Do players develop better in the NBA or NCAA? The stats are unclear

We are drawing ever closer to the NCAA Tournament, a time when college kids and their chronically unhappy coaches become the brightest stars in the American sports world. For NBA fans, it's also a particularly useful time to see which athletes look prepared for professional success. However, that scouting process has become increasingly more difficult in the time of rampant early entry. Draft prospects often look in process, if not entirely undeveloped, and it's anyone's guess as to how they'll adjust to the NBA.

Scouting is hard enough, in fact, that there's a debate raging over which organization allows players to develop best. First-year commissioner Adam Silver has spoken often of his desire to raise the age limit, citing the belief that players will be more prepared for the NBA (and therefore have better careers) after spending more time developing in college. Dallas Mavericks owner Mark Cuban has argued that this idea is hogwash, saying that the D-League, with its NBA-aping rules and total on focus on basketball, would be much better training for these young men.

Neither argument is especially scientific. So, Kevin Pelton of decided to run the numbers on a set of players who left college after their freshman (21) and sophomore (14) seasons and played 500 minutes in each of the three tested years. The results, while limited, indicate that the NBA develops players better:

We're not interested in the overall performance of these groups anyway. Instead, we want to focus on how they developed year to year. That's where my NCAA-to-NBA translations come in handy. They allow us to put college and NBA performance on the same scale (using player win percentage, the per-minute component of my WARP rating that is equivalent to PER).

That shows something remarkable. On average, the sophomores who returned performed only marginally better than they did as freshmen.

Amazingly, of the 14 sophomores who qualify (which requires playing at least 500 minutes all three seasons, a criterion that knocks out Blake Griffin, among others), nine rated worse as sophomores than freshmen. That includes basically all the high-profile freshmen who passed on the draft and saw their stock fall. [...]

Although [the one-and-done] group rated slightly better as NCAA freshmen, which makes sense given their perceived higher upside, 15 of the 21 improved as NBA rookies relative to their translated NCAA performance. On average, their win percentage went up by 10.5 percent, better even than we'd expect from players of this age.

Now, this study could be picking up on the superior potential of one-and-done prospects, a possible factor in why they generally were more coveted after one year in college. However, the development advantage disappears by the time both groups are in the NBA. In their third year out of high school -- the rookie season for the sophomores and second year for the freshmen -- the sophomores actually improve slightly more. But this difference isn't nearly enough to make up the development they missed out on between their two years of college.

The full charts are available in the Insider-only article, so I advise checking them out if you have the interest and ability to do so. While the one-and-done players would appear to have an edge in terms of overall quality (and they do), the sophomore group includes such talents as James Harden and Paul George. It's not as if the talent gap is glaring, and certainly not enough to make up for the attendant statistical difference.

At the same time, it's clear that these numbers have some limits. For one thing, the sample size is very small, both because the age-limit era has been relatively short and for the simple fact that not many players become rotation players as rookies and second-year pros. But it's also much more complicated than that. As Kevin says, some sophomores stayed in school when they might have been lottery picks, but more do so because they lack buzz as draft prospects at that point in their college careers. Paul George, for instance, would not have earned minutes as a rookie if he'd left after his freshman season -- he stayed at Fresno State because the alternative was an uncertain future in the D-League with a less certain chance at a roster spot moving forward. He stayed in school in part because that path was well trod and safer. Even if Cuban is right that the D-League would prepare players better, the fact that it's a new path introduces more variables into the equation of pro success. George entered the NBA after his sophomore season with a general sense of how his NBA team would move him along, and that might have helped him become a better pro if only because he knew more about what the future might bring.

This running of the numbers, then, mostly suggests that a player who already has a secure sense of his likelihood of holding down an NBA roster spot would do well to enter the draft as early as possible. Cuban's suggestion of using the D-League as a minor league, however, looks more like a way for teams to assess a player's readiness for contributing to a squad, not ensuring that the players will put themselves in the best position to succeed. It's a difference of perspective, but an important one that shifts from the side of management to that of the players. It could be that what we observe as on-court development depends on a player's comfort more than we are willing to admit. The NBA development of those players measured by Pelton might be impressive in part because each considered himself ready to play in the league. The relatively unimpressive marks of college sophomores could represent an important incubation period, even if they did not demonstrate sizable NCAA development.

Frankly, these sample sizes are too small and too limited to particular scenarios to tell us much about the effects of college or pro development for elite prospects. As ever, this is one question with an answer that will be defined primarily by the opinion a person has before looking at the data. Until there is a major change to the landscape -- perhaps by the introduction of paying college players and/or a questioning of the NCAA's broader practices -- we simply don't know enough about various situations and player paths to know which works best. This debate will rage for a while, and it's unclear when or if it will be resolved.

Of course, it could be that the age limit question has much more to do with the extent to which the NBA and its member teams can control the flow and quality of its labor force. Silver and Cuban, while possessing different opinions on the best way to do so, are both concerned with their ability to know which players will help the league most. An extra year in college, even if it keeps players from reaching their full potential as soon as possible, at least allows management to collect more information on incoming players. The house stays winning.

- - - - - - -

Eric Freeman is a writer for Ball Don't Lie on Yahoo Sports. Have a tip? Email him at or follow him on Twitter!

Stay connected with Ball Don't Lie on Twitter @YahooBDL and "Like" BDL on Facebook for year-round NBA talk, jokes and more.

What to Read Next