Week 3 of the Bilas Index brings some movement, with more to come as we enter conference play. As previously noted, the Bilas Index is not some random poll that has wild swings based solely upon a snapshot of this week's play. Rather, it is a combination of consistently reliable data and metrics (such as Kenpom.com ratings and Sagarin rankings) as well as the incredibly reliable eyes and mind of the Bilastrator himself. We can complain about "the eye test" all we want, but it is important to subtly adjust the positions of certain teams that benefit from or get hammered by the dreaded numbers of the pocket-protected, pointy-headed, pencil pushers who provide us with such revealing data.
However, I do not ignore the numbers in the Bilas Index and simply use the transitive property, either. As we get into conference play, there will be more data available for every team, and the Index can begin to "net out" data from meaningless games. By "meaningless games," I mean games played against the dregs of Division I, around the dreaded bottom third. The truth is, I don't really care how a team performs against bad teams. It is of no use to me in evaluating a quality basketball team. But, right now, that is most of what we have to go on. That is changing rapidly.
Against a bad team, good teams may not be motivated to bring their best and oftentimes do not. By the same token, a classic underdog may look average against another underdog in a fair fight. But against a Goliath, it may play well above that level. I am interested in how the Davids play when truly challenged. If they can hang in truly big games and have a chance to win, that scores points with me. After all, the only teams that play against bad teams in the NCAA tournament are the No. 1 and No. 2 seeds. Everyone else has to play somebody good.
Xavier still presents a quandary. What to do with this compelling team, when the performances of late make you want to drop them precipitously? Last week, I kept Xavier in place, but this week, I adjusted the Musketeers down. It's not the free fall their numbers and performance suggest should take place, but it's also not a free pass, which I gave them last week.
It is still early, which is why some of the rises and falls are so extreme. Once there is more data, the numbers will not change so much week to week, and the rankings will stabilize. There are still a few spots where I think a team is better than a few teams ranked ahead, but the numbers are holding them back. Later, I believe that will change.
Out: Kent State, Dayton, Denver and LSU
In: Middle Tennessee, Oral Roberts and Washington