In Defense of Marbury: Usage Rate and the Ball-Hog
This article was written by KnickerBlogger reader Michael Zannettis, who originally sent this to me in a different form. I’ve taken the liberty to edit the work to make it more befitting this space. Any grammatical or spelling mistakes are therefore mine. Additionally I sat on this article for well over a week, so I’ve attempted to update the stats where applicable. At the time the Knicks were doing much worse on offense and much better on defense, so it may suffer from it’s late publication. While it may not be best presented after Marbury’s 35 point outburst against Detroit, I feel that the piece should be heard, and believe that it still stands on its own.
Again any issues that arise from these changes are the fault of the editor, so save Michael from the voodoo doll pins that many of you are currently using to punish yours truly.
It is with great concern for the current competitiveness and future viability of our beloved Knickerbocker franchise that I have become distressed with the treatment of the Knicks? best player, Stephon Marbury. I was certainly a supporter of one of the great coaches of all-time, Larry Brown, being hired to lead this franchise back to the NBA playoffs, but his initial returns on player development are frustrating.
Most obviously, Mr. Brown?s poor treatment of his only star performer, Stephon Marbury, has collapsed a once-decent New York offense. I will not forget that the defense has made a turn around from 27th to 16th without the addition of even one frontline defensive player. Due credit will be meted out in time, but even Mr. Brown?s championship Detroit Pistons had an average offense to complement their superior defense.
Let the numbers decide the offense?s stature. Last year’s team was an average offense, ranking 17th (105.9pts/100 poss). This year, the team’s production has plummeted to 24th (100.7pts/100 poss), a decrease of 5.2pts/100 poss. Last year, the entire scoring load of the Knicks’ offense fell to its guards, Stephon Marbury and Jamal Crawford, both of whom were the only Knicks to average 15 points or more per game.
In the more advanced metrics, Marbury fared well while Crawford looked worse. The former led the team with a .690 player win percentage and a 21.9 PER, while the latter was an inefficient, if volumous, scorer with poor defense, whose win share was a replacement-level .344 with an average 15.4 PER. It was Crawford’s low shooting percentage, often forcing ill-advised shots, which killed his contributions to the offense. He didn’t help his cause by being a spectator on the defensive end either.
Marbury was criticized for his high Usage Rate (24.7%) but considering the teammates he was expected to pass to, it’s a wonder that he didn’t shoot the ball even more often. The only other Knick regular with Offensive Efficiency ratings at the league average or better were Jerome Williams and Mike Sweetney. Williams was a rebounding specialist whose 13.2% Usage Rate belied the fact that the only time he scored was off a tip-in or offensive rebound. Meanwhile the underrated and underutilized Mike Sweetney was a low-post scorer with a prodigious free throw rate, who neither was a pick & roll partner nor a particularly explosive finisher around the basket that would complement Mr. Marbury?s talents.
No other teammate, besides the underused Sweetney, approached offensive competence.
If the criticism of Marbury was that Sweetney should have received more touches in the low block, then I would certainly be in agreement. This was not the case. Rather, it was the Knicks’ coaching staff themselves who limited Big Mike’s production by playing him only 19.6 minutes per game, and all that behind inferior talent.
Sweetney even played less when Malik Rose joined the team through a mid-season trade with San Antonio. Rose was clearly finished as an offensive player, and shared Sweetney’s biggest weakness: being short. If Mike Sweetney was losing playing time for being an undersized 6?8? power forward with limited open court athleticism, then what exactly was Malik Rose, an undersized 6?7? power forward who could no longer hit a jump shot doing playing over 20 minutes a game?
Who then was Marbury expected to share the ball with? Because of his incompetent teammates, on any given possession the best option for the Knicks was Stephon taking the shot.
Examining the Usage Rate comparables of the 2004-05 season makes the ball-hog criticism even more inane. Marbury ranked 30th in the league in Usage Rate. For those of us keeping score at home, there are only 32 teams in the league, and since Usage Rate cannot exceed 100% by a team, the more one player’s ratio increases the more another’s must decrease (although if they are playing as substitutes one might not affect another’s directly). An obvious example would be Chris Webber?s high usage rate falling precipitously when he was traded to the Philadelphia 76er?s midseason. Allen Iverson led the league in Usage Rate, and promptly cut Webber?s dramatically, before it stabilized to a 17% decrease from his Sacramento rate.
As Dean Oliver explains in Basketball on Paper, a high usage scorer could be an asset to a team even if he is below average efficiency, because it permits his teammates to take fewer but higher quality shots. This improves the team?s overall efficiency. Ideally, of course, a team would like several high usage/high efficiency scorers. The Chicago Bull dynasty had this with the ultimate example of Michael Jordan and his Top-50 Player of all Time teammate Scottie Pippen. By the time they were done using the ball, the remainder of their teammates had only to use a small high quality percentage, which improved the team?s overall efficiency even more.
Compared to Stephon Marbury?s much maligned 24.7% Usage Rate, Scottie Pippen’s usage during the Bulls? championship seasons is comparable: 1990-91, 23.2%; ?92, 25.8%; ?93, 25.4%; ?96, 24.4%; ?97, 24.1%; ?98, 21.4%. And what about the one year Pippen had to lead the Bulls completely without Jordan? His Usage Rate in ?94 was 27.4%. Obviously, he should have passed the ball more. What a hog!
This additive function of Usage Rates would make it extremely difficult for any two teammates to be near the league leaders in Usage Rate, unless it was a classic pairing like Shaquille O’Neal and Kobe Bryant in Los Angeles during their championship run, or the aforementioned Jordan-Pippen combination. For the 2004 season, there were only four teams that had a pair of teammates who played with each other all season and both had a higher usage rate than Marbury’s 24.57%, 30th: Indiana Pacers (Jermaine O’Neal 32.32%, 3rd; Jamaal Tinsley 26.12%, 17th; Stephen Jackson, 25.03%, 29th); Miami Heat (Dwyane Wade, 29.03%, 5th; Shaquille O’Neal, 27.45%, 10th); Washington Wizards (Gilbert Arenas, 25.95%, 18th; Larry Hughes, 25.30%, 24th); Minnesota Timberwolves (Kevin Garnett, 25.88%, 19th; Sam Cassell, 25.86%, 20th). In other words, as the best player on the Knicks’ team Marbury only used the ball as much as the 24th most heavily used first-rate player. Including the 76ers and Warriors by projecting full season stats and therefore including Iverson/Webber and Davis/Richardson, only moves Stephon up to 22nd, still a below average rate for a team?s best player.
If Marbury’s reputation labels him a ball-hog, the statistical evidence does not support the hypothesis. Instead, he resembles a talented offensive player who creates his own shots and creates high quality efficiency. Teams need more of this, not less. Considering the Knicks general incompetence at the offensive end, it was a wonder that Stephon Marbury is not asked to increase his Usage Rate.