robertearle
Well-known member
Okay, here's one that would work: Calculate in decimal the proportion of points won against each of the other teams in the conference for games played and then total them. Undefeated, untied season would be 7. It would not matter if you played one game or four against any opponent, the impact of the season series against each opponent would be equal to the impact of the season series for every opponent.
That is the current method used in the Pairwise to determine the winner of the 'Common Opponents' component. They changed to that 'sum of percents' method from an overall winning percent in 2013 (and it cost Wisconsin a trip to the tournament, in favor of North Dakota).
I don't really see how one is better than the other. I guess that most of the time, they'll turn out the same, but - as per the example - every now and then, they won't.
-------------------
Adding, your idea also assumes that there will be at least one game between each pair of teams. What happens if, for example, the other series between Wisconsin and St Thomas also gets cancelled: does Wisconsin get zero because there was no game played?
Last edited: