On one hand, this ranking is a current ranking of a world. The most important inputs are the most recent data. If a world was a great world but went to hell in one season, it would take awhile for the ranking system to react. Generally, this isnt the case. A world isnt going to all of the sudden become top-heavy, or have a huge shift in the average fielding stats or era. Im sure there are exceptions, though.
If I do include prior season's data--it will be weighted much less. I already have identified one impovement I could make with my input for parity that should take care of the few cases where I think a world was penalized too harshly or not harshly enough based on my initial criteria. I think this improvement will reduce the benefit of looking at multiple years (meaning I may not go back to previous seasons--but implement previous seasons as I move forward).