ESPN’s BPI recognizes how powerful Willie Cauley-Stein and Kentucky have been this season.The college basketball season is about a quarter of the way complete and ESPN’s Basketball Power Index (BPI) has gathered enough information to rate the 351 NCAA Division I teams.

BPI is ESPN’s rating system that accounts for scoring margin, game location, pace, strength of opponent and key players missing. A full breakdown of BPI can be found here, but below are some quick comparisons of BPI to other systems. All of these systems account for outcomes of games, location (home/road/neutral) and quality of opponents, but BPI has at least one feature that differs from each.

Comparison of Prominent College Basketball Rating

SOS beyond opponents’ W-L

RPI is the most notable ratings system and has aided the NCAA’s tournament selection committee for years. It is the simplest of all of the systems and accounts for whom you play, where you play them and who won.

One major drawback of RPI is that it measures strength of schedule purely by opponent win-loss record. As of Dec. 16, Green Bay (5-2 vs D-1 teams) ranks fifth in RPI, largely because it ranks No. 5 in its SOS rankings. Green Bay’s schedule is deemed the fifth-hardest because opponents including Evansville and Florida Gulf Coast had strong records against lower-level competition.

Meanwhile, as of Dec. 16, Green Bay ranks 90th in BPI with the 161st-ranked schedule.

Scoring margin (with diminishing returns for blowouts)

A win is a win, but how a team wins can tell a lot about team strength. Beating Duke by one point is not as impressive as beating the Blue Devils by 40. RPI sees those wins as equal, but BPI and most other ratings systems account for scoring margin.

BPI, however, decreases the value of a blowout; a 30-point win is about 20 percent better than a 15-point win, not twice as good, which is how other methods can value it.

Four of the top five teams in scoring margin – Kentucky, Duke, Ohio State and Notre Dame – rank in the top 5 of BPI.

Pace of game matters

Top Teams in BPI Though Dec. 16

BPI takes score differential one step further by accounting for pace of play. Net efficiency (offensive efficiency - defense efficiency) in each game is captured before adjusting for other factors such as game site, scoring margin, opponent strength and missing key players.

All wins are better than losses (before opponent adjustment)

To capture the value of winning, all wins in BPI receive a game score above 50 on a 0-to-100 scale, and all losses are below 50. Once the opponent and site adjustments are implemented, however, a loss can be more valuable than a win.

Let’s look at an example. Wichita State lost by one point at Utah in overtime Dec. 3. The Shockers had a raw game score of 48 (out of 100), but after adjusting for the fact Utah ranks 11th in BPI and the game was on the road, the Shockers’ adjusted game score rose to 90. Conversely, North Dakota State won by one against Division II’s Minnesota-Crookston and received an adjusted Game Score of 9.

De-weighting games with missing key players

A key differentiating feature of BPI when compared with other systems is that it de-weights games in which key players are missing. Last season, Arizona began the season 21-0 before losing to California in a game where its best player - Brandon Ashley – suffered a season-ending foot injury after two minutes. Should Cal (and future Arizona opponents) receive full credit for beating the No. 1 team in the nation when it was not at full strength? Similarly, should Arizona be fully penalized? BPI accounts for key players based on minutes played and adjusts the importance of games when those players are not [are missing or are not missing.

Ultimately, BPI gives us a tool for rating teams that is more in-depth than other systems. BPI was not designed to be predictive, but in tests, it performed as well if not better than other systems in predicting NCAA tournament games.