In describing the last three seasons of Utah basketball one would not be tempted to use the word “progress.” In 2006-07 the Jazz finished with 51 victories and an appearance in the Western Conference Finals. Two years ago the team improved to 54 wins, but was bounced in the second round. Last year, the Jazz regressed to 48 wins and a first round exit in the playoffs.
Such a record would suggest changes would need to be made if the Jazz are hoping to reverse course. This suggestion, though, was ignored by powers-that-be in Utah. This past summer the only change Utah made to its roster was adding guard Eric Maynor in the draft. Every other player listed on this team’s roster at ESPN.com was on the team in 2008-09.
Given that the Jazz are returning the same roster, should fans of this team expect the same outcome?
To answer that question, let’s first examine the outcome from last season. Table One reports the Wins Produced and WP48 [Wins Produced per 48 minutes] for each player employed by the Jazz last season.
The Jazz employed five above average players last season (average WP48 is 0.100): Paul Millsap, Deron Williams, Carlos Boozer, Ronnie Brewer, and Andrei Kirilenko. Despite missing a combined 81 games this quintet still managed to produce 39.4 wins.
In the off-season Millsap was signed to an offer sheet by the Portland Trail Blazers. When Utah matched this offer it was expected that Boozer – who plays exactly the same position as Millsap – would be sent elsewhere. At the start of training camp, though, both Boozer and Millsap are still employed in Utah.
But who is the starter? At the moment, ESPN.com lists Boozer as the starter. So the depth chart for Utah would be as follows (with 2008-09 Win Produced and WP48 reported]:
Potential First String
PG: Deron Williams [10.9Wins Produced, 0.209 WP48]
SG: Ronnie Brewer [8.5 Wins Produced, 0.156 WP48]
SF: Andrei Kirilenko [5.2 Wins Produced, 0.138 WP48]
PF: Carlos Boozer [3.9 Wins Produced, 0.158 WP48]
C: Mehmet Okur [4.1 Wins Produced, 0.082 WP48]
Potential Second String
PG: Ronnie Price [-0.4 Wins Produced, -0.024WP48]
SG: C.J. Miles [1.3 Wins Produced, 0.038 WP48]
SF: Kyle Korver [3.2 Wins Produced, 0.081 WP48]
PF: Paul Millsap [10.9 Wins Produced, 0.229 WP48]
C: Kosta Koufos [0.4 Wins Produced, 0.031 WP48]
In addition to these ten players, the Jazz also have the services of the following players:
PG: Eric Maynor (Rookie, above average point guard last year in college)
SF: Matt Harpring [0.4 Wins Produced, 0.030 WP48]
C: Kyrylo Fesenko [-0.1 Wins Produced, -0.017 WP48]
Again, this team suffered some significant injuries last year, primarily to Carlos Boozer. Not only did Boozer miss games, his per-minutes performance also declined. Prior to last season Boozer’s career WP48 stood at 0.266. Had Boozer matched that performance last year the Jazz would have won 51 games and entered the playoffs as the 6th seed (as opposed to the 8th seed). And had Boozer been available the entire season – and maintained his career average – the Jazz might have won more than 55 games and taken the second seed. In sum, a healthy Boozer would have made a huge difference in 2008-09.
Understanding Boozer’s injuries gives us insight into the plan in Utah. If Boozer is healthy (and yes, that’s a big if), Utah can be one of the top teams in the West. Of course, there’s the issue of who gets to start at power forward? And perhaps more importantly, can Boozer and Millsap play together? If not, then the Jazz will always have a very productive player sitting on the bench.
If Boozer and Millsap can play together, though, the Jazz can field a line-up of Williams, Brewer, Kirilenko, Millsap, and Boozer. Such a line-up – with Boozer producing at his career average – would have an average WP48 of 0.200. To put that in perspective… a team where each player averages a 0.100 WP48 will win 41 games. So if the team doubles the average WP48, the team doubles its win total. Yes, this line-up is quite good.
Of course, other players will play. But the Jazz – with everyone playing and producing as they have in the past – can be quite formidable. And that decision to stay the course will look very, very smart (and many of my colleagues at Southern Utah will be very, very happy).
The WoW Journal Comments Policy
Our research on the NBA was summarized HERE.
Wins Produced, Win Score, and PAWSmin are also discussed in the following posts:
Finally, A Guide to Evaluating Models contains useful hints on how to interpret and evaluate statistical models.