You might think I haven't been writing because I'm gloomy. And that's partly true. Two Georgetown losses, an Illinois loss, and the Vikings getting eliminated from the playoffs will do that to you.
But the real reason I was away was lack of internet access. I'm back, but with plenty of coverage of the weekend action on Basketball Prospectus, let me just point out my favorite story of the weekend. No, I'm not talking about North Carolina going down. I'm talking about Oregon St. winning a Pac-10 conference game. Hey, no one deserves to go winless, and it was nice to see the Oregon St. fans finally get something to cheer for. And in that link Kevin Pelton covers the equally exciting story of how California not only beat Arizona this week, but also Arizona St. It is amazing what a difference a coach can make to a college basketball team and Mike Montgomery is having an impact in his first year back in the Bay Area.
Sample Size, Bench Minutes
But before we jump on or off the Cal bandwagon, or any bandwagon for that matter, I'm amazed how much faith people are putting in the early season results. People seem to assume that the Pomeroy Rankings or Sagarin Ratings have already determined a national champion. I particularly like this post by Marquette Blog Cracked Sidewalks where the Marquette bloggers sadly admit the team will have trouble doing better than 9-9 in Big East play.
But hold on a minute folks? Are we really ready to anoint the rankings as being fixed at this point of the season? There's still plenty of time for teams to improve or fall apart. And more importantly, do you really want to draw strong conclusions when most BCS teams have played only 2-5 BCS competitors at this point?
Certainly margin of victory or points per possession differential against cupcake teams can be meaningful. (I'm pretty sure Indiana is not very good.) But it can also be misleading because of how teams use non-conference games. Is the coach expirementing and giving major minutes to untested players, or did the coach narrow a rotation early because of injuries or lack of depth. Having watched Illinois, I know there were times when they had a big lead against a small team, but blew it after bringing in a bunch of young players who didn't know how to run the offense. Alternatively, Georgetown has had so few bench players that the subs have had only limited opportunities to lower the team's overall efficiency.
I suppose if this hypothesis is true, I should point to teams that might get better once they narrow the rotation. A key stat here might be Pomeroy's Bench Minutes. (But you have to be careful here to avoid the full-court pressure teams that always use long rotations like Minnesota and Missouri.) I think a good case in point is a team like Miami (FL). I think the Hurricanes may ultimately be much better than their Pomeroy Ranking of 26 indicates, once they settle on a shorter rotation. Miami has tried to mix in younger players like DeQuan Jones and Eddie Rios but with limited success. But once Frank Haith takes these players out of the full-time rotation, or focuses on these players strengths and defines a key role for each player, Miami can still compete to be a top level ACC team. Jack McClinton, Dwayne Collins, and James Dew are too good for Miami not to step it up.
(A few more notes on Miami: Everyone knows about McClinton, and James Dews was a solid back-court mate last year, but Junior Dwayne Collins has really emerged as a post option this year. Last year Collins, was solely a rebounder, but his emergence as an offensive post player, eFG% of 61.3% is huge for the Hurricanes.)
Remember Mississippi St?
Last year Mississippi St. ended up with the second best SEC PPP differential. But no one would have expected it after they lost to Miami (OH) and South Alabama in non-conference play and started the year 5-5. (To be fair, their pre-conf Pomeroy ranking was not horrible thanks to some blowout wins, but it wasn't very good either.)
And I guess that brings me back to my initial point about Marquette. Already, the Pomeroy Rankings have improved for the team. No longer predicted to go 9-9, the prediction is now 11-7 in conference play. And it will change again by next week.
[[Side note #12: Fabulous win for Marquette against Villanova. Yes, they probably should have won that game at home, but given how evenly matched they are on paper and on the floor, it was a very sweet victory. Villanova seemed to commit too many unforced turnovers in the game (traveling outside the three point line), while Marquette's turnovers were all in the course of trying to get the ball to the basket, but those two teams always make for a fabulous match-up.]]
But back to the prediction. Yes, Marquette probably does still face the toughtest final 5 regular season games of any team in the country, but the reality is that there is still a lot to learn about teams before then. Conference play is here! The time has come to sort out the contenders and pretenders, not only in the ranks of the unbeatens, but also in the margin of victory statistics.