surely the fairest way for comparisons sake, working off the fact that there is a drop in the number of teams, is to take a mean for the previous year based on ten teams.
In other words, take the top team and bottom teams averages out of the original equation and work off the remaining 10 teams, giving an average of 1400.
The remaining ten teams the following season attendance averages drop to 1400 meaning no real loss in attendance. And, of course, if attendances stabilise there is an actual increase.
To me, that would make more sense than the simplistic formula you have suggested Stu.
edit: (but, of course, im no Carol Vorderman either!!)