There is a clear cutoff for basketball recruiting royalty, and it comes right after Duke, North Carolina, and Kentucky. You can slice the data a few different ways to make it look like Kansas is close to being the third-best recruiting school in the recent past, but it's essentially impossible to put them above fourth.
Over the past ten years, the Blue Devils, Tar Heels, and Wildcats have combined to sign 91 top-100 recruits (nearly 10 percent of every top-100 prospect available) and 28 top-10 players (nearly 30 percent). They've put forth all of the top six classes since 2002 and 13 of the best 19. Kentucky has had the top class nationally three years running, North Carolina's ranked 2-3-5, and Duke's finished 9-6-2. In the past three years alone, these three programs have landed top-10 prospects DeMarcus Cousins, John Wall, John Henson, Mason Plumlee, Harrison Barnes, Kyrie Irving, Brandon Knight, Terrence Jones, Anthony Davis, Austin Rivers, Michael Gilchrist, James McAdoo, and Marquis Teague.
All of which is to say that these programs have an entirely different talent distribution to work with than anyone else in college basketball. We had a theory in place that, for these programs, the bar for playing time was being consistently set so high that it was barely worth recruiting players lower in the rankings.
Let me explain. These programs go after top-flight talent that they know may leave for the NBA, but back it with players they know will be in for the long haul. We theorized that the four-year players were actually getting crowded out consistently enough that they were transferring out just as fast as the higher-ranked players were heading for the pros.
So we put together a list of every Duke, North Carolina, and Kentucky commit since 2002 and checked how long they stayed with the program. Then we sorted them into groups for top-3 players, Nos. 4-10, 11-25, 26-50, 51-100, and unranked players, and looked at what happened.
1. The breakdown of talent was different for all three schools. North Carolina lives in that 4-10 range -- the great majority of their value comes from there. Duke avoids that set almost entirely, and works largely from the Nos. 1-2 and 12-40 ranges. Kentucky works in the top 3, clearly, and has a reasonable presence in each of the rest of the groups. They're also the only one of the three elite programs willing to use junior college transfers.
2. While only one of the top-3 players came back for his junior season (and only three of the ten who've had the chance even made it to their sophomore seasons), 11 of 13 in the 4-10 range were back as sophomores and nine of 12 came back as juniors (with Terrence Jones yet to decide). This makes it interesting that Duke has chosen to largely ignore this group. I feel like the staff normally tags this type of player "one-and-done" because they often did leave directly out of high school before one-and-done was instituted. Since the rule's been put in place, though, this player normally stays in college for a few years. These three schools inspired an abnormally high average career length for the 4-10 players, it's true, but they typically last more than two seasons at other schools, as well.
3. The transfer rate is so high at this level that lower-rated players rarely stay at the school for four years. Ten of the 15 players ranked 51-100, nine of the 17 unranked players, and just nine of the 18 players ranked 26-50 are projected to play all four years at the school that signed them. More of the players ranked 4-10 made it to their junior year at the school than did those ranked 51 and lower, and the rate was identical for 4-10 and 26-50. When you're at the absolute top of the recruiting game like this, those lower ranked players are crowded out anyway, and it rarely makes sense to chase them as four-year players.
The conventional wisdom, especially at the Duke-Kentucky-North Carolina echelon of the college basketball world, is that programs should build a foundation of four-year players, and augment them with supremely talented, short-term stars. I say this strategy is only taking into account the most likely scenario, without considering the way things will change in the future. It's reductive.
First, players recruited as one-and-done stars don't necessarily turn out to be one-and-done stars. The unstated assumption is that top-10 high school talents will actually comprise the ten best NBA prospects a year after high school graduation. And, sure, this is often true, on a player-by-player basis. When you're recruiting the No. 7 player in the country, he seems like an NBA prospect, and he is one. When you're recruiting the No. 96 prospect in the country, he doesn't seem like an NBA prospect, and he isn't one.
The problem with this reasoning is that things change. There's a decent chance the No. 7 player, by choice or by lack of choice, will find himself still in college as a sophomore. In the five years since the one-and-done rule went into effect, over 40 percent of top-10 players have returned to school for their sophomore seasons (and, once you exclude top-3 stars, it's essentially a 50-50 proposition). Even though Terrence Jones looked like a one-year player as a high school senior, he wasn't, and neither was Mouphtaou Yarou, for a different reason. On a case-by-case basis, it's easy to forget that players don't always progress as expected.
Secondly, recruits signed as four-year-type players don't necessarily turn out to be four-year players. The typical player ranked around No. 35 through the typical player ranked 100 have a remarkably stable average of three years at their original school. It's even more true at the highest level, though; see point No. 3 above. Between the demand for playing time, personal issues, and the occasional talent explosion, four-year players don't always progress as expected, either.
Every time I take a hard look at the mistakes being made in recruiting, they tend to fall under the umbrella of overspecifying needs. Rarely is it true that a team needs a four-year, prototypical power forward. It's much more likely that they're lacking a strong rebounder for the next season, or a long, versatile defender for the next two, but it's easier to describe that search in more certain terms that leave less out. If I were a coach at a top program, though, I would throw all the definitions out the window. I'd have a caveman-dumb recruiting strategy: I'd try to get as much talent on campus as possible while ensuring that each player fit my program. I wouldn't worry much about the team two and three years down the line except in understanding that that team will be better through the improvement of this year's team.
The one-and-done players and the program builders will make themselves known, and they'll do so whether or not I try to identify them beforehand.
Drew Cannon is a college student and a regular contributor to Basketball Prospectus. Follow him on Twitter at @DrewCannon1.
This free article is an example of the content available to Basketball Prospectus Premium subscribers. See our Premium page for more details and to subscribe.
Drew Cannon is an author of Basketball Prospectus.
You can contact Drew by clicking here or click here to see Drew's other articles.