OOTP Developments Forums - View Single Post

Garlon · 06-29-2008, 05:13 PM

1962 was the first season in Dodger stadium. I use 5 seasons worth of data to calculate the park factors. The previous 2 seasons, the next 2 seasons, ad the focus season which is counted 4x. In this case there are no "previous 2 seasons" since the park just opened. So 1962 is counted 4x in the calculation, plus 1963 and 1964 data. The data takes into account everything that happens in dodger stadium vs everything that happened in dodger road games, for both them and their opponents.

For Dodger Stadium that's:
AB: 33651
HR: 493
HR/AB: .01465

For Dodger road games that's:
AB: 33663
HR: 915
HR/AB: .02718

So now we just compare the HR per AB of Dodger Stadium vs the HR per AB of the road games. We get:

.01465/.02718 = .5389, which rounds to .539

For doubles and triples we don't use per AB, we use per non-HR Hit. 2B/(H-HR).

We don't have split data, so the factors generated are applied to both LHB and RHB.

The size of the park and walls are not used in creating the park factors. The real data of what actually happened over the course of 5 seasons is what matters. Remember, park factors are relative to the other parks in the league, they are not absolute. Wrigley Field hasn't really changed since 1938, but in the 1970's it was a much more HR-friendly park compared to the rest of the NL than it is today because now we have Coors Field, Minute Maid, and other even more HR-friendly parks in the NL than Wrigley.

Since we are comparing home vs road data it makes no difference how many total HRs your team hits and allows or your opponents because you are playing with and against the same teams when you go on the road as well. So we end up looking at the rate at which HRs occur at home compared to the road. We use 5 seasons worth of data, with the focus weighted at double. After 1962 when both leagues began using 162-game schedules that comes out to looking at a sample of 648 games at home and 648 road games each season to generate the park factors. This eliminates alot of random noise that could crop up if you were only looking at 81 games at home vs road.

06-29-2008, 05:13 PM	#11 (permalink)
Garlon All Star Starter Join Date: Jun 2004 Posts: 1,551 Thanks: 0 Thanked 52x in 26 posts	1962 was the first season in Dodger stadium. I use 5 seasons worth of data to calculate the park factors. The previous 2 seasons, the next 2 seasons, ad the focus season which is counted 4x. In this case there are no "previous 2 seasons" since the park just opened. So 1962 is counted 4x in the calculation, plus 1963 and 1964 data. The data takes into account everything that happens in dodger stadium vs everything that happened in dodger road games, for both them and their opponents. For Dodger Stadium that's: AB: 33651 HR: 493 HR/AB: .01465 For Dodger road games that's: AB: 33663 HR: 915 HR/AB: .02718 So now we just compare the HR per AB of Dodger Stadium vs the HR per AB of the road games. We get: .01465/.02718 = .5389, which rounds to .539 For doubles and triples we don't use per AB, we use per non-HR Hit. 2B/(H-HR). We don't have split data, so the factors generated are applied to both LHB and RHB. The size of the park and walls are not used in creating the park factors. The real data of what actually happened over the course of 5 seasons is what matters. Remember, park factors are relative to the other parks in the league, they are not absolute. Wrigley Field hasn't really changed since 1938, but in the 1970's it was a much more HR-friendly park compared to the rest of the NL than it is today because now we have Coors Field, Minute Maid, and other even more HR-friendly parks in the NL than Wrigley. Since we are comparing home vs road data it makes no difference how many total HRs your team hits and allows or your opponents because you are playing with and against the same teams when you go on the road as well. So we end up looking at the rate at which HRs occur at home compared to the road. We use 5 seasons worth of data, with the focus weighted at double. After 1962 when both leagues began using 162-game schedules that comes out to looking at a sample of 648 games at home and 648 road games each season to generate the park factors. This eliminates alot of random noise that could crop up if you were only looking at 81 games at home vs road. Last edited by Garlon; 06-29-2008 at 05:15 PM.