UConn’s Offense & CBB Overall

[UCFan1513]

I’ve seen hand-wringing about the offense being 26th in KenPom. Something I haven’t seen discussed is the overall improvement in offensive efficiency this year (and last) in CBB.

In 2021-24 our current offensive rating would’ve been top 5/10. In 2025 it would’ve been top 15. Defensive efficiency ratings during that time have maintained.

1. Why are offenses improving?
   1. Have teams taken what worked for UConn in 23/24 and started doing it better?
2. Is it common for highly rated offenses to regress in analytics after mid-season?

 

[uh-no]

1) the ratings are normalized by year. you can't compare year over year with any rigor.
2) continuation

 

[nelsonmuntz]

Three things that I see:

1) Kids shoot so much better at the youth level than they did even 10 years ago, and I think it is seeping into college efficiency stats. What constitutes a good shot has changed a lot over time because the shooting is so much better. So while 3 point shooting has not changed, it is holding steady against defenses designed to stop outside shooting. They are taking tougher shots and hitting just as many.

2) A return of effective interior and post up basketball. Think Reibe. Still too many low percentage kamikaze drives, but feels like less than there used to be.

3) because defenses are selling out to stop the 3 or defend the rim, the midrange is wide open. Basically undefended.

 

[auror]

Free throw rate and accuracy up, finishing 2pt% is up, three pointer attempt rate is up while making at the same rate, turnover rate down. offensive rebound rate up.

Talent infusions (best players from D2, international pros), continuation, better coaching, bigger players.

 

[UCFan1513]

1) the ratings are normalized by year. you can't compare year over year with any rigor.
2) continuation

Can you explain point #1? If offensive efficiency is points per 100 possessions, would it not be normalized to a common baseline of 100 possessions?

 

[UCFan1513]

View attachment 115629
Free throw rate and accuracy up, finishing 2pt% is up, three pointer attempt rate is up while making at the same rate, turnover rate down. offensive rebound rate up.

Talent infusions (best players from D2, international pros), continuation, better coaching, bigger players.

On your point of better coaching and bigger players - there’s been a stark improvement from even 2024 to this year. I have to imagine that the quality of player hasn’t gone up by that much in 2 years, but I could be wrong.

If it’s better coaching - do you think some other coaches have taken UConn’s secret sauce from 23/24 and improved on it?

 

[navery12]

Can you explain point #1? If offensive efficiency is points per 100 possessions, would it not be normalized to a common baseline of 100 possessions?

The offensive efficiency numbers are adjusted relative to the average D1 team each year, so can't necessarily be compared to previous years. If you assume the average team baseline is similar over the past few years it's still a fine comparison to make though

 

[nelsonmuntz]

View attachment 115629
Free throw rate and accuracy up, finishing 2pt% is up, three pointer attempt rate is up while making at the same rate, turnover rate down. offensive rebound rate up.

Talent infusions (best players from D2, international pros), continuation, better coaching, bigger players.

How can you tell that the 2 point percentage is only for finishes and not for jump shots too?

 

[uh-no]

Can you explain point #1? If offensive efficiency is points per 100 possessions, would it not be normalized to a common baseline of 100 possessions?

as someone else said, it is normalized to 100 possessions and THEN normalized to keep the average team at about 100 combined efficiency, and the standard deviation <something>


so if the overall distribution of the quality of teams (or their offense or defense) is different in a given season from others, then the system will end up normalizing everything differently for that season, making the results incomparable from year to year.

In general, you can only perform these kind of comparisons when the populations are evaluated against each other. But since we can't have, say, 2024 uconn play against 2026 michigan, there is nothing to normalize those two years against eachother and the actual ratings are just guesses. There may be generally observable trends, but things like "highest X since Y" are irrelevant.



All that mumbo jumbo leads to point 2. So is it actually up? Almost surely. Because teams get continuation now.

 

[OceanDoggie]

How can you tell that the 2 point percentage is only for finishes and not for jump shots too?

It's the percentage of 2-pointers made not attempted, or else there would be a column for 2PA%

 

[auror]

How can you tell that the 2 point percentage is only for finishes and not for jump shots too?

I can't, sloppy description.

 

[Iluvsuebird]

Other teams copying what UConn has done so well that lead to two titles? Inconceivable. Listen, Mullins is now getting into a playing time groove. Tarris was also out for a good stretch; even Hurley said that he was getting his legs from in shape. The eyeball test tells you that the team is loaded with offensive talent. History tells us the same thing. Alex, Silas and Solo have a documented track record that can inform us that they are going to be north of average. In Solo's case, he is better than last year, his 3 pt shooting notwithstanding. He is playing better D and he is penetrating and mastering the mid range shot. So how to explain being 26th in the silly metric known as Kenpon? Easy, the team has not been together and healthy, for terribly long. Michigan, Arizona even Iowa St., they have not lost two starters for two months, like we did. They are great teams, playing at a high level. But if we were them, we would be worrying sick about peaking too early. Conversely, we are a great team playing at 60% efficiency with our best game in front of us, not behind us.

 

[SoFla Steve]

Here are our KenPom ratings the morning after each game. Our defense is decidedly better while the offense is trending in the wrong direction.

 

[NJHusky]

Here are our KenPom ratings the morning after each game. Our defense is decidedly better while the offense is trending in the wrong direction.

View attachment 115637

Is the 3 pt shooting the main reason for the low rating ? Or is it something else?

 

[Waquoit]

Here are our KenPom ratings the morning after each game. Our defense is decidedly better while the offense is trending in the wrong direction.

View attachment 115637

I coulda told ya UConn is inconsistent from 3 without a chart.

 

[SoFla Steve]

Is the 3 pt shooting the main reason for the low rating ? Or is it something else?

KenPom is all about efficiency. It appears to be 3-pt performance as well as free throws.

Our 3-pt shooting is all over the place:

 

[nelsonmuntz]

It's the percentage of 2-pointers made not attempted, or else there would be a column for 2PA%

I was referring to what kind of shots the stat was referring to. Auror answered the question.

I can't, sloppy description.

One of the issues I have always had with "analytics offenses" is they ignore the defense. Since most teams play man or a stretch zone defense, the defenders are all going to be at the perimeter, with help collapsing at penetration to the rim. Jay Wright proved that having everyone post up was still a pretty efficient offense, and now the mid-range is wide open. I would just keep flashing to the high post, with the option to catch and shoot, drive, or reverse the ball to the weak side. That flasher would get a lot of wide open looks, which even a mediocre shooter should hit around 60% of the time.

I also like post ups because they result in a lot of offensive boards. A putback is the highest efficiency shot on the court.

I don't like 25 foot penetrations. The defense has to really suck not to help out on that play, making that a tough finish at the rim. The shooting percentages don't include steals, which happen a lot on those kamikaze drives.

 

[OceanDoggie]

Here are our KenPom ratings the morning after each game. Our defense is decidedly better while the offense is trending in the wrong direction.

View attachment 115637

Nice. Never seen the game log with rank.

Makes sense, turnovers account for roughly 25% of Ken Pomroy's offensive efficiency rating. He incorporates turnovers with his TO% ratio.

Turnover percentage is a pace-independent measure of ball security.

TO% = TO / Possessions

 

[SoFla Steve]

I coulda told ya UConn is inconsistent from 3 without a chart.

But now you know exactly how inconsistent they are. 😉