Talent distribution: the ACC
If you examine the rosters of NCAA Tournament champions they have a couple things in common. One, they're filled with elite recruits. Two, they have multiple players who end up as 1st round picks in the NBA. Those two things are not always interchangeable. Not every elite recruit pans out. Not all 1st round picks were elite recruits. But it's pretty darn close. Evaluating talent in basketball is easier than it is in other major sports. You could watch Michael Kidd-Gilchrist in high school (then, Michael Gilchrist) and see an NBA player. It's hard to look at a high school quarterback and see anything more than NFL potential.
This is the reason why I track where elite talent ends up. Those schools win titles.
Today, it's the ACC. For a data source I use RSCI Hoops, which, if you aren't familiar with this site, you should be. They take ESPN, Rivals, Scout, etc... rankings and combine them into one consensus ranking. Other sites have co-opted their methods (without citing them of course), but RSCI is the original. Run the Floor even creates some consensus rankings, but we use those as a stop-gap between RSCI updates. In the end, we defer to RSCI.
Combing through the ACC rosters it turns out that there are 56 players who were consensus top-100 recruits at RSCI. Here's how that distribution looks:
As expected, the Tar Heels and Blue Devils dominate the accumulation of talent. The biggest and baddest NCAA programs often have rosters like these, where most of the players were a top-100 recruit. In this case they each have nine on their rosters (from 13 available scholarships each). Georgia Tech and NC State show off their rapidly rising programs with seven and six players apiece. FSU and Virginia are at the upper-middle, and the rest of the ACC (minus one) all have three. Boston College (the minus one) has zero.
*Note: my original chart showed VT with three, but I had not reinstated Jerrell Eddie (No. 89, 2010) as the Hokies chose to do.
So what happens if we refine this search a bit. Limiting this to just the consensus top-50 players that same chart looks like this:
This shows the talent gap is expanding. Where UNC and Duke had 32% of the top-100 players in the ACC, they now have 47% of the top-50 recruits. NC State, FSU and Georgia Tech are still slightly above the rest of the conference, whereas Virginia Tech now joins Boston College on the wrong side of the chart. Dorian Finney-Smith was the consensus No. 22 recruit in the nation, but he transferred from the Hokies to the Florida Gators.
The final chart pares it down even more. Here is the same chart, but only with consensus top-25 recruits.
There are 12 consensus top-25 recruits in the ACC, and 8 (67%) play for UNC and Duke. Keep this in mind when you're projecting your pre-season champion.
Nice post!
A couple of things, from an NC State fan:
1. I hate the use of NCST. It's a pet peeve of mine and most State fans. Yes, it's what ESPN has programmed into the collective subconscious instead of the more-accurate NCSU.
2. If I can play devil's advocate, how many of these elite recruits see their rankings improve once they become UNC and Duke targets and then ultimately sign with them? Basketball recruiting is more precise than football recruiting, but it's not uncommon for players to wind up on all-star squads (McD's AA, for example) ahead of other, more-talented players almost certainly based on the jersey they committed for, not their actual skill level.
Because of their recent success, Duke and UNC certainly draw a lot of attention from the best-of-the-best, and both schools would lead in these types of roster comparisons regardless of whether rankings inflation occurs. But I wonder how many borderline guys go from a #31 to a #24, for example, or a #58 to a #50, based on their offer lists and commitments. You can see how a guy/recruiting service putting together these lists might think, "I've got him at 65 on my board, but UNC and Kansas are both offering...maybe I should take a closer look to see what they're seeing."
The use of RSCI is great, as it helps smooth out the aberrations.
An interesting backtest might be to compare the success rate (i.e. college awards, pro basketball success) of Duke and UNC's collection of top 100 recruits against all the remaining top 100 prospects over the last 10 years. As inexact as this stuff is, the numbers may not vary enough to definitively draw a conclusion about inflation one way or another, but who knows?
@RnR_NCSU Consider my mind blown. How have I never encountered the NCSU/NCST argument before? But Florida State (FSU), Kansas State (KSU), Washington State (WSU), etc... all follow a pattern which NCST does not. I feel programmed! NCSU it is.
As to ratings inflation, intuitively I don't buy it. But I don't have any data to back that up. I'd love to do a project to investigate that.
I definitely believe that there's a bias towards kids who mature earlier - in other words, take 2 hypothetically identical star HS seniors, and one has been on the radar since 8th grade, and one is new, the kid who has been scouted since 8th grade will be rated more highly by the services. But that type of inflation/deflation would be distributed in a more random pattern.
@Michael Rogner "How have I never encountered the NCSU/NCST argument before?" LOL, probably because we're the only folks who care about it. :) It's not uncommon to see instances where a school has a modified acronym when there are multiple schools with the same acronym--OSU means Ohio State; OkST means Oklahoma State. Same with USC/SCar.
But why NCST? Other than a Northern Colorado State U that may or may not exist, there aren't any other college programs that could share that acronym or be confused with NCSU. It seems like an unnecessary modification to avoid confusion that doesn't exist. I know State hasn't exactly lit the world on fire in basketball/football in the last 30 years, but they haven't taken our two NCAA titles back and we are still a founding member of the ACC and one of the key cogs of Tobacco Road. I think folks can figure out NCSU = NC State U.
I know what my diploma reads, and it's not NC State Tech, let's put it that way.
Keep up the good work!
