June 4th, 2013

Today's slate.

The Big Hurst - Felix Hernandez

skoormit - Jordan Zimmerman

La Osa Rosa - Tim Lincecum


skoormit: I like J Z vs the Mets. Oh, and the park factors thing: I recall a BP study that determined that there's so much randomness in the numbers that it takes something like two years to get meaningful numbers.
The Big Hurst: Well, that's a valid point.  It's really just a starting point to attack a very interesting question.  I agree that one year is a smallish sample size for a park factor. But we're approaching two years of numbers.  When there's that big of a change between the full season of 2012 and the one-third season of 2013, you've got to wonder whether this is noise or static.  Especially for the big ones, like both Chicago teams, Philadelphia, and Colorado.  Anything around a change of 0.400 is a huge variation.  I've got some data on this:

Between 2009-2010, only 1 team's park (NYY) changed more than 0.200.  There were 14 parks (out of 30) that changed more than 0.100.

Between 2010-2011, one park changed more than 0.300 (TEX).  Four parks changed more than 0.200.  And only six total parks (out of 30) changed more than 0.100.

Between 2011-2012, three parks changed more than 0.200 (TEX, CHW, & COL).  Eleven parks changed more than 0.100.

So there were 90 opportunities for a park factor to drift in this data.  Only once did a park change more than 0.300 in a year.  That's 1%.  Eight parks changed more than 0.200.  That's 9%.  Thirty-one parks changed more than 0.100.  That's 34%.  That feels about right.

What does this mean?  Year to year, there's some noisy drift in park factors.  But most values tend to stay fairly consistent.  I don't know what causes these numbers to change, but I can tell you that +/- 0.300 would be a fairly huge jump.  And I still wonder if there's a good reason in 2013.  I wish I knew fans in these cities, so they could write in and tell us whether these ballparks might've changed in the offseason.

As one example, historically, ARI's park has played at a significantly hitter-friendly 1.140 from 2009-2012.  But in 2013, it's now a slight pitcher's park at 0.953.  If they've changed something out there, I'd think that's a pretty big deal.

As a curiosity, I can report that the most consistent parks from 2009-2013 may be DET, OAK, WAS, CIN, STL, & LAD.  (Though there's a clear trend of CIN slowly getting more hitter-friendly over this time.)  The least consistent ballparks may be BAL, TOR, CHW, TEX, SEA, HOU, SFG, & COL.  (Though SFG was exactly the same between 2011 and 2012.)


RESULTS


The Big Hurst: With a traffic jam at the top, Tim Lincecum finishes in the money.  Zimmermann would've been higher, but for one little throwing error - not even his fault.  Baseball is maddening.  How come King Felix is great for you two and only mediocre for me?
skoormit: Felix is fickle. At least they let him go 100+ pitches this time.
La Osa Rosa: Sweet, precious Lincecum did well with me rooting him on! I’ll have to pick him again. It’s good for his career.


 

No comments:

Post a Comment