簡易檢索 / 詳目顯示
| 研究生： |
林政語 Lin, Cheng-Yu |
|---|---|
| 論文名稱： |
籃球數據致勝：技術統計對勝率的實證分析 Winning by the Numbers: A Data-Driven Analysis of Box Score Metrics |
| 指導教授： | 吳文傑 |
| 口試委員: |
孫秉宏
嚴萬軒 |
| 學位類別： |
碩士
Master |
| 系所名稱： |
商學院 - 國際經營管理英語碩士學位學程(IMBA) International MBA Program College of Commerce(IMBA) |
| 論文出版年： | 2025 |
| 畢業學年度： | 113 |
| 語文別： | 英文 |
| 論文頁數： | 44 |
| 中文關鍵詞： | 邏輯迴歸 、邊際效應 、技術統計 、P. LEAGUE+ 、勝率分析 、籃球數據分析 |
| 外文關鍵詞： | Marginal effects |
| 相關次數： | 點閱：208 下載：11 |
| 分享至: |
| 查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報 |
本研究旨在探討台灣 P. LEAGUE+ 職業籃球聯盟中，球隊的技術統計與勝率之間的關聯性。透過蒐集近兩個賽季的比賽數據，並以邏輯迴歸模型分析，包括進攻與防守端的多項統計指標如何影響球隊勝負。本研究特別強調邊際效應分析，以量化每一項統計變數對勝率的實質影響，例如增加一次防守籃板是否會提升勝率，或額外的一次失誤是否會降低勝率。
實證結果顯示，防守籃板對勝率的影響最大，對勝率有正向且顯著的影響；而進攻籃板與失誤則未如預期般對勝率產生顯著的負面關係。另一方面，二分球與三分球命中數皆顯著提升勝率，突顯出得分效率在現代籃球中的重要性。此外，對手的防守籃板與失誤數也展現出可觀的影響力，雖然其中部分變數未達統計顯著水準。
本研究不僅驗證了部分傳統籃球觀念，也挑戰了一些常見假設，如助攻數未必正向關聯勝率。此研究提供教練團具體的數據依據，作為訓練設計、戰術部署與賽前準備的參考，並可應用於系列賽對戰分析及球季成效追蹤。
This thesis investigates the relationship between box score statistics and team win probability in Taiwan's professional basketball league, P. LEAGUE+, using game data from two full seasons. Focusing on a single team’s performance, the study applies logistic regression and marginal effects analysis to evaluate how offensive and defensive metrics contribute to winning outcomes. The results reveal that defensive rebounds, three-point field goals, and two-point field goals are the most significant predictors of victory. Surprisingly, offensive rebounds and turnovers did not show a statistically significant effect, and assists were negatively associated with winning. These findings challenge conventional coaching assumptions and offer practical implications for strategy and training. By emphasizing high-efficiency scoring and possession control—especially through defensive rebounding—teams can meaningfully increase their chances of success. The thesis also proposes a data-informed feedback loop for continuous performance monitoring and outlines how these methods can be extended for playoff preparation and future opponent-specific analysis. This work contributes to the growing field of basketball analytics by adapting proven methods to a regional league context and translating insights into actionable coaching decisions.
1. INTRODUCTION 1
1.1. Research Background 1
1.2. Motivation and Purpose 2
1.3. Research Questions 3
1.4. Methodology Overview 4
1.5. Limitations of Existing Literature and Non-NBA Contexts 5
1.6. Relevance of This Study to P. LEAGUE+ 6
1.7. Contribution of This Study 6
1.8. Thesis Framework 7
2. LITERATURE REVIEW 9
2.1. Foundations of Basketball Analytics 9
2.2. Expanding Metrics: Win Shares, Box Plus/Minus, and RAPM 9
2.3. Machine Learning and Predictive Modeling 10
2.4. Shot Selection and Efficiency Analysis 11
2.5. Lineup Optimization and In-Game Strategy 12
2.6. Application to Coaching and Team Management 13
3. METHODOLOGY 14
3.1. Research Objective 14
3.2. Research Hypotheses 14
3.3. Model Specification 17
4. RESULT 25
4.1. Descriptive Statistics 25
4.2. Logistic Regression Results 27
4.3. Odds Ratio Analysis 29
4.4. Marginal Effects Analysis 30
5. CONCLUSION 34
5.1. Summary of Findings and Hypotheses Evaluation 34
5.2. Practical Takeaways and Future Application 36
5.2.1. Defensive Rebounding Must Be a Non-Negotiable 36
5.2.2. Emphasize Scoring Efficiency over Volume 37
5.2.3. Rethinking Assists: Not All Ball Movement is Efficient 38
5.2.4. Don’t Overcorrect Turnovers Without Context 39
5.2.5. Use Stats to Set Role Clarity and Player Buy-In 40
5.2.6. Build a Season-Wide Feedback Loop 40
5.2.7. Summary 41
5.3. Future Application 41
5.4. Additional Directions for Future Research 42
5.5. Closing Remarks 43
REFERENCES 44
Baumer, B. S., Matthews, G. J., & O’Neil, D. (2017). The Sabermetric Revolution: Assessing the Growth of Analytics in Baseball and Other Sports. University of Pennsylvania Press.
Csapo, P., & Raabe, D. (2019). Performance indicators in basketball: A systematic review. International Journal of Performance Analysis in Sport, 19(2), 185–202. https://doi.org/10.1080/24748668.2019.1579751
Goldman, M., & Rao, J. M. (2012). Effort vs. concentration: The asymmetric impact of pressure on NBA performance. Journal of Economic Behavior & Organization, 83(3), 602–617.
Kubatko, J., Oliver, D., Pelton, K., & Rosenbaum, D. T. (2007). A starting point for analyzing basketball statistics. Journal of Quantitative Analysis in Sports, 3(3), 1–22.
Miller, S. J. (2018). Sports Analytics and Data Science: Winning the Game with Methods and Models. FT Press.
Oliver, D. (2004). Basketball on Paper: Rules and Tools for Performance Analysis. Potomac Books.
Wooldridge, J. M. (2010). Econometric Analysis of Cross Section and Panel Data (2nd ed.). MIT Press.
P. LEAGUE+. Official box score data and game logs.
