forums

KenPom calls out the media after Kentucky's win over Miami

Advertisement
Pretty sad.

He built a great model that, in many ways, revolutionized CBB/sports gambling, and even the way people view the game. He realized there was something in the conventional wisdom that wasn't quite right.

Now Kenpom itself has become conventional wisdom, and he acts like it can't be improved upon. He's become exactly what he initially set out to fight against.

Other great analytics guys, like Bill Connelly, talk about their models' shortcomings all the time. You never hear that from Pomeroy.
 
Advertisement
hotshot said:
Those two numbers aren't mutually exclusive. There's a lot of good fortune (aka luck) involved in getting to the Final Four.
Click to expand...
There is definitely luck involved in March, but even as of Selection Sunday last year, our resume metrics were much stronger than our predictive metrics. That was also the case in 2022, and is the case so far this year.

At some point - and idk when that point is - we have to consider whether the metrics are missing something about this team.
 
hotshot said:
Those two numbers aren't mutually exclusive. There's a lot of good fortune (aka luck) involved in getting to the Final Four.
Click to expand...
I understand that, but we sat in the 30s for the majority of the season while winning ACC regular season, and having a strong year, then ended up making the final 4.

so somewhere there is a disconnect
 
gacaneaddict said:
I understand that, but we sat in the 30s for the majority of the season while winning ACC regular season, and having a strong year, then ended up making the final 4.

so somewhere there is a disconnect
Click to expand...
I think with one of UConn's Championship teams (mighta been last year even), they experienced this.
 
Advertisement
ALLoverCANE said:
There is definitely luck involved in March, but even as of Selection Sunday last year, our resume metrics were much stronger than our predictive metrics. That was also the case in 2022, and is the case so far this year.

At some point - and idk when that point is - we have to consider whether the metrics are missing something about this team.
Click to expand...

Exactly. The disparity between our strength of record ranking and kenpom ranking the last two years is the big problem in my opinion. And that’s created by ranking exclusively on what amounts to opponent adjusted margin of victory/defeat and completely ignoring wins and losses (a team that demolishes bad teams but loses close games to every good team will have a higher kenpom ranking than a team with a much better record and similar SOS that beats both good and bad teams by small margins).

Obviously our SOR was more predictive than our efficiency metrics for the last two postseasons.
 
hotshot said:
I think with one of UConn's Championship teams (mighta been last year even), they experienced this.
Click to expand...


Meanwhile they have won 24 straight OOC games by double-digits (going for 25 against Kansas tonight). This is the longest streak since the 1950s!

www.canesinsight.com

Around College Basketball - (Current topic: UCONN / Hurley 22 straight double digit OOC wins)

Let's kick it off with Bobby Huggins Note: This thread is dedicated to news and events taking place Around College Basketball as a whole.
www.canesinsight.com www.canesinsight.com

www.espn.com

UConn beats UNH by 20 to claim NCAA record

Led by a career-high 29 points from Donovan Clingan, No. 4 UConn cruised to an 84-64 win over New Hampshire on Monday -- setting an NCAA record in the process.
www.espn.com www.espn.com
 
EightiesCane said:
Did anyone else see the kenpom tweet?

Click to expand...



Guy is a ******* moron. We played Kentucky nearly even for the first half. We had one bad second half on the level of last year's game at GaTech or the Elite 8 game against Kansas two years ago.

But, yeah, this nutjob is going to act like 20 bad minutes is the "proof" or "validation" of his ****** model.
 
We're the only team to go to 2 consecutive Elite 8s the past 2 years. What does that have to do with the media?
 
Advertisement
AmherstCane said:
Exactly. The disparity between our strength of record ranking and kenpom ranking the last two years is the big problem in my opinion. And that’s created by ranking exclusively on what amounts to opponent adjusted margin of victory/defeat and completely ignoring wins and losses (a team that demolishes bad teams but loses close games to every good team will have a higher kenpom ranking than a team with a much better record and similar SOS that beats both good and bad teams by small margins).

Obviously our SOR was more predictive than our efficiency metrics for the last two postseasons.
Click to expand...
I just wouldn't go that far. In all sports, margin of victory/defeat matters. That's pretty conclusive at this point.

But I think there are some teams that are exceptions - for whatever reason - and we may be one of them.
 
ALLoverCANE said:
I just wouldn't go that far. In all sports, margin of victory/defeat matters. That's pretty conclusive at this point.

But I think there are some teams that are exceptions - for whatever reason - and we may be one of them.
Click to expand...

Of course it matters but it is not the only thing that matters. What’s strange is that he could easily get rid of the “luck” ranking (that is the most obnoxiously arrogant and awful metric) and incorporate an adjustment for winning against expectations and he could meaningfully improve his system.
 
AmherstCane said:
Of course it matters but it is not the only thing that matters. What’s strange is that he could easily get rid of the “luck” ranking (that is the most obnoxiously arrogant and awful metric) and incorporate an adjustment for winning against expectations and he could meaningfully improve his system.
Click to expand...
Maybe he can, maybe he can't.

Some things are not quantifiable
 
Advertisement
ALLoverCANE said:
Maybe he can, maybe he can't.

Some things are not quantifiable
Click to expand...
He can. His ranking system does not account for who wins or who loses. Winning by 1 is basically the same as losing by 1 in his system. He could weight for actual wins… very easy to do it but he believes wins and losses don’t matter because that’s the entire basis of his ranking system.
 
Dude is a clown.

So I guess TN is mid, since they got blown out at UNC…and Duke is for sure mid, since they lost to an unranked Arkansas squad.

Fact is, Mr. Pom-Pom’s computer metrics have consistently hated on Miami teams. I wonder why that is?

Reminds me of the old BCS computer poll that mysteriously screwed the Canes over in 2000 and kept us out of the title game, even though we had beaten FSU earlier that season, and our only loss was at a very highly ranked Washington team with an extremely young Ken Dorsey at the beginning of the season, and we still almost beat them.
 
By the way, this clown has UNC ranked 10th, and based on his “system” they’re mid because they’re barely going to beat FSU at home today.

Duke also has looked like buns, and he has them ranked 7th.

Miami is ranked 44th in his latest poll. Guy is a major hater.
 
Advertisement
Back
Top