Variance between RPI and Pomroy rating

[NersEllenson]

Pretty amazing to see some of the variance between RPI and Ken Pomroy's ratings.  From what I've followed this year, Pomroy's seem to be very accurate, and his game score and win/loss predictions are eerily correct.  Question:  If filling out a bracket, do you put more stock in Ken Pomroy than RPI rating?

Saw on ESPN last night William and Mary's resume against Miss State and Florida..and both Gotleib and Jay Williams chose William and Mary over Florida on the Blind Resume thing they do.  Their RPI is 57, yet KenPom has them at 113.

MU is 27 on Pomroy, 50 on Real Time RPI...

[oldwarrior81]

RPI takes into account only a Win or a Loss while ignoring the margin of victory.

Pomeroy and Sagarin both factor in the margin and generally are a more accurate indicator of a team's performance.

[Nukem2]

RPI takes into account only a Win or a Loss while ignoring the margin of victory.

Pomeroy and Sagarin both factor in the margin and generally are a more accurate indicator of a team's performance.

Agreed. Plus, the RPI has some weird math relative to homw vs. road losses.

[NersEllenson]

RPI takes into account only a Win or a Loss while ignoring the margin of victory.

Pomeroy and Sagarin both factor in the margin and generally are a more accurate indicator of a team's performance.

Agreed. Plus, the RPI has some weird math relative to homw vs. road losses.

Thanks for the feedback - I'm going to use Pomroy Rankings to help fill out my bracket..though that does mean I have to pick UW pretty far, as he has them 3! 

[oldwarrior81]

here's a link to Sagarin's.  UW is only a 14 if that helps.

http://www.kiva.net/~jsagarin/sports/cbsend.htm

[NersEllenson]

Digger just picked UW over MU - surprised?  No.  Anyway, onto computer rankings:

Pomroy has MU: 28, Washington 29
Saragin has MU: 24, Washington 42

of Note New Mexico:
Pomroy: 47
Saragin: 23

[Nukem2]

Obviopusly, the committee's thought process is more in tune with KPom/Sagarin than the useless RPI.