Thursday, January 2, 2014

RPI/SoS outliers

Now that it's January, we can start to look at the RPI and SoS numbers and not be deceived completely by them.  The numbers will still change rapidly, but everyone now has a non-con base around which the numbers are formed.

Let's see which RPIs and SoSs are much higher or lower than you think, and if any team should celebrate or panic about it.  This isn't a comprehensive list by any means, but should highlight points of interest:

UC Santa Barbara (RPI 27, SoS 26) - They went 7-4 in the non-con against a schedule that features only one team expected to be below RPI 200.  They went 2-3 against teams ranked 51-100 in the RPI right now, and have played 6 road games along with 2 neutrals.  All of this adds up to a computer profile that's better than the actual team.  This will dip a bit, so I don't expect any at-large hopes, but the seeding will look good on Selection Sunday if they get there.

North Dakota St (RPI 28, SoS 38) - A similar situation to UCSB.  Avoidance of bad teams plus 8 road games.  The bad news is that their quality win at Notre Dame is evaporating into little more than a paperweight.  The hidden downside to getting the surprise signature road win is that many times, the signature fades.  The home loss at Southern Miss is costly, as well as at North Dakota, even though it's a rivalry game.  The RPI and SoS won't hold.

UMass (RPI 2, SoS 5) - The RPI isn't much a surprise, given the great 11-1 record, but a top 5 SoS?  And the A-10 will hurt it, but not THAT much.  How did they manage it?  Again, great cupcake avoidance - the only really bad team is Northern Illinois.  (they still do have Miami(OH) and Elon on the non-con slate to come).  But the MAC scheduling agreement actually helped them this year - they got quality road games at Ohio and EMU - and got them both.  Neutral wins over New Mexico and Clemson.  They do lack the signature win (New Mexico and BYU have faded, LSU isn't there yet, they lost to Florida St), but that's ok.  They have too many good wins.  Now the question is how high a seed they can get.  Games against VCU and St Louis will determine the ceiling, while the other easier games determine the floor.

Indiana (RPI 89, SoS 135) - And a non-con SoS hovering around 200.  This is a dire situation.  No quality wins.  They'll have plenty of chances, but don't be fooled on name alone.  I do think they get the quality wins eventually, but it'll take work.  I do think 9-9 in the B1G would be enough.

Michigan (RPI 75, SoS 78) - Here's a B1G outlier that's more surprising.  However, the numbers will jump up because of the strength of the B1G (same for Indiana, by the way), and they do have neutral court wins over Stanford and Florida St in their hip pocket.  They also have a couple of anchors in the non-con SoS that B1G play will mostly erase.  They'll be fine.

Teams with better non-con SoSs than you think:
BYU (3) is going downhill fast, but if they ever correct themselves, they will have profile support.
New Mexico St (49) is making sure WAC play won't be a death knell to their seed.
South Carolina (19) is only .500, but could be a darkhorse in the weak SEC given their tough start.

Problematic non-con SoSs:
Wichita St (108) will be fine as long as they keep winning, but their opponents are struggling.
Louisville (102) has enough of a hard time already, but this combined with the AAC makes is really, really tough to get a top 2 seed.
Cincinnati (158) has the same AAC problem, and has a weak SoS for a presumed bubble team.
SMU (209) really, really has a problem.
Ohio St (83), Duke (75), and Arizona (73) don't have bad SoSs, per se, but the competition on the 1 line will be fierce.  Kansas, Wisconsin, Syracuse, Villanova, Kentucky, and UNC are in the top 30, and that's not an insignificant advantage to have.  It is worth noting all 3 are projected to gain strength in the SoS, and the numbers will improve, but the position relative to other top teams won't.
Need evidence the Valley is failing?  Indiana St (240) and Missouri St (252).

No comments: