The 2008 NBA Draft Preview

Today’s guest post is by Erich Doerr .  Erich first contacted me prior to the 2006 NBA Draft with a statistical preview in hand.  Each subsequent year has seen improvement in the depth and breadth of his analysis. This post continues the WoW Journal’s 2008 NBA Draft coverage. Outside of his basketball writing, Erich does consulting work for major software products by day and has started a fledgling sports-themed Open Source software initiative by night. 

Let’s start with the numbers.  Table One reports an analysis of the college prospects available for the 2008 NBA draft.

Table One: 2008 NCAA Prospect List

What follows is a discussion of these numbers. 

Basic Methodology

Since Bill James broke baseball down into numbers, similar statistical analysis has taken place for each major sport.  One recurring result is that scoring margin provides a strong predictor of future performance (even stronger than winning percentage).  Scoring margin, the difference between points scored and points allowed, can be broken down to individuals and transformed into statistics like Win Shares and Wins Produced.

In the Wages of Wins, David Berri, Martin Schmidt, and Stacey Brook use econometric analysis to generate an approximation for individual player contributions towards Wins Produced, and call the results Win Scores.  Their analysis lays out a simple formula for player evaluation that can be applied to any common basketball box score.

While their analysis was based on the National Basketball Association, the same metric can be applied to players in other leagues, including the NCAA, which is the NBA’s largest feeder league.

With the NCAA season over, Win Scores has been calculated for all the 2008 NBA draft prospects and an assessment can be made on draft worthiness.  As a basis for evaluation, DraftExpress‘s mock draft will be used to represent the industry consensus on prospect values.  From this basis, we will use Win Scores to identify over and undervalued players in the draft’s lottery (first 14 picks) and furthermore name a handful of valuable prospects likely to be available late in the draft. 

Before we get to the players, though, let’s discuss two comparative adjustments I think are necessary to discuss college numbers.

Strength of Schedule

Wins Produced and Win Scores were developed based on NBA statistics.  In the NBA, talent is relatively evenly dispersed when compared against lesser leagues.  While the NBA’s talent is spread among 30 clubs, the NCAA has 341 division I-A teams, running the gamut from talent-laden National Champion Kansas to the 0-29 New Jersey Institute of Technology

A player’s success depends not only on his prowess, but also on what his opponent allows them to do.  With this in mind, last year we compared a given player’s WS marks while playing against tournament-qualifying teams next to their marks against all other opponents.  Most players posted significantly better marks against non tourney teams.

This year, we will rely on a similar measure.  Specifically, Ken Pomeroy’s team ratings will define the 100 best teams, which will serve as the basis for strength of schedule analysis. 

By using Ken Pomeroy’s team ratings over the NCAA tournament field, we are able to exclude poor quality tournament teams such as Coppin State (ranked 310th out of 341 teams by Pomeroy’s stats).  Using 100 teams as a common basis also allows for larger sample sizes while at the same time maintaining a respectable level of quality competition.

In the 2008 preview table, these statistics are notated as KP100 for PAWS/M against top 100 teams.

Pace Adjustment

Given a variety of offensive and defensive schemes, a box score metric like Win Scores is susceptible to the number of possessions in a game.  Certain coaches slow the game down by applying conservative defensive principles and clock-eating offenses while others prefer aggressive defenses and high tempo offenses.  Given these disparities, a pace factor can be calculated by assessing the average number of team possessions in game.  The pace factor can then be applied to Win Scores to find a tempo-neutral value.

For an example, the first table in Tables 2-4 (below)compares two players from a fast paced team (North Carolina) against two similar players from slower paced teams (Georgetown and Washington State).  After adjustments, these player’s grades come out much closer than originally calculated.

Table 2-4: Additional Analysis of the 2008 Prospects

Pace adjustments for all players are included in the complete 2008 prospect list.

Projecting the Projected First Round

Looking at the projected first round we see players where the numbers suggest optimism, neutrality, and pessimism.  Let’s start with the three players where the numbers are encouraging.

Optimistic

In an Early Look at the 2008 Draft, Michael Beasley was dominating both headlines and Win Scores.  Nothing has changed.  Beasley remains the #1 talent available.

After Beasley we see two more big men.  While mock drafts have Kevin Love going anywhere from 5 to 15, Win Scores followers should consider Love the #2 or #3 talent in the draft, hands down.  Additionally, Florida big man Marreese Speights is one to watch.  Currently his name is bouncing around mock drafts in the mid first round, though by the stats, Speights seems to be surefire top 10 material, if not better.

Neutral

Derrick Rose is coming off of a strong national championship run, and with the rise of Deron Williams and Chris Paul, its going to be hard to pass on a tall, explosive point guard prospect. 

When I wrote January’s article, Derrick Rose had a .052 PAWS/M and warranted a warning rather than a recommendation.  He continued mediocre play until February 20th, and then – as Table 3 indicates (see above) — began playing like a top prospect. 

His late season statistical explosion, plus the high regard from the scouts suggest Derrick Rose is for real, but his early season struggles and high expectations prevent me from predicting a surefire success. 

Another top prospect is Brook Lopez.  He grades out high on PER, but Win Scores is skeptical given his non-impressive rebounding rate.  The numbers don’t warrant excitement, but compared to the current projected lottery (mostly listed below), Lopez grades out favorably.

