Performance evaluation of NBA teams: A non-homogeneous DEA approach

Abstract

National Basketball Association (NBA) is one of the four major sports leagues in North America. The performance evaluation of NBA teams is an important reference for team managers. However, non-homogeneity issues exist on both inputs and outputs sides for NBA teams evaluations. For example, some teams do not have “high-level” players as inputs and some teams are not qualified to play in the playoffs with respect to “wins in the playoffs” as an output. The current paper extends the existing non-homogeneous DEA method to address the non-homogeneous structure of NBA teams in performance evaluation by splitting NBA teams into types of homogeneous sub-units. The sub-unit, in this paper, consists of empirical input subset and output subset which satisfies both atomic property and maximum property. In addition, our method yields a unique efficiency decomposition of sub-units of NBA teams in 2018–2019 season without the need for imposing any additional conditions which are needed in some prior relative researches. As a result, detailed performance improvement directions for all teams can be provided.

Keywords:

• Data envelopment analysis (DEA)
• National Basketball Association (NBA)
• non-homogeneous
• efficiency decomposition

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

The authors are grateful to the comments and suggestions by two anonymous reviewers. This research is supported by the National Natural Science Foundation of China under Grants (No. 71771074, 71601067, 71971074, 71801068); China Postdoctoral Science Foundation (No. 2017M612071); the fundamental Research Funds for the Central Universities (No. JZ2018HGTB0239).