RPI Update – 5

State stopped the RPI slide yesterday with a big road victory at Clemson. Since the RPI now gives more credit for a road victory, all road victories (especially conference ones) qualify as “big”. However, State’s graph shows that it is easier to move down than it is to move up once you are competing for ranking against the top teams (especially with a weak schedule):

State’s RPI vs time

However, it looks like things are really volatile once you move down towards #50. Much to my surprise, GW stayed in the top-50 with a home win over #88 Rhode Island. GW has a big road game coming up Thursday night against Xavier (#50). A win Thursday should give GW a good boost up from their current spot at #43.

Seton Hall moved from 80+ into the top-50 with two huge road victories against NC State and Syracuse. Thus State’s record vs the top-50 currently stands at 2-4 with all of State’s losses coming against top-50 opponents.

Clemson’s home loss gave them a big hit in RPI ranking, knocking them out of the top-50. I think that GW, SH, and Clemson’s moves this week all point to how quickly things change once you move into the “bubble-range”.

Conference Update
Based on historical trending of RPI:





RPI Rank





Boston College




North Carolina St


North Carolina









Wake Forest



Florida St


Georgia Tech


Virginia Tech


UNC is currently sitting near the start of the bubble. Over the seven years worth of data that I looked at, every BCS school ranked 37 or higher received an at-large bid…so I put them in the tourney.

Approximately 90% of the bubble teams that received a bid had an RPI ranking of 36-55. So while UVa, Clemson, and UM, are technically on the bubble, they are at the far, outside edge. All three teams need to step up and play better to get an at-large bid. The overall strength of the ACC will likely work against Bubble Teams this year.

Several more RPI graphs that might be of interest:

The Wake Forest Collapse Continues

Virginia Digs out of a Hole

Clemson’s RPI

The ACC was ranked second to the Big 10 for the last several weeks. When I looked this morning, the ACC had dropped into third place. One can only surmise that Duke’s, Maryland’s, and State’s recent OOC losses had a big part in the conference ranking slide:

Top 10 Conferences

About VaWolf82

Engineer living in Central Va. and senior curmudgeon amongst SFN authors One wife, two kids, one dog, four vehicles on insurance, and four phones on cell plan...looking forward to empty nest status. Graduated 1982

General NCS Basketball Stat of the Day

13 Responses to RPI Update – 5

  1. choppack 01/30/2006 at 3:59 PM #

    Good stuff – I think your graph would be accurate if the tourney were held tomorrow. However, I’d be shocked if BC got a higher seed than us. We have a better record in conference and have won at their place quite convincingly.

  2. class of '74 01/30/2006 at 4:15 PM #

    It is amazing Maryland is as high as they are given some bad losses they’ve had.

  3. Rick 01/30/2006 at 4:15 PM #

    If UNC is anywhere near the bubble they get in on name alone.
    BTW, does anyone else find it absolutely incredible that a coach can return one player (that averageed 4 points), lose his top 7 players and still have a good team?
    We suck when one player goes down for a week or two.

  4. Clarksa 01/30/2006 at 4:23 PM #

    “If UNC is anywhere near the bubble they get in on name alone.”

    I think UNC-CH gets in with a 7-9 ACC record based on name alone.

  5. TVP 01/30/2006 at 4:29 PM #

    Sweet! Great stuff, thanks for putting this up.

  6. BJD95 01/30/2006 at 4:53 PM #

    7-9 plus name pls beating KY and AZ (even in down years) = bid.

  7. VaWolf82 01/30/2006 at 5:05 PM #

    Just don’t forget that most people (including me) had Maryland in last year with their two wins over Duke. 3-7 down the stretch moved the Terps into the NIT.

  8. JeremyHyatt 01/31/2006 at 3:08 AM #

    how the crap is GW ranked 10?? they don’t have a big win and lost to us. oh please have them play in our bracket.

  9. VaWolf82 01/31/2006 at 8:12 AM #

    Are you talking about RPI or the polls?

    If you are talking about polls it’s simple. They’ve only lost one game so far…and they dropped a few slots when that happened. Then since they haven’t lost since the State game, they move up every time someone above them loses.

    I rarely look at the polls any more. They have absolutely no meaning or impact on the season. Take a look at today’s entry on NCAAT seeding. Even a high RPI doesn’t guarantee a high seed in the NCAAT.

  10. site admin 01/31/2006 at 8:31 AM #


    That is exactly why I don’t pay a bit of attention to the subjective polls. But, be careful…this works both ways.

    There are programs who have been ranked in the Top 25 despite horrible records against Top 25 & Top 50 RPI competition that the case could be made isn’t really a Top 25 team.

  11. choppack 01/31/2006 at 5:04 PM #

    So quick question – since the rankings and seedings are based on a combination of a “book of work” and how well a team has played recently – which is better? A team that has had some impressive wins, but more embarassing losses or a team that has had fewer “embarassing losses”, but less impressive wins.

    The one thing State has had going for it is that it has had fewer, recent embarassing losses most years. This isn’t to say that a loss to St. John’s isn’t embarassing – but for a bubble team, it’s probably not out of the question.

  12. VaWolf82 01/31/2006 at 6:30 PM #

    which is better?

    Better for what? For RPI, it doesn’t matter. Winning percentage (adjusted), opponent’s winning percentage and opponent’s opponent’s winning percentage is all that matters.

    For seeding, I doubt that you will see a team get a high seed without some impressive wins.

    I think seedings are more heavily influenced by the play at the end of the season than for clearing the bubble. As we’ve seen with State in 2003 and 2005, 5-5 or 6-4 down the stretch is still good enough for an at-large bid. In 2005, Kansas had the #1 RPI ranking, but got a 3 seed….probably because they finished the year at 3-5.

  13. choppack 01/31/2006 at 9:33 PM #

    I agree w/ you point on the seedings…and I’d add that if we don’t beat Duke again this season, and go 11-5 in the conference, we’d probably end up anywhere from a 3 seed to a 5 seed.

Leave a Reply