I’m beta-testing a new predictions model based on 10 years of player stats. In this piece, I want to share a few preliminary numbers, while explaining where I disagree with other college basketball experts.

1. Is Michigan still a top 25 team?

In February, I could understand the love for Michigan. The Wolverines' projected 2012-13 lineup looked something like this:

ORtg

Pct Min

Pct Poss

Player

Rank

Class

Ft

In

119.3

79%

21%

Mitch McGary

3

Fr

6

10

112.7

69%

25%

Tim Hardaway Jr.

 

Jr

6

5

111.3

65%

24%

Trey Burke

 

So

6

1

117.9

53%

21%

Evan Smotrycz

80

Jr

6

9

107.7

49%

19%

Glenn Robinson

52

Fr

6

5

112.2

48%

19%

Jordan Morgan

 

Jr

6

8

104.4

42%

19%

Nik Stauskas

79

Fr

6

6

102.4

28%

18%

Max Bielfeldt

 

Fr

6

8

102.4

27%

18%

Caris LeVert

 

Fr

6

4

109.5

25%

14%

Matt Vogrich

 

Sr

6

4

102.4

 

 

Spike Albrecht

 

Fr

6

1

99.1

 

 

Jon Horford

 

Jr

6

9

Tim Hardaway really struggled with his three-point shooting last year, but based on his shooting prowess as a freshman (76 made threes), I expect his outside game to return. It is not uncommon for shooting guards to see their efficiency jump around, but if you were a good three point shooter once, odds are you can be again. Trey Burke was better than Hardaway in 2012, but based on Hardaway’s two-year profile (and particularly based on that three-point shooting), Hardaway projects to have a slightly more efficient season.Quick Notes:

If you think Jon Horford will crack the lineup, I won’t argue with you. But Horford didn’t play as a freshman, was injured in December last season, and basically hasn’t played meaningful minutes in a Big Ten game yet. That’s not a good profile and his projection reflects that. Based on John Beilein’s historic ability to turn unranked freshman into quality players, red-shirt freshman Max Bielfeldt actually looks marginally better on paper. But if you want to put Horford in the lineup as a backup post player, that isn’t crazy.

The Outlook Changes:

Unfortunately for Michigan, two things happened this spring. First, Mitch McGary plummeted in the recruiting rankings falling from 3rd to 26th. Second, Evan Smotrycz elected to transfer. The difference for Michigan is very significant.

ORtg

Pct Min

Pct Poss

Player

Rank

Class

Ft

In

109.6

71%

25%

Tim Hardaway Jr.

 

Jr

6

5

108.3

69%

25%

Trey Burke

 

So

6

1

105.0

61%

20%

Glenn Robinson

23

Fr

6

5

109.5

58%

20%

Jordan Morgan

 

Jr

6

8

105.0

58%

20%

Mitch McGary

26

Fr

6

10

101.8

44%

19%

Nik Stauskas

78

Fr

6

6

99.9

43%

19%

Max Bielfeldt

 

Fr

6

8

99.9

28%

19%

Caris LeVert

 

Fr

6

4

107.1

27%

14%

Matt Vogrich

 

Sr

6

4

99.9

27%

19%

Spike Albrecht

 

Fr

6

1

97.0

 

 

Jon Horford

 

Jr

6

9

Evan Smotrycz was Michigan’s under-utilized star. Smotrycz was already super-efficient, and as the only returning Michigan player to be ranked out of high school, his potential was still very high. With fellow three point sharp-shooters Zack Novak and Stu Douglass having departed, Smotrycz seemed poised for a breakout season. His loss pushes a bunch of players with weaker projections into a higher spot in the lineup.

More importantly, the historical difference between the 3rd ranked recruit and the 26th ranked recruit is immense. The 3rd ranked recruit is almost always an immediate star. The 26th ranked recruit will usually become a solid college player, but not necessarily make an immediate impact.

