Conference Tournament Results and the NCAA Tournament

On the other hand, basketball schedules, even schedules that would be considered good, have a lot of fluff, and only about 10-15 games for a team with a decent schedule are against tournament-level competition. For a team that advances deep in a conference tournament, they may increase the number of quality wins substantially. I thought I would take a look to see if the committee is properly weighing the effects of the conference tournaments.

I looked at the big six conferences since 2003: Atlantic Coast, Big East, Big Ten, Big Twelve, Pac-10, and Southeastern Conference. Using the same method I used in this post, I then compared how teams did relative to their seed expectation. I’m trying to adjust for expectations based on where a team is seeded in the NCAA tournament; we would expect that better teams would tend to advance further in conference tournaments, and would also tend to be seeded higher in the NCAA tournament. Here are the results, sorted by how far teams advanced in conference tournaments:

————————————————————————————

What I don’t see there is anything that would indicate that the committee overvalues the effect of conference tournament results systematically by moving a team up too far when it performs well in a tournament. Sure, some teams that made deep runs in a conference tournament and shot up their seeding level by 1 or 2 lines did flame out early, but that’s certainly not the rule.

In fact, the tournament teams from major conferences that bowed out at the quarterfinal stage (or earlier) have underperformed in the last eight years. Some specifics:

Teams that reached a conference championship game and were seeded #1 or #2 in the tournament reached the Sweet Sixteen 83% of the time (30 of 36), compared to 61% (8 of 13) for #1 and #2 seeds that lost in their tournament quarterfinals.
Among lower seeded major conference teams (seeded between 7 and 12), those that advanced to a conference semifinal or beyond won 61% of their first round matchups (25-16 record), while those that lost by the quarterfinals went 20-30 (40%) in the first round. The lower seeds that reached a semifinal were three times more likely than their early exit brethren to break through to the Sweet Sixteen over the last eight years. [note: I checked the seed distribution and the groups were nearly identical in composition]
Among the nineteen teams that won 2 or more games over average for their seed level, only 3 of them lost in the quarterfinals or earlier in conference tournament action–Michigan State in both 2005 and 2010 reaching the Final Four as a 5 seed, and Alabama in 2004 going to the Elite Eight as an 8 seed. In contrast, 7 teams that won a conference tournament title fell in this group, including 6 of the 8 national champions during this span.

Based on this, I would say that any belief that the selection committee improperly overvalues conference tournament play in its seeding process is unsupported. To the extent they are moving teams up or down in seeding based on getting quality wins on neutral courts or taking bad losses in the week before the NCAA tournament begins, it would appear to be justified by the results overall. What happens today and tomorrow does matter, and can be used by the committee to differentiate teams that are roughly similar.

[photo via Getty]