Those park effects appear to correlate with the data. Unfortunately, there is a lot of statistical noise that can cause these types of fluctuations (particularly when you further limit the sample sizes to left-handed and right-handed batters). Of course you can't rule out other things at play, particularly the weather, as well as alterations to the park or even composition of the team and pitching staff.
It does make me wonder if this is the right way to approach things. It appears that Strat uses single year park factors. Should they be using 3-year park factors in order to stabilize the data? I reckon that might be a better approach.
That said, I do appreciate the flavor that park effects bring to the game (I'm a Replay guy).
Seamheads is the place to go for detailed park breakdowns:
Those park effects appear to correlate with the data. Unfortunately, there is a lot of statistical noise that can cause these types of fluctuations (particularly when you further limit the sample sizes to left-handed and right-handed batters). Of course you can't rule out other things at play, particularly the weather, as well as alterations to the park or even composition of the team and pitching staff.
It does make me wonder if this is the right way to approach things. It appears that Strat uses single year park factors. Should they be using 3-year park factors in order to stabilize the data? I reckon that might be a better approach.
That said, I do appreciate the flavor that park effects bring to the game (I'm a Replay guy).
Seamheads is the place to go for detailed park breakdowns:
https://www.seamheads.com/ballparks/ballpark.php?parkID=BAL11&tab=pf1
As you can see, the data is a lot less "crazy" when applying 3-year park factors:
https://www.seamheads.com/ballparks/ballpark.php?parkID=BAL11&tab=pf3