Did Minnesota Hire the Right Rambis?

Posted on August 9, 2009 by


Kurt Rambis has left the World Champion LA Lakers to be the head coach for the Minnesota Timberwolves.  Marc Stein of ESPN.com reports the following:

Long considered a potential heir apparent to Los Angeles Lakers coach Phil Jackson, Rambis was lured away from Jackson’s staff by a four-year deal believed to be worth in excess of $8 million.

Wolves owner Glen Taylor confirmed ESPN.com’s report in an interview with the St. Paul Pioneer Press late Saturday, telling the newspaper that he expects Rambis to formally sign the contract Monday.

“Kurt, with his experience playing and coaching for a winning organization and working under people like Pat Riley, will bring a lot to our organization,” Taylor told the Pioneer Press. “In meeting with him and talking to him, I know he’s very excited about having an opportunity to help build his own team.”

We don’t know what kind of team Rambis will build in four years.  We do suspect, though, it will be a better team than what he’s inheriting.  Here is what the veteran players Minnesota will employ in 2009-10 did last season (according to the roster reported at ESPN.com, WP48 = Wins Produced per 48 minutes):

Ryan Gomes: 2,614 min., 0.6 Wins Produced, 0.012 WP48

Kevin Love: 2,048 min., 8.9 Wins Produced, 0.208 WP48

Quentin Richardson: 1,894 min., 3.9 Wins Produced, 0.098 WP48

Al Jefferson: 1,836 min., 6.1 Wins Produced, 0.160 WP48

Darius Songaila: 1,521 min., -0.8 Wins Produced, -0.024 WP48

Brian Cardinal: 909 min., -0.6 Wins Produced, -0.033 WP48

Damien Wilkins: 626 min., -0.6 Wins Produced, -0.046 WP48

These seven players are the only players who logged at least 500 minutes last season.  The team also has Chucky Atkins (-0.120 WP48 in 413 minutes), Corey Brewer (0.053 WP48 in 307 minutes), Oleksiy Pecherov (-0.014 WP48 in 277 minutes), and Bobby Brown (-0.045 WP48 in 256 minutes). 

The Wins Produced of all these players only sums to 16.5.  The team has also added Jonny Flynn and Wayne Ellington (and maybe Ricky Rubio) in the draft.  Of these, Flynn is most likely to play significant minutes in 2009-10.  Flynn’s performance in college, though, suggests he may not be very productive next year (although that’s just a suggestion, college performance does not predict perfectly).

All of this suggests that the Timberwolves might benefit more from hiring Rambis the player, rather than Rambis the coach.  Of course, Minnesota would have to turn back time to get Rambis the player.  If they did, though, it would help.  How much would it help? Before we answer this question, let’s ask another:  How many veteran players currently on Minnesota’s roster posted a higher WP48 last year than what Rambis did for his entire career?

Rambis played fourteen seasons in the NBA (from 1981-82 to 1994-95).  Across these seasons he produced 68.2 wins in 16,299 minutes.  This results in 0.201 WP48.  Only Kevin Love produced at a higher rate last year.  So if the T-wolves could have Rambis the player, the team would be improved.

But even with Rambis, this team is still going to struggle.  There are simply too many below average players; and too many of these are in the negative range.  So it’s a good thing Rambis has a four year contract.  In four years this roster should change dramatically.  And it better. Because if it doesn’t, the head coaching career of Rambis is going to be quite short, and he might wish he would have just waited for Jackson to retire in LA.

– DJ

The WoW Journal Comments Policy

Our 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.