The extent to which unbalanced schedules cause distortions in sports league tables

The Australian Football League (AFL) has operated its fixture on the basis of an unbalanced schedule since the league expanded from 12 to 14 teams in 1987. This system contains a number of factors (some random) determining the set of bilateral combinations of teams that play each other on an extra occasion during the course of the season, not least of all maximising attendances. While the status quo may be unavoidable to some extent (it is also a bone of contention to many fans), its implications for within-season measures of competitive balance are nonetheless obvious. This is because of the potential for biases being created in the end-of-season league table as a result of the unbalanced schedule. This paper uses a modified model to correct for this inherent bias over the seasons 1997-2008, and the results are discussed in detail. The model is also generalisable to many unbalanced schedule designs observed in professional sports leagues worldwide.