Starting Point for Analyzing Basketball Statistics" by Kubatko, Oliver, Pelton and Rosenbaum (2007)

Concept:

Like almost all other sports, basketball is a game of numbers but there has been a dearth of peer reviewed research concerning the basic quantifiable variables such as per-minute statistics, offensive and defensive ratings, plays, true shooting percentage, pace adjustments, effective field goal percentage, and rebound rates, among others, that are used to measure player and team performance in terms of efficiency with the basketball.

Based on the centrality of the amount of time players have possession of the basketball during games, Kubatko and his associates conducted a comprehensive analysis of this data over the course of four seasons to provide a benchmark for future research in this area. The analytical approaches described by these researchers included plus/minus statistics, Four Factors, linear weights metrics, counterpart statistics, individual possession usage, individual efficiency, the Bell Curve method and the Pythagorean method, as well as their respective strengths and weaknesses for analyzing basketball team efficiency. For their analysis, Kubatko et al. (2007) also employed a series of formula to estimate possessions based on the available seasonal box score data.

Results:

The results...
...

By contrast, the points scored and allowed per game are valid indicators of a basketball team's efficiency as well as the pace at which a given game was played; and efficiency with the ball was found to be the optimal measure of the quality of a team's defense and offense.

Summary:

As noted above, the overarching purpose of this study was to help fill a peer-reviewed gap in the existing body of knowledge concerning time of possession of basketball teams because this metric is among the more valid predictors of team efficiency. For example, Kubatko and his colleagues (2007) point out that, "Possession estimates have existed for approximately two decades, but the various formulas have sometimes created confusion" (p. 20). The systematic analysis of this and other metrics using the analytical tools described above showed that are more valid approaches to calculating possession time and basketball team efficiency available.

Discussion:

Given the exploratory research design used by Kubatko et al. (2007), there was no basic question being asked…