College Sports (Basketball)

[Fly Illini Fly!]

Right, 63rd in points allowed per game as an assistant coach. One hell of a reach dude.

That's because they played very fast. More possessions = more points.

On KenPom they were the #1 defense (not 63rd):

.....

Points scored / given up is useless because it doesn't account for pace.

Illinois was the #2 offense, but was 28th in points per game (due to playing very slow).

What matters is efficiency -- points per possession.

.....

You seem to have racked up quite the track record lately.

-- Coleman is our "indisputable" PG; not Vaaks

-- Musselman over Brad

-- and now Michigan the 63rd best defense??

 

[Fly Illini Fly!]

Obligatory "not to dog pile", because it's easy to argue against Michigan having the best defense in history last year;

But I was curious about actual stats for where Michigan, the number #1 defense last year, was ranked next to other #1 defenses, so I took the last 10 years as a sample set, and used Kenpom's Drtg. It turns out that Michigan was the lowest-rated #1 defense of the last 10 years:

View attachment 50762

It seems like the only thing truly remarkable about their defense is that they were the first team since 2013 to be the team with the #1 defense that won the Championship. Otherwise, they were just a physically imposing team, which is why they also had a highly rated offense.

I think the raw points per possession might explain this.

2026: Michigan kenpom #1 defense | #8 in raw ppp

2025: Houston kenpom #1 defense | #2 in raw ppp

2024: ISU kenpom #1 defense | #2 in raw ppp

2023: Tennessee kenpom #1 defense | #2 in raw ppp

2022: TTU kenpom #1 defense | #2 in raw ppp

2021: Memphis kenpom #1 defense | #1 in raw ppp

So, in terms of raw points per possession surrendered, Michigan was #8 this year (few cupcakes in there: Marist, UC Irvine), while many of the others were #1 or #2.

It's just that they played the #2 offensive strength of schedule, and that strength of opponent puts them over the hump as the top defense.

 

[AirTran]

Obligatory "not to dog pile", because it's easy to argue against Michigan having the best defense in history last year;

But I was curious about actual stats for where Michigan, the number #1 defense last year, was ranked next to other #1 defenses, so I took the last 10 years as a sample set, and used Kenpom's Drtg. It turns out that Michigan was the lowest-rated #1 defense of the last 10 years:

View attachment 50762

It seems like the only thing truly remarkable about their defense is that they were the first team since 2013 to be the team with the #1 defense that won the Championship. Otherwise, they were just a physically imposing teaIm, which is why they also had a highly rated offense.

I'm confused - Michigan the first team to have the top KenPom defense rating to win the NCAA championship.

The Kenpom ratings don't actually determine anything and certainly do not determine game winning Pct. and tournament advancement rate.

Michigan ranked 7th in Team Defensive Efficiency according to Team Ratings
Michigan ranked 14th in Team Defense according to Warren Nolan
Michigan's KenPom Defensive Rating for the 2025 - 2026 regular season? That was 106.7 or 182nd in the nation.

 

[wc_illini]

I'm confused - Michigan the first team to have the top KenPom defense rating to win the NCAA championship.

The Kenpom ratings don't actually determine anything and certainly do not determine game winning Pct. and tournament advancement rate.

Michigan ranked 7th in Team Defensive Efficiency according to Team Ratings
Michigan ranked 14th in Team Defense according to Warren Nolan
Michigan's KenPom Defensive Rating for the 2025 - 2026 regular season? That was 106.7 or 182nd in the nation.

You might want to double check these.

They were 106.7 in KenPom defense in 2024, not 2026 (when they were 88.5)

TeamRankings' defensive efficiency is not adjusted for opponent.

Not sure where you got the WarrenNolan ranking from. They were #8 in defensive rating, but that also isn't adjusted for opponent.

 

[AirTran]

You might want to double check these.

They were 106.7 in KenPom defense in 2024, not 2026 (when they were 88.5)

TeamRankings' defensive efficiency is not adjusted for opponent.