Two other neutral prospects are DeAndre Jordan and Donte Green.  Jordan, a Texas A&M center, posted a full season .047 PAWS/M, but seems to have picked on the little guys, as his scores against tougher competition get ugly quickly.  Meanwhile Green had an excellent start last season, but he actually hurt his team with a sour second half and wound up with a negative PAWS/M (which is far worse when Syracuse’s fast paced play is considered).  If he returns to his early form, he’s well worth a lottery pick. If his second half is more like it, be afraid.  Be very afraid.

Pessimistic

O. J. Mayo’s .018 PAWS/M does not impress at all.  I would not have advised any agent to provide him with a big screen TV or pay advances, and I would advise lottery drafting teams to stay away.  As a slight ray of hope, Mayo did seem to improve in the second half, which is more than I can say about…

Eric Gordon.  Win Scores pegs Gordon as a net drain on his college team which got worse as the season progressed.  Save your NBA team millions of dollars and do not draft Eric Gordon.

Furthermore, teams should stay away from Anthony Randolph, given his poor .005 PAWS/M and horrific -.044 PAWS/M against the top 100 college teams. 

No Guarantees: Beyond the Projected First Round

Past the projected top 30, there are a handful of Win Score favorites that appear likely to pleasantly surprise their new organizations.  Richard Hendrix, Chris Lofton, and Joey Dorsey all appear to be excellent NBA prospects per Win Score.  These gentlemen are coming out of strong programs yet seem underappreciated by the scouting majority. 

While Draft Express’s scouting report on Richard Hendrix is positive, his mock draft spot winds up south of the first round.  Hendrix’s body of work is solid, posting consistently high PAWS/M regardless of competition (marks of .143 vs top 25, .162 vs top 100, and .139 vs NCAA65).  To lock this guy up for 4 years would be a solid investment for any NBA team.

This year, Tennessee guard Chris Lofton posted an initially unimpressive .052 PAWS/M, though that comes with an asterisk.  Lofton’s biggest opponent this year was not on the Volunteer’s schedule, it was cancer.  Lofton’s story was kept quiet until recently and all signs point to a full recovery.  Prior to this season, Lofton has consistently put up high marks, posting PAWS/M’s of .134, .124, and .145, which put him in similar territory to Brandon Roy’s 2006 .166, Rodney Stuckey’s 2007 .123, and Rajon Rondo’s 2006 .123.  While Lofton’s measureables may come up short against the other three, he certainly seems to be undervalued by mock drafts as some even have him going undrafted.  Despite his shortcomings, Lofton clearly seems to be worth more than a second round pick.

Finally, Memphis’s Joey Dorsey just plain puts up amazing Win Scores.  Looking at his history, you’ll understand why I sighed a big breath of relief as Dorsey fouled out of the National Title Game.  Dorsey, not Derrick Rose, was the engine that drove the Memphis Tigers through the regular season, and both players stepped up for a fantastic tournament run.  Rose may go 30 picks earlier, but Dorsey can well prove to be the better value given the cheap projected price of a 2nd rounder.

Outside the big programs, there are three smaller school guards – Lester Hudson, J.R. Giddens, and George Hill — that are intriguing, though, the case for a full-fledged endorsement is hampered by a small sample size against serious competition.

Tennessee-Martin guard Lester Hudson posted great games against top competition, but the sample size was a woeful 1 game against the top 25 and 5 games against the top 100 NCAA teams.  (PAWS/M .254 Top 100, .203 all games)

New Mexico guard J.R. Giddens destroyed weaker competition in gathering a PAWS/M of .205 in 2007, but had no games against the top 25 and posted a poor .058 in WS/M his junior year.

IUPUI’s George Hill posted a solid .183 PAWS/M and a .108 PAWS/M in his 5 games against top 100 competition.  Furthermore, Hill posted a decent .119 PAWS/M in his freshman year, showing 2007-2008 may not be a fluke.

And finally, let me toss out the following one-liners

NCAA scoring leader Reggie Williams posted a strong pace-adjusted .209 PAWS/M, but played a mere one game against a top 100 team. 

Nebraska center Aleks Maric came on strong with a .192 PAWS/M in his senior campaign along with two monster games against Kansas State. 

Rider’s Jason Thompson was the NCAA’s 2nd place finisher in rebounding (after Beasley) and offers an impressive .188 PAWS/M, though, most of that comes against weak competition.

Beasley & Love versus Durant & Oden

Win Scores may be optimistic on Beasley and Love, but how do they measure up historically?  To answer this, I dug up last year’s numbers and ran a quick comparison between Win Scores favorite big men, 2007 versus 2008.  The results are reported in Table 4 (see above).

It appears Beasley and Love grade out solidly higher than Oden & Durant.  In general, the class of 2008 generally can claim higher Win Scores than the class of 2007.  Maybe the 2009 All-Star Rookie-Sophomore Challenge will be competitive for once…

Thanks for reading, and check back here before the draft for more! 

-Erich Doerr

The WoW Journal Comments Policy

The WoW research on the NBA was summarized HERE.

The Technical Notes at wagesofwins.com provides substantially more information on the published research behind Wins Produced and Win Score

Wins Produced, Win Score, and PAWSmin are also discussed in the following posts:

Simple Models of Player Performance

Wins Produced vs. Win Score

What Wins Produced Says and What It Does Not Say

Introducing PAWSmin — and a Defense of Box Score Statistics

Finally, A Guide to Evaluating Models contains useful hints on how to interpret and evaluate statistical models.

Comments are closed.