Tuesday, April 07, 2015

One and done,
Spoils the fun.

The Final Four is a very small sample on which to rank the teams who play in it.  Just to make that point, below are the Sagarin rankings for the top twenty teams based on the entire season.  Sticking with just the top three, the Final Four ranking would be 1. Duke, 2. Wisconsin, and 3. Kentucky.


The Final Four had added interest because two of the teams, Kentucky most famously but Duke as well, featured the new one-and-done model, with the teams featuring freshman stars who are likely to declare for the NBA draft, while the other two teams, both from the Big Ten, employed the more traditional model where seniors and juniors form the bulwark for the clubs.  Each game played in the Final Four was new model versus old.  The results gave the new model 2 victories and 1 defeat.

Invariably, given this conclusion, the new model is taken to be triumphant and Coach K is looked at as a genius for having embraced it.

Duke won its fifth national championship with four freshmen combining for 60 of the Blue Devils' 68 points and a rookie accounting for every single Duke point in the second half. Tyus Jones finished with 23 points, and Grayson Allen, who had 18 in five combined NCAA tournament games prior, finished with 16.

It is anathema to what Duke had long been, a team built on wily seniors who stuck around and eventually won a championship. The last time the Blue Devils won a title, in 2010, Krzyzewski had mostly avoided one-and-done players. By the end of June he might be saying goodbye to three who remained in Durham, North Carolina, for only one season.

Yet what's maybe even more astounding is that Krzyzewski has not only made the change.

He has loved it.

There isn't enough data to support these conclusions, though that won't stop the sports pundits from making them.  But here I'm more interested in what the coaches actually think rather than what the pundits tell us about them.  Are the coaches of one mind on this?  Or are there some in both camps?

It seems pretty clear that the new model can work only at a few elite programs at most - there isn't enough talent to go around for it to become the norm for most programs.  What is less clear is whether among other programs that go the traditional route if at least a few will emerge that give these elite teams a run for their money.  To me, the most interesting game in the tournament was Wichita State beating Kansas.  Yet Wichita State seemed outgunned when playing Notre Dame.  A few years ago Butler was a compelling story.  NCAA basketball needs that type of story to keep up fan interest.  Wisconsin provided that type of story this year, even though they play in a so-called bit-time conference.

If those stories begin to dry up, college basketball will start to seem like it is replicating the income inequality in society as a whole.   Coaches are under a lot of pressure from their fan base, who want to win now.  I wonder whether some coaches at elite programs who actually prefer the traditional model will nonetheless make the switch to the new model because the fans demand it.

One potentially offsetting factor is the Moneyball effect.  By this I mean that for players in high school who are better than their competition it is unclear how they will perform once they face a higher caliber competition. So the becoming a McDonald's All American and the various rankings of players at the time they are being recruited out of high school should be thought of as noisy (in a statistical sense).  The more the noise, the less good should the new model perform over time.   But the talent pool should be thought of as a pyramid.  Is there as much noise near the top as there is at the base?  Even if there is not, is it still noisy at the top?

Lebron is Lebron and Kobe is Kobe.  There wasn't much noise in their case.  (Lebron was drafted number 1 so you might say there was no noise in his case.  Kobe was drafted 13th.  Given that Michael Jordon wasn't drafted number 1, I prefer to think there always is some noise.)   But players like that aren't in every draft.  Last year Michael Carter-Williams was the NBA Rookie of the Year.  He was drafted 11th.

The other factor that could eventually offset the new model is evidence that players' growth in basketball skills and general maturity happens more in college (at least at some programs) than it happens in the NBA.  This would give talented players at such programs some incentive to not declare for the NBA draft that is other than that they simply are having a great time in college.  I have a sample of one in mind in thinking about this.  Deron Williams left Illinois after his junior year and a season where the team made it to the NCAA Championship game.   He might have gone out after his sophomore year, but I believe he timed it right doing things as he did.  There would need to be many other such examples to convince kids it is good for them to stay.  Some of the Kentucky players are sophomores, but it is unclear whether for them that was the reason or if  they felt their talent wasn't up to par.

Division 1 men's college basketball serves multiple functions.  One, clearly, is as a minor league for the NBA.  Another is as entertainment unto itself, with a fan base some of whom care more about the college than the pro game.  The old model respects the multiple functions.  The new model, not so much.  That's the issue.  

No comments: