The ESPN park factors are totally screwed up, seeming to randomly recalculate themselves when you re-load the page at times. FWIW, it's now shwing AZ as a 1.138 for runs, in line with the past.
Park factors do show fluctuation from year to year which can be the result of normal randomness like all stats---ballplayers have a true talent level, but they don't perform the exact same way every night, month, or year--as well as other factors. Park factors can be impacted by weather, wind, time of day (check out Philly's day versus night stats), modifications in the field, and other things.
Most stat guys use park factors averaged over a three year period to try to minimize the impact of the normal flux in the stat.
You're exactly right about the ESPN park factors being totally screwed up. I figured that out recently when I was going back-and-forth between Runs, HR, etc. and the numbers kept changing. Down at the bottom of the screen, I clicked on "report a bug" and told them the problem I was having and they seem to have fixed it. I just went there and the factors look like they should now, with Cincy being 3rd-best in runs scored, rather than 27th, which is what I got earlier. If the park factors are now correct, one interesting note is that the Phillie's park has a higher PF than Coors Field, which is a bit of a shock considering how Coors Field has played over the years. BronXBombers51 wrote
I'm like Rynman. I have no idea how the parks are taken into account with some of these stats. How is it calculated?
It appears that different places factor them slightly differently, depending on how much math work they want to do.
A basic way to do this is to take a team, say Boston.
For this example, Boston scores 25 runs in 5 home games and 20 runs in 5 road games.
Park Factor (Boston)=(runs at home/games)/(runs on the road/games)
What that says is that the Red Sox score 25% more runs at home than on the road.
A more advanced way would be to use runs per at-bats or runs per plate appearances, rather than games. If the Red Sox win all 5 of those hypothetical games, they might have fewer at-bats or plate appearances because they could go 5 games without batting in the 9th. It would be more accurate to use runs per plate appearances.
I don't know what ESPN uses.
Also, as mentioned above, it's generally a good idea to use 3-years of park factors for accuracy because within a season, you get some small sample size factors.
On the ESPN Park Factors page, they have Runs, HR, Hits, 2B, 3B, and BB. All of these occurences can have different park factors.
At this link:
http://www.baseballthinkfactory.org/btf ... ffects.htm
The guy takes his park factors into even more detail because, if you think about it, the Red Sox are playing a lot more games in Yankee Stadium, Camden Yards, Skydome, and Tropicana Field than other teams, so the writer of the linked article tells how he breaks down his park factors to take into consideration the different visiting parks that a team plays in.
I hope that's not too complicated. The basic idea is that a neutral park is 1.00. Anything above that is a hitter's park, below that is a pitcher's park.
U.S. Cellular Field (White Sox) HR park factor--1.568 (56.8% more homers are hit at The Cell than in a neutral park).
Citizen's Bank Park (Phillies) HR park factor--1.544 (54.4% more homers hit at CBP.
RFK Stadium (Nationals) HR park factor--0.558
So, big OUCH if you have any hitters on the Nationals. Although, I would expect they'll change the dimensions of that park prior to next season after the recent revelation that the dimentions are mismarked.
Roger Angell: I was talking with Bob Gibson and I said: 'Are you always this competitive?' He said: 'Oh, I think so. I got a three-year old daughter, and I've played about 500 games of tic-tac-toe with her and she hasn't beat me yet.'