First step in the process is to determine regression scores for each team. Next, schedule strength is calculated as the average of the regression scores of the team's opponents, with a correction for home-court advantage.
Games involving CIML teams versus non-CIML teams are the sole basis for ranking CIML's overall conference strength.
Within the CIML generally top teams have a easier schedule.
The following is seemingly is a bit of a paradox. If a team wins by 30 and is expected to win by 10, the rating of the opponent and the winner's schedule strength both decrease. In reality this means that the model had underestimated the winner's strength and overestimated the loser's strength.