Reactions to this week's news that Mike Brown will replace Phil Jackson at the helm of the Los Angeles Lakers have run the gamut. Brown remains something of a punchline dating to the 2007 NBA Finals, when his Cleveland Cavaliers team had no Plan B on offense after LeBron James and was swept by the San Antonio Spurs on the league's biggest stage. When the Cavaliers reemerged as one of the league's top teams behind elite units at both ends of the floor, Brown was rewarded as Coach of the Year, and a reevaluation of Brown's reputation began despite the fact that he was fired last year.
So which coach is the real Brown, the defensive specialist who rode James to offensive respectability, or a top defensive mind who tailored his offense to the talent available? Evaluating coaches remains one of the most difficult tasks for statistics, but there is some evidence that can help differentiate myth from reality when it comes to the Lakers' new coach.
During Brown's first season in Cleveland, counterintutively, the Cavaliers improved at the offensive end and not noticeably on defense. The Cavs' D went from 0.6 percent better than league average on a per-possession basis to 0.8 percent better than average under Brown, while the offense took a larger jump from almost exactly average (+0.1 percent) to 1.7 percent better than average. The change can largely be traced to James' development in his third NBA season. He increased his usage rate from 29.7 percent of Cleveland's plays to 33.6 percent without sacrificing efficiency.
It was during Brown's second season as coach, which resulted in the unexpected trip to the Finals, that Cleveland earned its defense-first reputation. The Cavaliers allowed 4.7 percent fewer points per possession than league average, going from 14th in the NBA to fourth, but the offense slipped below average. With Zydrunas Ilgauskas aging, all of Cleveland's top five options on offense save James posted below-average True Shooting Percentages.
The Cavaliers repeated their ranking of 19th in the league in offensive rating in 2007-08, but things changed dramatically the following season. With Mo Williams giving the team more scoring punch from the point and a full season from Delonte West and Wally Szczerbiak, Cleveland's offense was less LeBron-centric. At the same time, James took his game to the next level, winning MVP for the first time.
Brown delegated offensive duties to assistant John Kuester that season, and Kuester naturally got much of the credit for the Cavaliers' improvement. Parsing out Kuester's role is impossible from the statistics, though it is instructive that Cleveland repeated its fourth-place finish in per-possession scoring in 2009-10, with Michael Malone stepping into Kuester's role as offensive coordinator. During Brown's last season with the Cavaliers, just as his first, the team was actually better on offense than on defense.
Cleveland's roster changed so dramatically last summer that any comparison to 2010-11 is unfair to Byron Scott. Beyond James' departure, the Cavaliers lost three other rotation players and saw defensive stalwart Anderson Varejao suffer a season-ending foot injury in early January. Looking strictly at lineups that featured only returning players, Cleveland's defensive rating was similar to what it had been in 2009-10. Yet the Cavaliers' rookies and other replacements were so bad defensively than the Cavaliers slipped to 29th in the NBA in per-possession defense. Without James, Cleveland was also 29th at the other end of the floor.
Adjusting for personnel
The problem with any team-based analysis is that it's impossible to determine what is coaching and what is simply the product of the players on the floor. That's where an innovative new technique comes in. Adjusted plus-minus ratings have been used to evaluate players for nearly a decade, with mixed results. It's possible to consider coaches using the same method by considering them essentially a sixth player on the floor. To the extent a lineup performs better or worse than expected based on its results over several seasons, a coach can be credited for the difference.
Researcher Jeremias Engelmann has used data from a five-year period to evaluate coaches by adjusted plus-minus. The results generally pass the laugh test, though they are not especially kind to Jackson (at +0.7 points per 100 possessions, not much better than average).
The numbers are more favorable for Brown. Accounting for the Cavaliers' lineups, this method suggests Brown improves his team by 1.5 points per 100 possessions, ranking him 12th among all coaches over this period. Most of Brown's success is indeed traced to defense, where his coaching is responsible for shaving 1.2 points per 100 possessions off his team's defensive rating, but Brown comes out above average on offense as well.
Coaching adjusted plus-minus isn't nearly reliable enough to compare Brown to Jackson, or most other coaches, with any kind of confidence. What it does say with some certainty is that Brown has a positive influence as a coach and is not the kind of offensive disaster he was once considered.
In a more general sense, Engelmann's results confirm the conventional wisdom that coaches make a greater difference at the defensive end of the floor. Coaches' ratings, both positive and negative, are much larger on defense than they are on offense. That suggests that it is wise to favor defensive-minded coaches over their offense-first peers. A team's success on offense depends largely on personnel, as Brown's track record in Cleveland shows. Talent matters on defense, too, but Brown has shown the ability to consistently mold defenses better than average, no matter what kind of lineups he's given.
From that standpoint, the Lakers' future is less likely to be decided by who's on the sideline and more by an issue we've already addressed -- whether Kobe Bryant and company can avoid the impact of their advancing age.
A version of this story originally appeared on ESPN Insider .
Kevin Pelton is an author of Basketball Prospectus.
You can contact Kevin by clicking here or click here to see Kevin's other articles.