| 摘要: | 本研究旨在利用差額變數模型(slack-based model, SBM)與非意欲產出模型對NBA(National Basketball Association,NBA) 2012-2022各賽季球隊進行經營績效與球員效率分析，將須支付豪華稅定義為抱團，研究球隊抱團對經營績效與球員效率是否有正相關。球隊方面以球團支出作為投入項，例行賽勝場數、場均淨勝分差、球團營收和場均購票率作為產出項，進以衡量2012-2022各賽季球隊的經營效率，第一階段歸納抱團球隊的經營效率是否高於非抱團球隊。第二階段計算NBA 30支球隊2012-2022賽季的平均效率值，分析抱團次數較高的球隊與平均效率值之間的關聯性，並分析平均效率值較高與較低球隊投入與產出項的差異。研究結果顯示，第一階段2012-2022賽季共330支球隊中有106支球隊效率值大於0.9，抱團球隊佔6隊；49支球隊校率值小於0.6，抱團球隊佔18隊；這表示多數抱團球隊投入大量的支出卻沒獲取相對應的產出，最終導致效率不佳。第二階段2012-2022賽季30支球隊的平均效率值以New York Knicks、Utah Jazz 與 Boston Celtics效率值最佳，抱團次數分別為3次、2次與3次，與平均球隊抱團2次相差不大，這表示球隊經營效率並不因球隊抱團次數多寡而有顯著的差異。進一步分析可以發現抱團球隊普遍經營績效反而偏低，原因為球隊支出拉高，但無相對應球團收入或勝場數。
球員方面以球員薪資與場均時間作為投入項，產出項分為意欲產出:場均得分、場均助攻、場均籃板、場均抄截、場均封阻、球員效率值與勝場貢獻值與非意欲產出:場均失誤、場均犯規。研究結果顯示控球後衛與中鋒相較於其他位置有較多具有效率的球員，進一步分析跟控球後衛的助攻數與中鋒的籃板數高於其他位置有關，球團可以視球隊方向以相對合理的薪資網羅球員。第二階段分析抱團球隊在戰績與球員效率方面是否有相關性，實驗結果顯示球隊實施抱團策略確實對球隊勝率有幫助，但由於過度的投入在球員效率方面並無顯著地幫助。本研究結果可作為球團在經營球隊時之參考。
;This study aims to employ Data Envelopment Analysis (DEA) to evaluate the operational performance and player efficiency of National Basketball Association (NBA) teams across seasons from 2012 to 2022, and to investigate whether there is a positive correlation between teams paying the luxury tax and their operational performance and player efficiency. For team analysis, the expenditure of the team is considered as the input variable, while the number of wins in regular seasons, average net points difference per game, team revenue, and average ticket sales rate are considered as output variables, to measure the operational efficiency of NBA teams during the seasons 2012 to 2022. In the first phase, the study summarizes whether teams that pay the luxury tax have higher operational efficiency than non-luxury tax-paying teams. The second phase calculates the average efficiency scores for the 30 NBA teams over these seasons and analyzes the relationship between the frequency of luxury tax payments and average efficiency scores, as well as the differences in input and output variables between teams with higher versus lower average efficiency scores. The results indicate that, in the first phase, out of 330 teams in the seasons from 2012 to 2022, 106 teams have an efficiency score higher than 0.9, of which 6 are luxury tax-paying teams; 49 teams have an efficiency score lower than 0.6, of which 18 are luxury tax-paying teams. This suggests that most luxury tax-paying teams invest significantly but do not achieve corresponding outputs, resulting in poor efficiency. In the second phase, the New York Knicks, Utah Jazz, and Boston Celtics have the best average efficiency scores among the 30 teams in the seasons 2012-2022, with luxury tax payments of 3, 2, and 3 times respectively, which is not significantly different from the average of 2 times for other teams, indicating that operational efficiency is not significantly impacted by the frequency of luxury tax payments. Further analysis reveals that luxury tax-paying teams generally have lower operational performance due to high expenditure without corresponding revenue or number of wins.
Regarding player analysis, player salaries and average playing time are used as input variables, while average scores, assists, rebounds, steals, blocks, player efficiency ratings, and win shares are considered as desirable output variables, and average turnovers and fouls are considered as undesirable output variables. The results show that point guards and centers tend to be more efficient compared to players in other positions, which is associated with higher assist numbers for point guards and higher rebound numbers for centers. Teams can consider this in recruitment strategies with relatively reasonable salaries. In the second phase, the study analyzes whether luxury tax-paying teams have a correlation in terms of performance and player efficiency. The results show that implementing a luxury tax strategy does indeed help the team′s winning rate, but due to excessive investment, it does not significantly aid player efficiency. The findings of this study can be used as a reference for team management in operating NBA teams. |
