If you watched Kentucky against Kansas on Tuesday, (or Memphis against Belmont for that matter), you know that recruiting is incredibly important in college basketball. But people often note that while Bo Ryan does not get the best high school talent, he finds players that can execute his system. 

So let us throw out the recruiting rankings for a moment. Which coaches get the highest level of performance from their freshmen, on average? And, since player development is equally important, how much does the average player improve over time?

The following table lists the average ORtg for Freshmen, Sophomores, Juniors, and Seniors under all the long-tenured coaches in BCS conferences. (When calculating the average ORtg I include all players from 2002-03 to 2010-11 with over 25 possessions for their team in the year.)

- Mike Brey may be at the top of this list, but his numbers are extremely misleading. As noted last week, Brey almost never gives playing time to freshmen. And because his freshmen rarely get to play or shoot, they rarely waste possessions. 

- Duke has the classic pattern with a huge jump in efficiency from freshmen to sophomore year. 

- Kansas, Duke, Louisville, and Texas typically have great recruiting classes and their freshmen tend to live up to the hype.

- But North Carolina has had a number of unproductive freshmen over the last eight years. Certainly, the mass defections after the pair of national titles left Roy Williams with two very inexperienced teams. (It is much harder to produce as a freshman without veteran players to complement you.) But I also think we do not give Roy Williams enough credit for player development. Think of how much Harrison Barnes developed throughout the year last season. Many of Roy Williams players have not been efficient out of the box, but if they stick around until their junior year, they are usually stars.

- South Florida is getting such a poor caliber of freshman, that no matter how much the players improve under Stan Heath, the team still is not very good.

- You will notice that a lot of these teams do not get peak efficiency out of their seniors. This may be a bit of a surprise given that players tend to improve over time. But remember that teams like Kansas typically lose their best players to the NBA every season. Thus Kansas gets it peak performance out of its juniors.

- Even with attrition, Frank Martin’s pattern is disturbing. You don’t want your seniors to be your worst performers.

- Don’t let the lack of tournament success fool you, Bo Ryan’s players have been more efficient than Tom Izzo’s over the last nine years.

- Keep in mind the averages at the bottom of the table. Most of these long-tenured coaches exceed the BCS average. Meanwhile coaches below the BCS average tend to get fired.

- But the averages also help evaluate the patterns in the data. While Tubby Smith’s freshmen have exceeded the D1 average, his seniors are below the average. That’s a disappointing amount of player development for the Gopher’s coach.

- On the flip side, notice how Bruce Weber, Matt Painter, and Leonard Hamilton’s freshmen have performed rather poorly. But their seniors are better than the BCS average. That is why Weber, Painter, and Hamilton get credit for developing talent.

For completeness, I also include the numbers for the newer BCS coaches. I would not draw too many conclusions for these coaches based on the small sample size: 

- John Calipari is only entering his third season at Kentucky, (but throwing out Mike Brey and his system that does not play freshmen), Calipari’s Kentucky freshmen have clearly been the most efficient in the nation.

- For new coaches that inherited unproductive seniors (see USC, DePaul, and Indiana), it is very hard to win.

About four years ago, I started asking the question how much of a coach’s success was due to recruiting and how much was due to player development. But I was never satisfied that there was a great methodology to separate these two factors. The best coaches also tend to recruit the best talent, and when you look at changes in the number of Top 100 recruits within programs over time, talent tends to be less important to explaining success than you might think. (See this post from a few weeks ago about how Florida and Georgetown had their best seasons with fewer Top 100 recruits.) But by looking at the player data, I believe we can take the next step in evaluating coaches. 

From showing that Bill Carmody’s players typically do not improve much relative to the BCS average, to showing that Jim Boeheim’s players typically experience a huge sophomore year leap in production, we can learn a lot about why these coaches have succeeded or struggled with their current programs.

More can still be done with this data. I should account for “inherited” talent. What Mick Cronin has done is even more impressive when you look at how depleted the Cincinnati program was when he took over. And I should do more to account for attrition. John Thompson’s numbers looks similar to Roy Williams’ numbers, but a significant amount of JT3’s player improvement has come because his most inefficient players have transferred out. Still, it is impossible to look at the coaches in this table and not see that every coach has a story.