College Success and the NBA Draft

For the elite players in college basketball, a career is validated on that special day in June when an NBA team shows an interest in actually paying them to continue playing. Their NBA career could fizzle quickly, but on that day, the premier basketball talent evaluators in the world determined that they were worthy of a draft pick. The NBA Draft therefore gives us a blunt instrument with which to measure the amount of elite talent in each conference, so I've compiled the data from the last 10 NBA Drafts, using the awesome resources at Statsheet. First, the summary:

Note that, for now, I'm looking at BCS conferences only. This isn't to say that elite players don't come from other conferences (they certainly can and do), but I wanted to limit the scope of this post to something more manageable, and the vast majority of drafted college players come from one of these six conferences. The first two columns list the number of players drafted in the 1st and 2nd round from 1999-2008 from each conference. The next two columns take those numbers and divide them by the number of teams in that conference - this gives us a "per capita" number. The final column is just the addition of the two previous columns.

One last piece of semantics - I used the current composition of each conference and then went back team-by-team. For example, the teams that joined the Big East in 2005, like Louisville, are included in the Big East count all the way back to 1999. This maybe isn't ideal, but, again, I wanted to keep this manageable, and I think the tradition of the teams CURRENTLY in each conference is the most interesting thing to analyze.

So, looking at the data, it's clear that the ACC and Pac-10 are a notch above everyone else in terms of producing NBA Draft Picks. Between those two, the ACC has an edge in first-rounders, while the Pac-10 has a bigger edge in second-rounders. The average ACC team had 5 players drafted over the past 10 years, while the average Pac-10 team had 6 drafted.

There's not a lot of difference between the other four conferences (including our dear Big Ten). The average team from all of these conferences produced 4 draft picks over the past 10 years (2 first round, 2 second round).

Are there any overarching trends in the number of draft picks produced by each conference? Here's a look (1st round picks only):

We see that the ACC was generally among the best at producing 1st round picks, while the Pac-10 stayed within the pack until a big spike in 2008.

As far as the Big Ten goes, there's not really much of a trend. The two best years for the conference were 2000 (Jamal Crawford, Joel Przybilla, Mateen Cleaves, and Morris Peterson) and 2007 (Greg Oden, Mike Conley, Daequan Cook, and Alando Tucker), with four 1st rounders each season.

Looking at this data, I wondered how much of a correlation existed between NBA talent and NCAA success, when viewed at the conference level. Fortunately, Ken Pomeroy has ratings for each conference over the past 10 years. I took that data and ran a correlation between the Pomeroy conference rating and the total conference NBA draft picks for each season:

So, most seasons, there's a correlation between NBA picks and conference quality, but it's not typically as strong of a correlation as I'd expect. In fact, in one season (2003-04), having more NBA draft picks per team actually correlated (weakly) with having a worse conference. Still, when you average the past ten seasons, the correlation coefficient is .409. That shows that there certainly is a relationship between NBA talent and conference strength, but the bigger point is that NBA-level talent isn't the end-all in college basketball.

This makes logical sense - the number of college players drafted into the NBA is very small when compared with the number of players that make contributions to their team. Just considering the BCS conferences, there are 73 teams. Let's say the top 8 players on each team are contributors, which means there are 584 players every season that could be considered rotation members on a major conference team. In an average season, just 33 of those players are drafted - the top 5% of all rotation members. Needless to say, the performance of that other 95% can make a big impact on a team, or conference's, overall strength.

I'm wondering if it's more than mere coincidence that 2007, the first year of the one-and-done rule, is also the year with the strongest correlation between NBA picks and conference strengh. This idea may be worthy of deeper exploration.

If we look at individual teams, the overall leader in draft picks isn't much of a surprise:

Duke has the most first-round picks and the most overall picks over the past ten years. Not surprisingly, the winners of the last 10 National Championships are all contained in this list, but it's interesting to see the teams that ranked highly here and didn't win a title. UCLA, Arizona, and Texas are all teams that averaged at least one draft pick per season yet never won it all. Syracuse is the lowest ranked team on this list to win a National Championship (with Carmelo Anthony in 2003), and they had 7 NBA picks over the past 10 years.

It's clear that NBA-level talent has been a prerequisite for a National Championship over the past 10 years. Is that an earth-shattering statement? Of course not, but at least we now have the data to back it up.

As far as the Big Ten goes, here's the draft pick tally over the past ten years:

Honestly, I'm a little surprised that every team has at least one draft pick over that period. Of all BCS conferences, the only teams without a draft pick are Baylor and Oregon State. LaceDarius Dunn figures to get Baylor off the schnide in the next year or two; Oregon State may not be so fortunate.

To sum this all up, I think we've reached a couple interesting conclusions:

  • Over the past 10 years, the ACC and Pac-10 have produced the most NBA draftees per team and are clearly ahead of all other conferences.

  • There's no appreciable difference between the other four BCS conferences in terms of producing NBA draft picks.

  • There is not a consistently strong correlation between a conference's average quality and the number of draft picks produced.
    If anyone has any other ideas of what they'd like to see done with this data, drop a comment.