?😉

Interesting stuff Cory and posses a very enjoyable debate.

I also have to say that the findings you guys have come up with about pitching the outside vs pitching the inside is an extraordinary find for baseball, and strategy applied to pitching. Great job!

-Alex in Chicago-

You guys didn’t put up Tuesday’s show… Is that going to go up eventually?


Charles in DC

Thanks that was fun stats.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: