Justin of On Baseball and the Reds adds a nother part to his player value series with an improvement in park factors. He helps negate the disparity between good hitters and weak hitters that comes from ratio park factors by using regressed 5-year park factors to make an additive form. Essentially, how many runs a park affects a typical player per PA, and then adding or subtracting that from a player. Not only does it make sense, it's also extremely easy, and he puts it in an ultra-nifty table.
Also, I did not realize Arizona was the best non-Coors park for hitters.