MUScoop

MUScoop => Hangin' at the Al => Topic started by: KipsBayEagle on February 10, 2019, 05:25:29 PM

Title: Kenpom Rating System
Post by: KipsBayEagle on February 10, 2019, 05:25:29 PM
https://kenpom.com

We currently sit at 29, the school in front of us is Wofford.  Why are the Kenpom ratings still a thing? 
Title: Re: Kenpom Rating System
Post by: Pakuni on February 10, 2019, 05:28:26 PM
https://kenpom.com

We currently sit at 29, the school in front of us is Wofford.  Why are the Kenpom ratings still a thing?

Wofford actually isn't too bad.
That 14-10 Texas, with losses to Radford, Okie State and Georgia, is sitting in the top 25 is a travesty.
Title: Re: Kenpom Rating System
Post by: brewcity77 on February 10, 2019, 05:29:57 PM
Wofford actually isn't too bad.
That 14-10 Texas, with losses to Radford, Okie State and Georgia, is sitting in the top 25 is a travesty.

You mean that 14-10 Texas, with wins over North Carolina, Purdue, Kansas, and K-State? They just have a weird resume.
Title: Re: Kenpom Rating System
Post by: MU82 on February 10, 2019, 05:59:59 PM
kenpomnomatta
Title: Re: Kenpom Rating System
Post by: TAMU, Knower of Ball on February 10, 2019, 08:05:04 PM
https://kenpom.com

We currently sit at 29, the school in front of us is Wofford.  Why are the Kenpom ratings still a thing?

IMHO KenPom is the most accurate system. Not perfect but no system is. Wofford is a very solid team this season though probably a little overranked. Their only four losses this season are North Carolina (home but might as well be on the road) and Oklahoma/Kansas/Mississippi State all on the road.

Do you have a suggestion for a better system than KenPom?
Title: Re: Kenpom Rating System
Post by: wadesworld on February 10, 2019, 10:23:47 PM
My biggest question about Kenpom is are the rankings updating today based on results from our games in November? For example, Kenpom had us beating Nova by 2, we won by 1, so there’s very little movement for MU based on that result.

What about our game with Louisville? Louisville is coming off a not great season and going through a coaching change, and there’s very little to base them off of in November. I don’t know what Kenpom had it going into the season, but I’d guess he had MU about a 5 point favorite. We won in OT so we get dinged because we didn’t beat Louisville by as much as his algorithm predicted, but since then, in the 3 months of basketball played between then and now, Louisville has proven to be much better than what they were believed to be. But we played them in November so to Kenpom we underachieved?

And also when it’s looked at for NCAA Tournament purposes, it should no longer be about predictive measures and should be based on actual results. The selection process should be about rewarding what had been done up until that point, not predicting what an algorithm would suggest should happen moving forward. You earn a postseason berth/seed based on what you actually achieved over the course of a season.
Title: Re: Kenpom Rating System
Post by: MU82 on February 10, 2019, 10:26:18 PM
Is KenPom more "accurate" than the media and coaches (SIDs) polls?

Maybe so. I don't know. Just askin'.
Title: Re: Kenpom Rating System
Post by: MUDPT on February 10, 2019, 10:44:19 PM
Is KenPom more "accurate" than the media and coaches (SIDs) polls?

Maybe so. I don't know. Just askin'.

Picking a bracket with KenPom’s rankings beats a chalk bracket basically every year.
Title: Re: Kenpom Rating System
Post by: TAMU, Knower of Ball on February 10, 2019, 10:52:19 PM
My biggest question about Kenpom is are the rankings updating today based on results from our games in November? For example, Kenpom had us beating Nova by 2, we won by 1, so there’s very little movement for MU based on that result.

What about our game with Louisville? Louisville is coming off a not great season and going through a coaching change, and there’s very little to base them off of in November. I don’t know what Kenpom had it going into the season, but I’d guess he had MU about a 5 point favorite. We won in OT so we get dinged because we didn’t beat Louisville by as much as his algorithm predicted, but since then, in the 3 months of basketball played between then and now, Louisville has proven to be much better than what they were believed to be. But we played them in November so to Kenpom we underachieved?

And also when it’s looked at for NCAA Tournament purposes, it should no longer be about predictive measures and should be based on actual results. The selection process should be about rewarding what had been done up until that point, not predicting what an algorithm would suggest should happen moving forward. You earn a postseason berth/seed based on what you actually achieved over the course of a season.

Yes, the "prediction" gets updated as the season goes on. Our win over Louisville as only gotten better.