"KenPom is bull***t" Sit back an enjoy this four minute dialogue from Doug Gottlieb on the KenPom rankings. (mentions St. Thomas, Minnesota)

Donate To GopherHole

For context...I have heard lots of coaches say the same thing...especially smaller schools.
And they just lost lost to the number one team in the league by 8 and as a coach he wants respect for his team. Green Bay is number 4 in the league after being ridiculed at the bottom last year.
 

I tend to agree, but…. the KenPom rankings do have some validity. If, however, you rely on them too much when making your March Madness brackets, you will make your friends and co-workers happy.
 

Some of what he’s saying, doesn’t account for injuries or recency, etc, is a known issue with any ranking model. And I think the NCAA refuses to consider those factors.

My biggest issue with Kenpom is that it doesn’t limit blowouts, and as a result, teams are rewarded for blasting terrible teams. By my math, us whomping Alcorn St early on currently moves us up about 8 spots. If our entire noncon was sub-300 SWAC/NEC type teams and we strafed them all, despite never playing a competent team, we’d enter B10 play in KenPom’s top 20. Obviously a flaw. But because Ken is trying to rank offensive and defensive efficiency and then reverse engineer an overall ranking, he has no choice.
 



I didn't watch this because I wasn't interested but I would like to throw a compliment to UW-Green Bay for how their season has progressed. They were 1-4 after losing that November overtime game to the Gophers but they've gone 11-8 since.
 
Last edited:

Go4POV said:
I tend to agree, but…. the KenPom rankings do have some validity. If, however, you rely on them too much when making your March Madness brackets, you will make your friends and co-workers happy.
Click to expand...
Actually I’d say the opposite. People bet too much on the underdogs and the fun splash upset (they undoubtedly will occur but it’s too hard to predict which ones that will be). If you just let KenPom operate you would probably have won the March Madness pool last year.
 


The Deliberate Stranger said:
I agree with Doug and FWIW the committee doesn't take kenpom into account, at least there not supposed to
Click to expand...
Au contraire, mon frere. See below for an excerpt from the official link NCAA selection criteria

Men’s Tournament​

Key Metrics Used:

  • NET Rankings (NCAA Evaluation Tool).
  • Quadrant Wins and Losses (more details below).
  • Results-based metrics (Wins Above Bubble, KPI and Strength of Record): These metrics help evaluate a team’s resume. Wins Above Bubble, a new metric for the 2024-25 season, shows how many more, or fewer, wins a team has against its schedule versus what a bubble team would expect to have against the same schedule.
  • Predictive-based metrics (Torvik, BPI and KenPom rankings): These metrics help evaluate the quality of teams. Torvik was added to the team sheet list for the 2024-25 season.
 



I decided to watch the clip, and it wasn't what I thought it'd be. He was trying to make an argument that the Summit League representative will deserve better than a 16 seed (especially if the regular-season champ loses in their conference tourney, i.e. if GB happens to win). One of his more valid points is that the predictive metrics don't take into account who's healthy/available for every game when evaluating a team's overall performance - this is very true.

However, the way I know KenPom isn't BS is that if you look at his predicted point spreads and the point spreads that Vegas puts out, you will see extreme correlation.
 

Ken Pom is good as one tool


But efficiency isn’t the most important stat
Win-loss is
Efficiency is predictive of win loss…but when it isn’t Ken pom calls it luck

I call it knowing how to win close games
 

FritoPendejo said:
Some of what he’s saying, doesn’t account for injuries or recency, etc, is a known issue with any ranking model. And I think the NCAA refuses to consider those factors.

My biggest issue with Kenpom is that it doesn’t limit blowouts, and as a result, teams are rewarded for blasting terrible teams. By my math, us whomping Alcorn St early on currently moves us up about 8 spots. If our entire noncon was sub-300 SWAC/NEC type teams and we strafed them all, despite never playing a competent team, we’d enter B10 play in KenPom’s top 20. Obviously a flaw. But because Ken is trying to rank offensive and defensive efficiency and then reverse engineer an overall ranking, he has no choice.
Click to expand...
There’s definitely a little of that, but in the flipside, if you ONLY beat a team like Alcorn State by 20, you’re getting docked. It’s also not going to help your O efficiency if you boat race the 350th defensive team (I don’t know what they are, probably something like that).

I think with AI, these algorithms are going improve significantly, and begin factoring injuries and more nuance.
 

You must log in or register to reply here.



Top Bottom