NCAA Division III men's basketball efficiency ratings calculated based upon Ken Pomeroy's method.
Below is an attempt at replicating Ken Pomeroy's excellent Division I efficiency ratings for Division III teams. Only D3 vs. D3 games are included in the data set and team records. No Division I, II, NAIA, NCCAA, or other opponents are included.All stats come from NCAA.org. If box score data hasn't been reported to that site, it won't be in this data set yet.
Preseason ratings are generated with a MARCEL-like projection using each team's last three seasons of data plus some regression toward Division III average. This will be very inaccurate for teams with lots of roster turnover, but it is simple and will help stabilize early-season results. The preseason ratings will get less and less weight as the season goes on and will be completely gone by the end of December.
A brief explanation of the ratings and method:
AdjEM - Adjusted Efficiency Margin - This is the number of points (per 100 possessions) by which the team in question would be expected to beat an average team. This is that category by which teams are ranked overall.
AdjO - Adjusted Offense - This is the number of points (per 100 possessions) that the team in question would be expected to score against an average defense.
AdjD - Adjusted Defense - This is the number of points (per 100 possessions) that the team in question would be expected to allow against an average offense.
AdjT - Adjusted Tempo - This is the speed, in possessions per 40 minutes, at which the team in question would be expected to play against an average team.
SOS Win50 - This column represents the median rating of all off the opponents a team has played so far this season. This strength of schedule indicator is less susceptible to swings caused by a small handful of particularly poor (or strong) opponents.
OppEM/O/D - Opponents - These columns represent the average adjusted efficiency margin, offensive rating, and defensive rating of a team's opponents. These give an idea of a team's strength of schedule.
NCSOS OppEM - Non-conference - This column is the average adjusted efficiency margin of a team's non-conference opponents.
Adj - Adjusted - Raw offensive, defensive, and tempo ratings are calculated for each game using box score data. These numbers go through an automatic iterative process until they stabilize in an attempt to account for game tempo and quality of opposition. For example: if Calvin plays a "system" team like Grinnell in a game that results in 95 possessions, they shouldn't necessarily be seen as a fast team (simply because they played a fast opponent). Their adjusted game tempo would be a more NCAA-average 75 (or so) since 95 possessions per game is how fast average teams play against Grinnell. The same process applies to offensive and defensive efficiency ratings. Each game number is adjusted to account for quality of opposition (a 'good' offensive output against a lowly defense might end up looking only average). The number next to each column is the team's overall rank in that category.
'Update' Column - The rightmost column (you might have to scroll) indicates which teams had data missing from the NCAA stats website at the time of the last update run. UPDATE means a game is missing from a previous day (it might only be an NAIA opponent that wouldn't count anyway). TODAY means the team in question had a game scheduled for the day of the last update that isn't yet included in the data set. The date and time of the last update can be seen at the top left of the embeded spreadsheet.
Conference Ratings - Scroll below the team list for ratings of conferences. Conference ratings have been generated by averaging the metrics of member teams.
Score Predictions - Also, below the conference ratings, please find DAILY SCORE PREDICTIONS or predict a custom matchup. These use the adjusted figures to predict a score based on each team's tempo, offense, and defense.
Previous Seasons
2017-2018
2018-2019
2019-2020
2021-2022
Link back to this page if you can. It helps! Also, please feel no obligation to give but you can help support my data-collection endeavors by donating using the button below.
Feel free to reach out to Matt Snyder @FFTMAG for feedback or if you notice an error.
Check out the D3 Datacast on YouTube using this link!
Predict Your Own Matchups!
Use the dropdown menus to select a home and away team (results will also predict a neutral site game). (This might not be working right now!)Daily Score Predictions
These are daily Division III men's basketball score and pace predictions. The main scores, margin, and totals include a standard 3.5 point home field advantage. For neutral site games look at the right-hand columns for score and margin predictions.Games including NAIA or Division II (etc.) opponents have 'no data' listed and do not include a prediction.