With Mitch McGary plummeting in the recruiting rankings, not only does this lower McGary’s projection, it also has spillover effects to the rest of the team. First, Max Bielfeldt (or Jon Horford) is now likely to play more minutes in the post. Second, the weaker overall lineup hurts everyone. Everyone (from Tim Hardaway to Caris Levert) will likely get fewer wide-open shots, and their ORtg projection falls slightly.

For all these reasons, I just don’t see Michigan as a Top 25 team anymore. If you love recruits, even Wisconsin brings in a higher ranked post prospect (Sam Dekker) than Michigan at this point. (Of course knowing Bo Ryan, Dekker will probably red-shirt.) But at least on paper, I have a hard time generating a lineup where Michigan’s offense looks scary enough to deserve an elite ranking.

Caveats:

There are two ways you can generate a stronger prediction for Michigan. First, maybe Michigan’s defense will finally be elite this year. McGary’s biggest impact may very well be on the defensive end. But in 10 years, John Beilein has never had a tempo-free defense ranked in the Top 25 nationally.

Second, Beilein will likely use a tighter lineup than what you see here. I’m projecting more limited minutes for Burke and Hardaway than last season, as Beilein has a bunch of young players that deserve a chance to crack the lineup. Perhaps by the time the Big Ten season comes around, a clear rotation of 7 players will evolve, and Beilein can short-lineup his team to an effective offense. But for his deep recruiting class to pay dividends, those freshmen need a chance to play. And after a rough final year of high school, even Mitch McGary no longer projects as a can’t-miss freshman at the college level.

2. Should Duke be the ACC favorite?

I understand why most people feel lukewarm about Duke heading into this season. Last year was not a vintage season, and while Duke has a few key prospects coming in, none of them have received the hype Austin Rivers did last year. But is Duke really the third best team in the ACC as I’ve seen in some publications? I think the answer all comes down to what you expect Duke to do on defense this year.

Duke

Adjusted Defense

Average Ht. Inches

Returning Minutes

2004

85.4

77.21

75%

2005

84.9

77.19

64%

2006

89.8

77.09

69%

2007

85.6

77.56

40%

2008

87.6

77.10

82%

2009

90.8

77.78

80%

2010

85.9

79.13

60%

2011

88.7

78.12

58%

2012

95.8

77.96

61%

 

Prediction

   

2013

90.2

77.82

61%

In my eyes, 2012 looks like a fluke. It was clearly Mike Krzyzewski’s worst defensive team in a decade. And the last time Krzyzewski had an adjusted defensive rating above 90, his team bounced back substantially.

But another way to look at this data is to say that Duke’s defense has been steadily slipping in recent years. Duke has had four seasons with sub-86 defensive ratings, but only one since 2007. And that year Duke had a 7 footer, Brian Zoubek, manning the interior. (Notice that the solid defense in 2010 coincided with an increase in team height.)

I don’t think that is quite fair though. Zoubek was not, by any stretch of the imagination, a dominant shot-blocker. In fact, Mason and Miles Plumlee both had better block rates in 2012 than Zoubek did in 2010. Perhaps Zoubek’s additional size did help alter shots, but Duke is certainly not lacking size by any historic metric.

The trouble, as many people identified last year, was that Andre Dawkins, Seth Curry, and Austin Rivers were liabilities on defense. You can afford to have one slow-footed shooting guard on the court at once, but putting Curry and Dawkins on the floor together was usually a recipe for disaster. And despite his nice shooting touch, Dawkins saw his playing time slip because of his defense. Now Dawkins is taking a non-basketball related red-shirt and Rivers has left for the NBA. That should translate into better perimeter defense. Freshman Rasheed Sulaimon, Tyler Thornton and a healthy Quinn Cook are likely to be better defenders this season.

Most importantly, I believe that coaching is the most important input on defense, and I cannot see Mike Krzyzewski letting Duke’s defense slip for a second year in a row. He can see the stats as well as I can, and he knows he needs to emphasize that side of the ball from the start of practice in the fall.