[Orcin]

After reading the threads linked above, I am more convinced than ever that ballparks in PT should not be customizable except for name (and other cosmetic items) and have park factors of 1.000 locked in.

Originally Posted by Questdog

Park factors not totaling to a 1.000 average across the board is not a problem as far as skewing the overall stats of a league since OOTP sets the LTMs with the park factors in place to get the overall league stats in line with what they should be.

The problem with the factors not averaging to 1.000 is that you cannot know exactly what the actual affect of the factors are by just looking at the settings.

For instance, if the park factors of one category in a league average out to say 1.500, then a park with a 1.000 setting is actually below average in that league and not neutral. A park factor setting of 1.200 would actually be below average and so when you look at the home/road stats for that category the team should perform better on the road.

If all the factors average to 1.000 then you'd know to expect a team with a 1.200 factor to perform better at home.

Originally Posted by Charlie Hough

The way that OOTP handles park factors is based on real life park factors and how they are calculated. That's why they have to total 1.000 if you don't want deviation from adherence to the league totals that you have set. In theory, if you adjust one and it offsets the balance, you are supposed to adjust the others.

Park factors also play a huge role in historical OOTP games, particularly if you're not using neutralized stats. If you're not using neutralized stats, then using park factors will effectively double the impact of your ballparks. This is because, theoretically, the player stats for that historical season were already influenced by the ballparks, and if you add park factors on top of the player's ratings, which were determined by his real stats, then you're potentially going to skew results. So, the guy who hit 50 home runs in a hitter's ballpark that season might hit 65 because the park factors are double-influencing his performance. This is why people who don't play with neutralized stats are advised to set each park's factors to exactly 1.000 for every category in historical games. But this is problematic when you consider that players can change teams in historical games, and they can take on a life of their own. So that's why some people recommend playing with neutralized stats, so players can move around and their performances can be influenced with relative ballpark realism.

However, this real-life basis is also why there has been so much confusion over the park factors in OOTP over the years. As I pointed out, most people think of park factors intuitively and assume that they should reflect a park's real, long-term impact on statistical results and not merely the single-season statistical output within that park relative to the output among all other parks in an MLB season.

So, this is why I have brought up the real life calculations, to help explain how and why OOTP handles park factors the way that it does. And I also bring it up because I think the real-life park factor methodology is flawed, and we need a better approach. At the very least, I don't think it's advisable for people to ensure that their park factors total 1.000 if they are not trying to strictly run an MLB season and base things on exact league totals. For example, let's say that you've created a league based on the MLB, but you want half your ballparks to be much friendler to pitchers. Well, you're not going to see any major change in results if all your ballpark factors add up to 1.000. Instead, they should add up to less than 1.000, which will realistically reflect the potential impact of half the ballparks suddenly being pitcher-friendly parks.