Ranking Non-Conference Strength of Schedule (Part 1)

The college basketball season just days away tipping off, and this season I wanted to add more to this blog than just my weekly updated bracketology projection. Throughout the season I plan to a post “Games of the Week” article, where I will pick a game or two from each night of the week that has important implications for NCAA Tournament selection and/or seeding. Once late January comes around, I will join the other hundreds of bracketologists and post a semi-weekly Bubble Watch article.

But in anticipation of the start of the college basketball season, I wanted to write a few articles on topics that influence the selection process and seeding of teams in the bracket. The first of those is comparing the non-conference strength of schedule (SOS) for every team in the Power 5 + Big East + American Athletic Conference + the “Best of the Rest” (e.g. Gonzaga). Non-conference SOS is one of the most important variables to consider when comparing teams on the bubble. The committee has historically left teams out with gaudy records but very poor non-conference SOS (see NC State last year). Some other ideas I have for articles to post over the coming weeks include identifying the most likely 1 seeds for this year’s tournament, an initial list of teams’ status relative to being selected to the NCAA tournament (i.e. a Preseason projection with Bubble Watch flair), and ranking the holiday tournaments and their potential impact to the NCAA Tournament bracket.

Let’s begin with ranking the non-conference SOS. With this post I intend to offer a table ranking all 92 teams in this analysis and a graph to help visualize. A second article will be posted later in which I will profile the top 10 schedules and give a conference breakdown.

First, the methodology. I used KenPom’s preseason rankings to determine the quality of opponent. For the holiday tournaments, only the first game for each team is given, therefore I projected the additional round (or two, in some cases) by assuming the higher rated KenPom team wins each matchup. This introduces some bias into he results since upsets happen routinely in college basketball, but I wanted to get a feel for the strength of the field in those holiday tournaments. Finally, I ranked all 92 teams based on the average preseason KenPom ranking of their opponents. In addition, I noted for each game which quadrant it falls into. Recall that the selection committee uses a quadrant system to better visualize the quality of schedule and team’s record against relative quality of opponents. Games in Quad 1 are extremely valuable, as wins in this quadrant are like gold to the committee, whereas games in Quad 4 for power conference teams are “cannot lose” type games, games where the power conference team could place their best IM basketball team on the court and the game should be competitive. The committee does not use KenPom's rankings, but their rankings by the end of the season will be similar enough to the KenPom rankings that I felt comfortable adding the quadrant system into this analysis.

Below is a table of all 92 teams in this analysis and a graph, sorted by average KenPom rank of opponents. Stay tuned for more in-depth analysis on these rankings.

Legend: navy=SEC, Orange=AAC, Red=Big 12, Black=Big Ten, Green=ACC, Yellow=Pac-12, Gray=Big East, Purple=Other