Saturday, September 13, 2008

Nationals Park: How's It Playing?

One of the big questions going into this season was how the new Nationals Park would play. RFK was famously one of the league's most extreme pitchers' parks, with only Petco Park in San Diego suppressing runs more.

No matter what, Nationals Park would be more favorable to hitters. Guys like Austin Kearns (we thought) stood to gain from the move, while we all held our breath for pitchers like Chad Cordero and Jason Bergmann.

Headed into the season, Stan Kasten said that the park would play "fair," not an extreme hitters' or pitchers' park, and based on dimensions alone, Nationals Park indeed looked like a slight, but not extreme, pitchers' park:
LF: 336
LC: 377
CF: 402
RC: 370
RF: 335

Right-center field is a bit of a short-porch, the 8th shortest distance in the league, although the high fence above the out-of-town scoreboard out there reduces that effect a bit. The centerfield fence is right in the middle of the pack compared to other stadia, while left field, left center, and right field fences are all in the bottom third in baseball (meaning the farthest third).

Of course, park effect is a funny thing. Dimensions are the most obvious factor, but wind, altitude, humidity, heat... all these elements and more factor in. For instance, many speculated that Nationals Park's proximity to the river and gaps in the grandstands created the potential for wind tunnel effects that could cause balls to carry, especially to that short porch in right.

To really get a better handle on the park effect, you need to look at a stat called park factor. It's really pretty simple--park factor compares how the home team scores at home to how they score on the road.

The most common calculation goes like this:
PF = ((homeRS + homeRA)/(homeG)) / ((roadRS + roadRA)/(roadG))

A rate higher than 1.000 favors the hitter. Below 1.000 favors the pitcher. maintains various PF stats here.

There are a lot of problems with this stat, mainly that there's just a lot of random statistical noise in there. Things just don't even out over the course of 162 games. Even if every environmental factor and the players themselves were all kept exactly even all season, there would be some statistical randomness, like when you flip a coin 100 times you don't always get 50 heads and 50 tails. But of course all those other factors aren't equal. If the Nationals happen to catch Santana and Hamels twice each at home, that will skew the results.

Statheads prefer to average out the park factor over three years to reduce the distortions, but even that's not perfect, as it doesn't account for roster changes, unbalanced schedules, injuries, etc. Hardball Times did a detailed column last year on the shortcomings of this basic park factor stat, if you're interested.

Nonetheless, at this point in the season we can look at the numbers and begin to get a sense of how it's shaping up. Here are the park factors for a set of key offensive stats (league rank in parentheses).
Runs: 1.023 (16)
HR: 0.915 (21)
H: 0.994 (17)
2B: 1.016 (14)
3B: 0.905 (17)
BB: 0.980 (19)

Overall, the rankings are pretty much right in the middle. For runs and doubles, Nationals Park has been a very, very slight hitters' park, while HRs, hits, triples, and walks have been suppressed slightly. (If there's a stat you would think wouldn't be affected by the park, it's walks, but it is. Pitchers in hitters parks tend to nibble, the air at altitude reduces drag and makes breaking balls move less, etc.)

Add it all up, and it's what we expected looking at the dimensions: a slight pitchers' park. Nothing like RFK, but far from a bandbox. Stan probably had it right--it's fair.

However before we can conclude for sure how Nationals Park plays, we'll need to wait another couple years and allow the data to even out. The biggest thing I look for going forward is what will happen if/when the Nationals finally get a left-handed power bat or two. That short-ish porch in right could cause Nationals Park to creep into the slight hitters' park category. Our total lack of left-handed power (no, Willie Harris doesn't count) is a major variable that god willing will change soon.


Mike said...

I think as long as it's no longer RFK, and also not Citizen's or Great American, they'll be fine. It doesn't appear that Nationals park is going to trend toward either of the extremes. And, while I guess it depends on how the rest of the skyline shapes up, I can't see any wind vortexes being created as we go forward.

The two biggest examples of "surprise" park effects are Petco and the Ballpark at Arlington.

Petco was, just from the dimensions, thought to play as a hitters park. The wind and other climatological factors spoiled that (as well as Brian Giles later years).

The Ballpark at Arlington was supposed to be somewhat neutral, but the construction of the stadium apparently collects the wind, swirls it, and shoots it from home plate into center field giving the field a sort of launching effect. I'd have to dig for the article, but I remember a beat writer remarking that, as he stood on home plate before a game, it felt like a 30mph wind was constantly blowing on him from the backstop.

Steven said...

I agree with all that. I prefer a slight pitchers' park because I think the park effect is more important to the development of young pitchers than hitters.

Your comments about Petco and The Ballpark are interesting. I didn't know that.