NowRanked #5 in the KenPom ratings
- Thread starter Fansince72
- Start date
We moved up 3 in kenpom from Thursday... And above Kansas
I think the reemergence of Cline and with Haarms playing the way he is, we are looking great!
Sets of three are the most luck --- agreed and add DEFENSE
By the way, kudos to Cline with the 2 blocked shots yesterday!
Moved up 6 in RPI to 15
Tennessee @ 6
W Kentucky @70
2-1 against top 50
4-1 against 50-100
Although RPI is set to be have less influence in seeding... I’ll believ it when I see it
Tennessee @ 6
W Kentucky @70
2-1 against top 50
4-1 against 50-100
Although RPI is set to be have less influence in seeding... I’ll believ it when I see it
is this the year they stop using RPI to determine seeding or no?
Not sure how things like BPI and Kenpom etc aren't used with more weight
But at this point as long as we take care of business the rest of the way, there is no reason we cant be atleast a 3 seed with an outside chance at a 2 if we win the games we are supposed to win and make the big ten championship game
way to far to look ahead but after the 2 losses in a row this is a far turnaround thats for sure
Not sure how things like BPI and Kenpom etc aren't used with more weight
But at this point as long as we take care of business the rest of the way, there is no reason we cant be atleast a 3 seed with an outside chance at a 2 if we win the games we are supposed to win and make the big ten championship game
way to far to look ahead but after the 2 losses in a row this is a far turnaround thats for sure
Appears there will be some changes this season to the team data sheets the committee uses for selection and seeding. Some tweaks grouping wins and losses into 4 quadrants, giving more weight to road games. But is still using RPI as a basis.
It does appear KenPom and other predictive metrics may be making their debut on the team sheets based on the link below. I'm not sure if any of this has been set in stone. And of course, old habits die hard, so not sure how much consideration the "new" metrics are going to be given. Baby steps..
http://www.ncaa.com/news/basketball...ion-committee-adjusts-team-sheets-emphasizing
So far it looks like the best place to find the resumes according to the new quadrant system is the NCAA itself. Its daily "Nitty Gritty" reports now reflect the new setup. You can find the reports here (note: more than mens bball is listed so pick the correct sport): https://extra.ncaa.org/solutions/rpi/SitePages/Home.aspx
As of today Villanova (6-0), Duke (5-1), and Purdue (5-2) are the only teams with at least 5 combined wins in quadrants 1 & 2.
Purdue is now up to fifth in the composite rankings:
https://www.mratings.com/cb/compare.htm
Purdue fairs the worst in the human polls.
https://www.mratings.com/cb/compare.htm
Purdue fairs the worst in the human polls.
looks like arizona st was 100th in kenpom preseason. and preseason rank does not completely fade from his calculation until mid/late january. so still in consideration, kinda like what fans hate about human polls.
I don't have an answer, but Houston depended on metrics the most of any team in baseball ... and they're World Champions.
At least KenPom backs himself with data. His blog, it goes into great detail on how he developed his weighting of the preseason ratings and how he arrived at January 23rd as the date at which the preseason weighting tapers off over years of data, with evidence that even 10 or 15 games into the season the sample size is too small to make accurate predictions.
With that said, the fact that we are #5 and started #17, shows how little effect the preseason rating has at this point in time. ASU just simply has not played a very difficult schedule and is dreadful on defense.
ya I enjoy the ratings way more than polling. but don't think any 1 is perfect. as you point out, pu has jumped 12 spots whereas asu has gone 70+ but barely sits in kenpom top 30
also interesting when a 2 loss Cincinnati is ranked 10 (compared to undefeated asu at 29), and cincy's SOS is more than twice as bad.
Last edited:
Similar threads
