I returned from a road trip up the California coast to what I will gently
call a flood of, um, "constructively critical" e-mail from Yankee
fans, responding to last Thursday’s piece for ESPN.com. Buried amidst the
profanity, though, was some good response to
the Daily Prospectus of the
same day, which dealt with Curt Schilling‘s transformation.
Schilling, I noted, had dramatically dropped his pitch counts and walk rate
after being traded to the Diamondbacks last July.
As much to deflect anticipated questions as anything else, I mentioned in
the piece that one possible explanation was park effects. The Vet increased
walks last season (walk park factor from Stats, Inc.: 110), while Bank One
Ballpark was essentially neutral (WPF of 99).
This caused a few people to ask how a park could impact walk rates. Most
readers of this site are familiar with park effects on run scoring and on
the elements thereof, particularly home runs. But there is variation among
the frequency of walks from ballpark to ballpark.
Is the variation random, or an actual feature of the park? In the same way
that Coors Field increases run scoring every year, or Dodger Stadium
decreases scoring, a park that actually has an impact on walks should
appear to do so over time. If walk park factors are real, we should see
some year-to-year correlation among the leaders and trailers.
Here are the top walk park factors for the parks in use for all three
seasons from 1998 through 2000 (all data courtesy Stats, Inc.’s
League Handbook, a resource I strongly recommend):
2000 1999 1998 Park WPF Park WPF Park WPF St. Louis 119 Philadelphia 118 Cincinnati 122 Baltimore 112 Minnesota 109 Minnesota 113 Philadelphia 110 Cincinnati 108 Chicago (A) 107 Texas 107 New York (A) 108 Philadelphia 106 Colorado 105 Anaheim 106 Cleveland 106
The only park to appear in the top three all three seasons was Veterans
Stadium, which perhaps gives credence to the idea that the change in
Schilling’s control is in part a change in park. Cinergy Field and the
Metrodome appear twice each, and eight parks appear once. Of those eight,
only Busch Stadium in St. Louis increased walks in both of the other two
Same chart, bottom five in WPF:
2000 1999 1998 Park WPF Park WPF Park WPF San Diego 88 Atlanta 88 Arizona 81 Montreal 90 Arizona 93 Baltimore 87 Kansas City 91 Los Angeles 93 Oakland 89 Boston 91 Chicago (A) 93 Tampa Bay 91 New York (N) 92 Kansas City 94 LA/Tex/NY (A) 92
The first thing I see here is that of the 11 parks to appear in the top
five in WPF from 1998 through 2000, just three make appearances in the
bottom five in any year (Comiskey Park, Yankee Stadium, Camden Yards).
Also, we again see parks like Kauffman Stadium, Dodger Stadium and Bank One
Ballpark making multiple top-five appearances.
This isn’t an exhaustive or conclusive study, but it’s enough to indicate
that parks do have some impact on walks, if not the influence they have on
The more interesting question is, " why does this happen?" Some
of this is speculation, but the most significant factor is probably
visibility for hitters. Does the park have good lighting? Is there a good
"batter’s eye" in place that enables hitters to pick up the pitch?
Other factors include the condition of the mound (theoretically standard,
heights have been said to vary, and the surface usually does) and the
amount of foul territory in a park (fewer foulouts should lead to a few
more walks). There may be other factors as well, but these are the ones for
which an initial case can be made. I’d be interested to hear some other
ideas about why, say, the Vet is a good walking park or the Bob isn’t.
Joe Sheehan is an author of Baseball Prospectus. Contact him by