Not sure where you got the WarrenNolan ranking from. They were #8 in defensive rating, but that also isn't adjusted for opponent.

You are correct about the 2024 KenPom D- rating. I misread - my bad

But I think you are missing the point. KenPom is the most "bet against" rating in sports. Meaning - it is the least predictive statistical indicator in the history of sports. Having the best KenPom rating in anything - cannot be quantitatively substantiated. And your initial statement proves it - only once in 10 years has a KenPom top rated defensive rating won an NCAA title.

9 of the the KenPom overall top 16 teams made the sweet 16 in 2026 - and his top 16 teams failed to cover the point spread in 52% of their games.

I think you get my point. KenPom doesn't have the best teams at the top of its rating list. There is statistical bias to the ranking and a subjective

If you are trying to pick the National Champion - these stats will work better than KenPom

 

[March2TheArch]

https://twitter.com/x/status/2073783867207483691

 

[lstewart53x3]

You are correct about the 2024 KenPom D- rating. I misread - my bad

But I think you are missing the point. KenPom is the most "bet against" rating in sports. Meaning - it is the least predictive statistical indicator in the history of sports. Having the best KenPom rating in anything - cannot be quantitatively substantiated. And your initial statement proves it - only once in 10 years has a KenPom top rated defensive rating won an NCAA title.

9 of the the KenPom overall top 16 teams made the sweet 16 in 2026 - and his top 16 teams failed to cover the point spread in 52% of their games.

I think you get my point. KenPom doesn't have the best teams at the top of its rating list. There is statistical bias to the ranking and a subjective

If you are trying to pick the National Champion - these stats will work better than KenPom
View attachment 50815

This has nothing to do with the original posters point btw. The original posters point was that Michigan did not have a historically good defense last year and was simply using KenPom as a data point to prove that.

 

[the juiceman cometh]

https://twitter.com/x/status/2073783867207483691

I think he's going to be good but let's be real here. This is U-17 international ball, meaning he's playing against high school players (and even then some of the best high schoolers would be aged out) from counties with a much shallower talent pool. Likely this level of competition is below high level AAU ball if we're being honest.

 

[jamarvelous31]

I think he's going to be good but let's be real here. This is U-17 international ball, meaning he's playing against high school players (and even then some of the best high schoolers would be aged out) from counties with a much shallower talent pool. Likely this level of competition is below high level AAU ball if we're being honest.

Yeah, completely fair on the talent.

I think this is one of many data points at this stage though. Pretty sure he's been impressive back to when he was 15 playing with KJ. Averaging 16 and 7 for barca u & 17 and 7 at eurocamp also wasn't bad.

That said, I trust Scheyer to underperform with him.

 

[Fly Illini Fly!]

Michigan ranked 7th in Team Defensive Efficiency according to Team Ratings
Michigan ranked 14th in Team Defense according to Warren Nolan
Michigan's KenPom Defensive Rating for the 2025 - 2026 regular season? That was 106.7 or 182nd in the nation.

These are completely wrong.

 

[wc_illini]

You are correct about the 2024 KenPom D- rating. I misread - my bad

But I think you are missing the point. KenPom is the most "bet against" rating in sports. Meaning - it is the least predictive statistical indicator in the history of sports. Having the best KenPom rating in anything - cannot be quantitatively substantiated. And your initial statement proves it - only once in 10 years has a KenPom top rated defensive rating won an NCAA title.

9 of the the KenPom overall top 16 teams made the sweet 16 in 2026 - and his top 16 teams failed to cover the point spread in 52% of their games.

I think you get my point. KenPom doesn't have the best teams at the top of its rating list. There is statistical bias to the ranking and a subjective

If you are trying to pick the National Champion - these stats will work better than KenPom
View attachment 50815

I've seen you make similar claims a few times, and each time I've provided links refuting your claims and asked for your data/source. I'm still interested if you have it.

That said, 52% ATS isn't profitable after the vig, and one tournament is a tiny sample size. With enough games, computer models generally end up really close to 50%.