Kids today, Part II

We're going to take a quick breather from Big 10-focused posts, and move onto a topic that's a bit broader - freshmen.

Trying to evaluate high school basketball players can be tough. For one, adjusting for the level of competition can be challenging. Unlike college teams, high schools generally don't travel great distances and play against several teams in several conferences. And even then, most high school teams are filled with players who will never play at a higher level. What does it tell us when a 6-6 player dominates a team whose tallest player is 6-1? Furthermore, the only time the best high school players in the country are going against each other is on the AAU circuit. In case you haven't seen an AAU game recently, it bears little resemblance to the college game. Heck, it bears little resemblance to basketball generally. And somehow, from all of that, talent evaluators are supposed to whittle down to the best 100 players or so, and they must somehow decide whether the 6-6 SF from Georgia is better than the 6-1 PG from New Jersey, even though they have only seen each play for about 10 minutes.

And then, on top of that, they have to account for the fact that while some of these kids appear to be fully formed adults, some are 20 years old before they play their first conference game, and some hit a growth spurt after they are largely ignored by scouts. It's not an easy job.

Nonetheless, taken together, rankings of high school players do mean something. No doubt, rankings aren't everything. I'd be much more comfortable predicting a player's career armed with a year's worth of college data than a ranking. But I think we can deduce something from rankings. I've explored this topic previously, with the class of 2007. The quick summary on that class:

  • The top 10-15 players were excellent, star-of-the-team, high-efficiency players
  • The top 40 or so players were generally high-usage, efficient players
  • While lower ranked players were, generally speaking, of lower caliber than higher ranked players, the variance increased further down the rankings. In other words, it's a good bet that the #1 player is better than the #20 player, it's not such a good bet when comparing #100 to #80.

Now, I bring you the class of 2006. In case you don't remember, the class of 2006 was the year of Greg Oden and Kevin Durant. While the media attention certainly surrounded those two, it's clear to me that there were other stars in that class.

In looking through the data, I threw out guys who played less than 10% of the available minutes. That's to get rid of guys who were injured, guys who were playing behind great talent (like Marreese Speights was), and to avoid polluting the data with small sample numbers. It's a lot easy to look really good (and really bad) in 3 minutes a game. Why 10%? Totally arbitrary. Also, the rankings are RSCI.

Ranked 1-10:
Min%: 65.5
ORtg: 112.3
Shot%: 26.0
eFG: 54.6

These stats look a lot like those for the class of 2007's 1-10...but just a touch below. Less minutes (I know Oden missed a lot of time with injury, but even that wouldn't make it up), a bit less shots, but very similar. Overall, these are go-to players that can carry a team.

Ranked 11-20:
Min%: 46.6
ORtg: 103.3
Shot%: 21.25
eFG: 54.0

While there were some very good players in this group, there were also a couple of duds. Lance Thomas was probably not the 20th best player in America. Also interesting was that only one player in this group (Javaris Crittendon) more than 65% of the available minutes. It would appear that there were less superstars in the class of 2006 than the class of 2007.

Ranked 21-30
Min%: 55.1
ORtg: 104.8
Shot%: 19.5
eFG: 51.4

While these players still give efficiency, the steady decline in Shot% continues. These players are capable of coming in and starting, but would be overmatched in carrying a team.

Ranked 31-40
Min%: 47.0
ORtg: 96.0
Shot%: 22.4
eFG: 44.4

I'm not quite sure what to make of this group. The Shot% was high because there were several players taking a ton of shots (DeShawn Sims, Raymar Morgan, Scotty Reynolds), and for the most part, they didn't shoot it well. They were probably overmatched on that shot diet. However, that ORtg isn't really representative of the group, and is pulled down by James Keefe, who had one of the worst seasons of any freshman in a BCS league.

Ranked 41-50
Min%: 45.0
ORtg: 103.0
Shot%: 16.2
eFG: 51.9

This is a group that wasn't asked to do too much, and flourished in that role. The Shot% is indicative of a role player.

Now, for some pretty graphs on the top 100 players...

While there's an overall steady downward trend in each of these, there's appears to be a bit more randomness than with the Class of 2007. Also, while the Class of 2006 enjoyed higher ORtg numbers, they didn't take as many shots. Indeed, only one range of players (51-60) put up a higher Shot% than its counterpart in the Class of 2007. The Class of 2007 definitely had more superstars as well, so I think it's fair to give them the overall nod.

Up next, the class of 2005 (where we can see the impact of the Stern Rule)...