To give you some context as to how the Draft Rater has performed in past drafts, I've included its projections for each draft going back to 2002.
Each chart shows the top 12 college players picked by the Draft Rater, as well as where those players were actually selected among collegians on draft night. As the charts show, the Draft Rater has generally held up very well; in fact, in most years it has produced better results than the actual draft.
As I pointed out in the story on this year's class, the Draft Rater rates "pro potential," which isn't always the same things as "pro performance" -- sometimes you'll have a Michael Sweetney eat his way out of the league or a Michael Redd work his way into the league by becoming a deadly jump shooter. Unfortunately, there's no stat for off-court habits.
Despite that limitation, the Draft Rater had performed extremely well prior to 2008, when it seems to have been tripped up by several one-and-done freshmen (the rankings have tended to be least reliable for that subset of players).
As mentioned in the larger story, from 2002 to 2007 there were 15 players who were taken among the first 10 collegians but whom the Draft Rater excluded from its top 12. None of them has played in an All-Star Game, and the only two long-term starters in the group have been Kirk Hinrich (who was 13th) and Charlie Villanueva.
Additionally, the Draft Rater has come up with some big draft steals: Carlos Boozer in 2002, Josh Howard in 2003, Danny Granger in 2005, Rajon Rondo in 2006, Rodney Stuckey in 2007 and Mario Chalmers and George Hill in 2008, among others.
A couple points we must make before we get started. First, the projected PERs for nearly everyone will seem a little low. This happens for two reasons.
First, the players who become superstars are almost always outliers not only in terms of ability but also in how much they improve in their first few seasons, so it's insane to project superstardom as the most likely outcome for any individual player.
Second, it's the nature of the beast: Using regression analysis has a lot of value in terms of ranking the players, but the one drawback is that the estimated PERs tend to clump around the 10-11 range much more than they do in real life.
With that said, let's take a look at the Draft Rater's top 12 players from each season since 2002 and throw in a few comments:
2002
2002 was a strong season for the Draft Rater, as it correctly recommended avoiding four players (Wagner, Ely, Haislip and Jones) selected among the top 10 collegians and made only one bad recommendation (Logan) in their stead. Boozer, Salmons and Prince all ended up as great value picks, with Boozer probably the best recommendation in the Rater's history -- the 26th collegian taken, he rated second here and is an All-Star.
2003
Draft Rater got the top two right but pushed down Bosh, Kaman and Hinrich a bit too far. On the other hand, it correctly identified David West and Josh Howard as strong prospects, and Luke Walton as a second-round steal. It also correctly avoided Hayes.
2004
2004 is the most similar to the real draft -- 11 of the top 12 college selections were the same. Only two of the recommendations were bad ones -- both Snyder and Jackson were busts, and the Rater had each a bit higher than they went in real life. On the other hand, most teams would be happier with Deng or Iguodala rather than Okafor, and the Rater found a second-round steal in Chris Duhon. Kevin Martin, a late first-round steal in real life, was 13th here.
2005
The Rater judged Deron Williams and Charlie Villanueva too harshly, but more than made up for it by pushing Chris Paul all the way to the top and Danny Granger to the No. 3 position. May came highly recommended and has played well when he's been on the court but has been undone by conditioning and knee issues. Jack, Garcia and Robinson were strong sleeper recommendations; Julius Hodge, not so much.
2006
This is arguably the Rater's best draft: It nailed five of its top six picks with only the Shawne Williams placement at No. 3 derailing it. That's a big improvement on what really happened on draft day, when Adam Morrison, Shelden Williams, Randy Foye, Patrick O'Bryant and J.J. Redick were five of the top 10 collegians taken. Only Morrison got into Draft Rater's top 10, and he was seventh rather than second.
2007
Moving Durant to the top spot and promoting Stuckey and Young were good calls, partly offsetting the demotion of Horford to No. 9 and the promotion of Crittenton to No. 6. Though highly anticipated, this draft has turned out to be a bit short on star talent thus far; we'll see in a couple of years how good this projection turned out to be.
2008
2008 saw, by far, the Draft Rater's most radical departure from the actual draft, and also seems likely to go down as the Rater's worst overall season. Love led all rookies in PER and Beasley was fourth among '08 collegians, so the top two picks are solid. After that, it's a mess. One-and-done freshmen Derrick Rose, O.J. Mayo, Anthony Randolph and Eric Gordon all ranked too low, as did solid rookies Brook Lopez and Russell Westbrook.
Of the players moved up in their place, Chalmers, Hill, Douglas-Roberts and Speights seemed like good recommendations; Speights finished second only to Love in PER from this draft class. On the other hand, Alexander, Arthur, Anderson and Jackson still have much to prove.
John Hollinger writes for ESPN Insider. To e-mail him, click here.