Comparing ACC Team Computer Rankings With the Preseason: Pitt Up, BC Down…Posted by Brad Jenkins (@bradjenk) on December 12th, 2013
With approximately 30 percent of the regular season already played, it’s a good time to check out how ACC teams are currently rated by some of the most highly regarded computer rating systems. It’s also a good time to compare that with each team’s preseason expectations, where we find that there are definitely some surprises.
The chart below lists the 15 ACC schools according to their current computer rankings. This ranking is based on an average of three of the most respected basketball computer gurus – Ken Pomeroy, Jeff Sagarin, and Kenneth Massey. Keep in mind that these ratings are updated daily and this table represents data from Monday, December 9. The first column is the average national ranking for ACC teams in these computer systems. The next two columns compare the conference ranking for each team using the computers versus the official ACC Media Preseason Poll.
With the unpredictability of college basketball it’s not surprising that several teams are quite a bit above or below expectations so far this year. The first team on the list has the third biggest variance between current computer ranking and ACC Preseason Media ranking (five spots). The computers love Pittsburgh this year. Pomeroy has the Panthers fourth in the nation despite a less than challenging schedule. Basically it means that Pitt has been beating bad teams convincingly. It is still reasonable to project Jamie Dixon’s team higher now than at least two of the teams that were ranked ahead of them in the preseason poll — namely Virginia and Notre Dame. It’s worth noting that the Panthers were also something of a computer darling last season. At a similar point in the 2012-13 schedule, Pitt was seventh in Pomeroy’s ratings and eventually ended the season at #11, despite entering the NCAA Tournament as a #8 seed and losing to Wichita State badly in its opening round game.
The other two big variances belong to Clemson and Boston College, each team at opposite ends of the spectrum. Clemson is an astounding eight spots higher than its predicted finish of next to last. On the other hand, Boston College is the worst team in the ACC according to the computers, which is seven spots worse than their preseason ranking. The explanation for both of these extremes lies with the defensive performances of each of these teams. Clemson’s defense is rated sixth in the nation according to Pomeroy while the Eagles are currently a pathetic 268th. But before we accept the predictive value of the current computer situation for these teams, consider this fact — last year at this time Pomeroy ranked Clemson #64 and Boston College #178. Sound familiar? The end result was that the Tigers ended up 11th in the ACC standings and dropped to #124 in Pomeroy, while the Eagles finished eight in the league and improved to #96. There’s a lot of basketball left to be played.
This leads to two very important truths. First of all, a sample size of eight to 10 games is very small and tenuous from which to draw conclusions — these integrated computer systems are not as accurate now as they will be later. Secondly, teams develop at different rates and progress or regress over the course of a long season. Still it’s interesting to see how teams are performing with the data at hand and compared with how the media thought they would before the season began.
Note that we are not including the RPI ratings in our analysis. The NCAA Selection Committee may use it in March to rank-order, select and seed the field, but we are not going to use it to try and judge teams now. As for why, consider this example: Even this early in the season, should any legitimate rating system have a 3-3 UC Santa Barbara team at #50, 10 spots higher than a 7-2 Duke squad? On a final note, all three computer ratings used here have the Big Ten as the top overall conference, with the ACC averaging third place.