Ratings Explanation + Glossary

Ratings

We have baseball ratings for NCAA Division I, Division II, and Division III from 2012-present and for NAIA from 2018-present though only 2022 ratings are currently loaded on the site. Score data is obtained from https://ncaa.com and from https://masseyratings.com. Current season scores are gathered from both sources and discrepancies are reconciled, often by viewing individual team schedules to confirm score, site, and/or date. Game data used in the ratings are score and site of each contest with only same-division games included. The following factors were tested in each division over all past seasons to find the best predictive combination:

  • How deep into the season to keep preseason rating (currently previous year-end figure) as part of the formula
    • Testing ranged from 0 to 60 games
  • Include full margin of victory or institute cap
    • Testing included no cap and range of 1 to 20
  • Value of home field advantage
    • Tested ranged from 0 to 1 runs at 4 decimal places
  • Should preseason rating be only previous year’s figure or include data from earlier seasons
    • Multiple linear regression was used to find the best combination of prior year(s) and tested against preseason rating

Current year ratings are determined by using the best combination of factors from the list above, which varies by division.

Glossary

  • SOS (Strength of Schedule)
    • The value for our SOS is based on the expected winning percentage of an average team against that schedule
    • A lower value means a tougher schedule and vice versa
  • SOR (Strength of Record)
    • This is a plus or minus value showing how the team has fared against their schedule vs the expected record of an average team
    • A positive value means the team won more often than an average team would have and vice versa
    • This levels the SOS and illustrates which teams have performed best against their schedule regardless of difficulty
  • +/- Exp Wins
    • This is a plus or minus value comparing the team's actual record vs the expected record according to their current rating
    • A positive values means the team has won more often than their record would indicate and vice versa
    • Some rating systems and sites call this "luck" but we look at it simply as over/underachieving your expected record