Bear Rankings

These rankings currently use several sport-specific parameters that are subject to change:

  1. The home field advantage parameter. Every so often I re-optimize the algorithm for this parameter, but it doesn’t change a whole lot.
  2. The margin of victory parameter. This parameter is based on data from the last 10 years in each sport; I try to determine what’s a reasonably expected margin of victory, and what’s considered running up the score. Using this data is a better indicator than trying to nail down that number subjectively.

The teams are ranked by their WVARA score. I provide the projected win-loss ratio for the remaining games if the season is currently on-going.

The two remaining fields are the SW (strength of wins) and the SO (strength of opponents). I struggled for a while to find a good metric to express a strength of schedule – the current systems just don’t cut it for me. I finally realized there are two important components to a good strength of schedule – how hard it is (SW), and how good it is (SO). Both metrics are scored from 0-1 with 1 being ideal. It usually turns out that the teams at the top of the pile have really low SW and SO; not sure what to make of that yet.

  1. SW is a measure of how hard it would be for a team to go undefeated against their current schedule. This doesn’t take into account whether they actually are undefeated or not.
  2. SO is a measure of how good a team’s schedule is, that is, how good are the teams they play during the season? This is based on the average WVARA rank index of their opponents.