Well, scratch the resolution idea from the previous post, I found a much better way to create frictional ball parks. Earlier this evening I had wanted to find real life parks that seemed to result in a Baseball-Reference-esque type number.
It was a futile exercise. There were wild swings in DMB park data - a 120 / 100 / 75 / 49 park might add up to Neutral. I didn't want to use something like this for all 1800s batters and pitchers in a neutral park.
Instead, I believe I have devised a method to auto calculate any generic park factor I wish:
- The Historical league average player hits .257 / .331 / .375
- In 550 AB lets estimate he hit 20 doubles, 3 triples, and 12 home runs
- Reverse engineering we get 133 Runs Created
- We know RC is a fairly good estimator of teams runs scored
- We know the BR park factor equates to the percentage of runs scored above or below average. Ex: 104 = 4% more runs scored.
- Therefore, we can surmise that the average batter would increase their runs created by 4% as well.
- Knowing all the above, we can back into a 4% Runs Created increase by monkeying with a batters singles, doubles, triples, and home runs.
- The percentages we change those stats, are the percentages we change the DMB park factors.
- Whoila! Below is the chart I will use
If for some reason you are still reading this post, the left hand column is the BR Park Factor. The next four are the park factors I will use for DMB purposes. And the rest are the stat lines of the average player in each of those parks.
The best part (or worst) my original way of doing it was probably close enough anyway. Testing out a 110 BR BPF the old way (110 across the board) yields 10% more Runs Created. Mother of all that is good and holy.
No comments:
Post a Comment