Much has been written about Derrick Rose‘s importance to the Chicago Bulls, and whether the team is too reliant on the young MVP. Much less has been written (this season, at least) on that same topic in relation to Kobe Bryant and the Los Angeles Lakers.
Yet, often when talking about how central a player is to a team’s offense, we look to usage rate* as a measurement of involvement, and on that score, Bryant seems more important to the Lakers than Rose does to the Bulls: the NBA’s 2010-11 usage rate leaders were Bryant (35.1%), Rose (32.2%), and Carmelo Anthony (32.0%). However, in Rose’s case, I believe usage rate underestimates how vital he is to the Bulls, for three reasons.
*[Usage rate is an estimate of the percentage of plays (FGA + 0.44*FTA + TOV) used by a player while he is on the floor.]
1) Usage Rate Doesn’t Include Assists
Usage rate only counts plays that were directly ended by a player via field goal attempts, free throw attempts, and turnovers. But that means it ignores Rose’s role as the Bulls’ primary distributor. To account for that, we can look at assists. Rose’s percent of assists while on the floor was 44.7%, compared to 30.3% for Bryant, and 17.7% for Anthony. OK, but what does that mean in terms of plays used?
About 21.2% of Chicago plays ended with an assisted field goal. Multiply that by Rose’s 44.7% “assist usage rate” and we can estimate that besides the 32.2% of plays he directly ended, he made the assist that led to the end of about 9.5% more plays. Compare that to 6.4% for Bryant and 3.6% for Anthony.
2) Usage Rate Ignores Minutes Played
A player can have a high usage rate in very few minutes played. Imagine a player that only played 2 minutes per game, but shot the ball every single time down the court. He’d have a usage rate of 100%, but we would be foolish to say he’s more important to his team than Rose is to Chicago. To account for this, we can look at minutes played.
Rose played 37.4 minutes per game. Bryant averaged 33.9, and Anthony averaged 35.7. We can divide these by 48, multiply by the player’s usage rate, and find the percentage of a team’s plays a player used per game. That calculation gives us: Rose 25.0%, Bryant 24.8%, Anthony 23.8%.
3) Usage Rates Considers Both Good And Bad Plays
Remember, usage rate is an estimate of the percentage of plays (FGA + 0.44*FTA + TOV) used while a player is on the floor. But we can split plays into two basic types: scoring ([FG made] + 0.44*[FT made]) and non-scoring ([FG missed] + 0.44*[FT missed] + TOV). Using those definitions, we can calculate “positive” and “negative” usage rates.
Rose’s positive usage rate is 32.6%, and his negative usage rate is 31.8% (+32.6/-31.8). For Bryant, those values are +35.7/-34.6, and for Anthony +33.1/-31.1. Admittedly, these are a small differences, and they actually penalize Rose more than Anthony. The point here is not that usage rate underestimates the contribution of Rose relative to other stars, but that it underestimates all above-average players relative to below-average contributors. For example, while Charlotte’s Stephen Jackson is 14th in the NBA in usage rate at 27.4%, he has positive and negative usage rates of +25.4/-28.8, meaning he is quite a bit more involved in unproductive Bobcat plays than he is in productive ones.
Putting It All Together
Combining the three ideas above, we can create a simple metric that estimates the fraction of a team’s scoring plays in each game that involve the player in question.
First, using the definition of scoring plays above, and adding in assists, we can calculate the number of scoring plays a player contributes to per game:
([FG made] + 0.44*[FT made] + Assists)/Games
Then, the team’s scoring plays per game is given by:
([Team FG made] + 0.44*[Team FT made])/Games
Finally, we just divide the player’s value by the team’s value. The calculation for Derrick Rose looks like this:
( ( [711 Rose FG made] + 0.44*[476 Rose FT made] + [623 Rose Assists] ) / [81 Rose Games] )
/ ( ( [3042 Bulls FG made] + 0.44*[1492 Bulls FT made] ) / [82 Bulls Games] ) = 42.2% Scoring Involvement
So, in a typical Bulls game, on 42.2% of Chicago’s scoring plays Derrick Rose either does the scoring himself or gets the assist.
Just how high is 42.2%? Pretty freaking high, at least for a team with designs on the winning the 2011 NBA Playoffs.
I looked at every NBA champion since the merger (1976-77 season), and found the player with the highest scoring involvement. I also jotted down their age, and who their best teammate was. Here are the 15 championship teams with the most involved leaders, along with a few still-alive 2011 teams.
Here’s what jumps out at me from the above table:
- 2011 Derrick Rose would be second on the list, behind only 2006 Dwyane Wade (who got to play with Shaq).
- Every young player (under the median age of 27) had an all-time great big-man as a sidekick (either Shaq or Kareem Abdul-Jabbar). Rose has … Boozer? Deng?
- Regardless of age, 13 of the top 14 seasons are from Michael Jordan, Magic Johnson, Kobe Bryant, or Shaquille O’Neal. The other was Dwyane Wade, who played WITH Shaq.
What this says to me is that if Rose is able to win a championship this year, it’ll either be a massive outlier on this list, OR will look reasonable in retrospect after he goes on to become one of the all-time greats. Now, that exact same logic applies to both LeBron James and Russel Westbrook as well. But of those three, who is most likely to end up viewed as one of history’s best? LeBron is about 2 rings from that status already.
I think it goes almost without saying that this isn’t meant to be predictive, but let me say it anyway: this isn’t meant to be predictive. (And I wouldn’t place a bet based on this info.) This is more about providing context for Rose’s attempt to lead the Bulls to a title. The take away is that if he succeeds, it will be a pretty remarkable accomplishment, perhaps remarkable enough that on this evidence alone, we’d have to start considering the possibility that Rose will end up as an inner-circle Hall-of-Famer.