Measuring the Quality of Basketball in the NBA

Posted on 07/17/2010 by


Note: I know I said I was doing a post on Regression and then the Knicks but my muse really is insistent and hates schedules.Those posts are coming though.

Basketball pundits spend a lot of time arguing about the quality of one NBA season versus another. We hear about the 80’s being a golden age in the game after the NBA/ABA merger. About the too much, too soon era in the late 90’s.  We hear about the league being better than ever now. We as fans accept a lot of these observation as true simply because they match our subjective experience of having watched those games.

But human beings tend to romanticize the past and it led me to wonder if we were remembering correctly. Were the 80’s really a renessaince for basketball? Were the late 90’s a lost generation? Is there a way to measure and evaluate this objectively?

To create an objective measurement of the quality of basketball in the NBA, a good place to start is  how we measure the quality or value for each player productivity (WP48 and Wins Produced, for an explanation go here). Our Measures for player productivity WP48 (Wins produced by player per 48 minutes played) and Wins Produced can be broken as follows:

WP48= ADJP48(Player’s Productivity per 48 minutes)- Position Adjustment(Average of all Player’s Productivity per 48 minutes for players playing the same position) + .100 (The average productivity guaranteeing 41 wins)

Wins Produced = WP48*Minutes Played = Player’s Productivity – All Player’s at Same position Productivity+ .100 *MP

In a previous post (The Talent Pool and Marginal Value in the NBA) I took a look at the change in average production by position over time (See below, ) :

The Position Adjust is meant to represent the average production by season at each position. So using the changes in the position adjustment by position we should be able to come up with a metric to determine the quality of basketball being played in a particular NBA season. My proposed Metric is as follows:

Average Team Productivity(season) = Sum of all Position Adjustments (season)

This simple metric will give us an idea of how good or bad the quality of basketball being played was for a particular season. So If I calculate the Team Productivity It looks as follows:

So by the simple expedient of calculating Average Team Productivity, I can objectively conclude:

  • 85 thru 92 was the best basketball the NBA has seen since the merger
  • 1992 was the Apex and 1999 the nadir for Basketball quality in the NBA
  • 2008,2009,2100 represent a new era of excellence for the NBA

Stats really can do anything.

Posted in: Uncategorized