College basketball rankings: Conference results trickle in

The Friday Facts were originally developed to help evaluate conference performance during this era of mega-leagues and unbalanced schedules. In the Big Ten, for instance, each team plays a home-and-home with only five of the other 13 teams. That leaves plenty of room for in-conference strength of schedule to affect the conference standings. And even in leagues with true round-robins, the standings at any given time can be deceptive based on who's played whom, and where.

For many years, John Gasaway has attempted to cut through the noise by publishing his "Tuesday Truths," which track raw efficiency margins for the Big Ten and other conferences. Last year, I developed the Friday Facts to take this analysis a step further, by adjusting the raw efficiency margins for strength of opponents and locations of games. For example, raw efficiency margin credits a one-point win over Rutgers at home as being better than a one-point loss to Wisconsin on the road, when in fact the opposite is true. The Friday Facts fix that.

This year, I can take things even one step further, because with the T-Rank power ratings I can actually track adjusted offensive and defensive efficiency for every team in conference games only.

Two weeks into the conference season, we've got enough data to make the in-conference Friday Facts worth taking a look at:

Allow Me to Explain

To break this down, the "Raw EM" is the raw efficiency margin: point differential divided by possessions (aka, the Tuesday Truths).

The "EM+/G" column is a calculation of adjusted efficiency margin -- specifically, how each team has performed, on a per-possession basis, compared to how an average D1 would be expected to perform against that team's conference schedule. This is how the Friday Facts were calculated last year. Astute readers may notice that the number is bigger for every team this year; that's because last year the adjusted EM was based on how an average Big Ten team would be expected to perform, and this year it's based on the average D1 team. Since Big Ten teams are better than the average D1 team (ahem -- usually, Rutgers), the adjusted EM figure is bigger now.

The adjusted offense and adjusted defense numbers are each team's adjusted efficiency numbers in Big Ten games only. These are used to calculate a Pythagorean Expectation, which for Friday Facts-branding purposes I call the "Barthag." This more or less tracks the adjusted EM column, but in some cases the order will differ, and in those cases the Barthag trumps. Those are this year's Friday Facts.

Finally, there is a projected conference record for each team. That's calculated from the team's total-season T-Rank (noted in the first column), not the conference-only results.

So... we have a bit of a surprise in that Michigan State is the current Friday Facts leader even though Wisconsin is the undisputed, sole leader of the Big Ten at 3-0. How could this be? Well, MSU has two impressive wins -- romps over a decent Indiana squad at home and a good Iowa team on the road -- and its only loss was a two-point, double-overtime affair against No. 26 Maryland. By contrast, Wisconsin has played one of the easiest schedules so far (Penn State, at Northwestern, Purdue) and although the Badgers have performed fine, the margins haven't knocked anyone's socks off. In particular, you can see that Wisconsin's adjusted defensive efficiency in conference games is a decidedly mediocre 96.3, good for 11th in conference play.

The other, perhaps surprising, result is that Minnesota comes out middle-of-the-pack despite being mired at 0-3. This is rather easy to explain, as Minnesota has played the toughest conference schedule so far (at Purdue, at Maryland, Ohio State) and has lost by an average of just six points per game (including an OT loss to Ohio State).

This Week's Big Ten Games

This week's slate is headlined by some big home games for the league's tourney hopefuls, including Minnesota, which will be desperate for a win against Iowa.