Chapter 22: Player Performance Prediction - Further Reading

Foundational Texts

Basketball Analytics Foundations

Oliver, D. (2004). Basketball on Paper: Rules and Tools for Performance Analysis. Potomac Books. The foundational text for basketball analytics. Chapter 8 on player evaluation provides essential context for understanding projection challenges. Oliver's work on possession-based statistics and the Four Factors framework underlies modern projection systems. Essential reading for anyone serious about basketball analytics.

Hollinger, J. (2005). Pro Basketball Forecast. Potomac Books. An early systematic approach to projecting NBA player performance. Hollinger's work on PER and his annual player projections demonstrate practical application of statistical methods. While methodologically dated, the conceptual framework remains relevant.

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). Academic treatment of basketball statistics foundations, including discussion of measurement challenges relevant to projection. Available online through the journal.

Statistical Methods for Prediction

Silver, N. (2012). The Signal and the Noise: Why So Many Predictions Fail—But Some Don't. Penguin Press. Nate Silver's accessible treatment of prediction methodology across domains, including sports. Chapter 3 covers his work on baseball projections (PECOTA), which shares methodological DNA with basketball projection systems. Essential for understanding prediction philosophy.

Gelman, A., Carlin, J. B., Stern, H. S., Dunson, D. B., Vehtari, A., & Rubin, D. B. (2013). Bayesian Data Analysis (3rd ed.). CRC Press. The definitive text on Bayesian statistical methods. Chapters 1-5 provide foundations essential for understanding regression to the mean, prior specification, and posterior updating in projection contexts. Mathematically rigorous but accessible to determined readers.

James, G., Witten, D., Hastie, T., & Tibshirani, R. (2021). An Introduction to Statistical Learning (2nd ed.). Springer. Excellent introduction to statistical learning methods relevant to projection modeling. Chapters on regression, model selection, and validation are directly applicable. Free PDF available at www.statlearning.com.

Projection System Documentation

Public Projection Systems

FiveThirtyEight CARMELO Projections - Documentation: https://fivethirtyeight.com/features/how-our-nba-projections-work/ - Silver, N. & Fischer-Baum, R. (2015). "How Our NBA Projections Work." - CARMELO uses player similarity and aging curves to generate probabilistic projections. The methodology documentation provides practical insight into building production-quality systems.

FiveThirtyEight RAPTOR - Documentation: https://fivethirtyeight.com/features/how-our-raptor-metric-works/ - Successor to CARMELO incorporating player tracking data. Documentation of the metric provides context for projection inputs.

Basketball-Reference Player Projection Framework - Sports Reference's approach to projecting player statistics, documented through their methodology pages - Demonstrates practical implementation challenges

Historical/Comparative Systems

Tango, T. M., Lichtman, M. G., & Dolphin, A. E. (2007). The Book: Playing the Percentages in Baseball. Potomac Books. While baseball-focused, Chapter 2 on regression to the mean and Chapter 10 on player projection provide methodological foundations applicable to basketball. The Marcel projection system described here influences basketball projection methodology.

Schuckers, M. E. (2011). "An Alternative to the Plus-Minus for Determining Contribution of NHL Defensemen." Journal of Quantitative Analysis in Sports, 7(3). Hockey analytics paper demonstrating cross-sport applicability of plus-minus adjustment methods used in player evaluation and projection.

Academic Research

Aging and Career Trajectories

Wakim, A., & Jin, J. (2014). "Functional Data Analysis of Aging Curves in Sports." arXiv preprint arXiv:1403.7548. Statistical treatment of aging curve estimation using functional data analysis. Addresses survivorship bias and provides methodology for sport-specific aging curve construction.

Fair, R. C. (2008). "Estimated Age Effects in Baseball." Journal of Quantitative Analysis in Sports, 4(1). Rigorous treatment of aging effects that applies conceptually to basketball. Fair's methodology for separating age effects from selection effects is relevant to basketball aging curve construction.

Schultz, B. & Schultz, A. (2020). "NBA Player Age Analysis: When Do Players Peak?" MIT Sloan Sports Analytics Conference. Conference paper specifically addressing NBA aging curves with modern data. Examines position-specific and skill-specific aging patterns.

Projection Methodology

Berry, S. M., Reese, C. S., & Larkey, P. D. (1999). "Bridging Different Eras in Sports." Journal of the American Statistical Association, 94(447), 661-676. Addresses era adjustment in historical comparisons, essential for similarity-based projection systems. Demonstrates statistical approaches to comparing performances across different competitive environments.

Albert, J. (2010). "Improved Season Predictions." Chapter in Anthology of Statistics in Sports. ASA-SIAM. Academic treatment of season-level prediction in baseball with general principles applicable to basketball player projection.

McCullagh, P. (2002). "What is a Statistical Model?" Annals of Statistics, 30(5), 1225-1267. Foundational paper on model specification and uncertainty quantification. Provides theoretical grounding for projection model construction.

Machine Learning Applications

Miljkovic, D., Gajic, L., Kovacevic, A., & Konjovic, Z. (2010). "The use of data mining for basketball matches outcomes prediction." IEEE 8th International Symposium on Intelligent Systems and Informatics. Early application of machine learning to basketball prediction. Useful for understanding ML approaches and their limitations.

Cao, C. (2012). "Sports Data Mining Technology Used in Basketball Outcome Prediction." Master's thesis, Dublin Institute of Technology. Comprehensive treatment of prediction modeling in basketball context, including player-level applications.

Applied Analytics Resources

Online Resources and Blogs

Cleaning the Glass (Ben Falk) - https://cleaningtheglass.com/ - Professional-quality analysis including player evaluation. Subscriber content demonstrates practical application of projection concepts.

Nylon Calculus (Various authors) - https://fansided.com/nba/nylon-calculus/ - Collection of analytical writing including projection-related analysis. Good source for applied methodology.

Positive Residual (Seth Partnow) - https://positiveresidual.com/ - Former Bucks analyst providing analytical perspective including player evaluation frameworks.

Thinking Basketball (Ben Taylor) - https://www.youtube.com/c/ThinkingBasketball - Video analysis demonstrating integration of statistical and film-based evaluation relevant to projection.

Data Resources

Basketball-Reference - https://www.basketball-reference.com/ - Comprehensive historical statistics essential for projection model building

NBA Stats Official - https://www.nba.com/stats/ - Official NBA statistics including advanced tracking data

PBP Stats - https://www.pbpstats.com/ - Detailed play-by-play derived statistics

Technical Implementation

Programming and Tools

McKinney, W. (2017). Python for Data Analysis (2nd ed.). O'Reilly Media. Essential reference for data manipulation in Python. Chapters on pandas provide tools for processing basketball statistics data.

VanderPlas, J. (2016). Python Data Science Handbook. O'Reilly Media. Comprehensive treatment of Python data science tools. Available free online at https://jakevdp.github.io/PythonDataScienceHandbook/

Wickham, H. & Grolemund, G. (2017). R for Data Science. O'Reilly Media. R-focused data science introduction. Relevant for analysts working in R environments. Free at https://r4ds.had.co.nz/

Statistical Software

Stan Development Team (2021). Stan User's Guide. Documentation for Stan probabilistic programming language, useful for Bayesian projection models. https://mc-stan.org/users/documentation/

scikit-learn Documentation - https://scikit-learn.org/stable/documentation.html - Essential reference for machine learning implementations in Python

Conference Proceedings

MIT Sloan Sports Analytics Conference

Annual proceedings available at: - https://www.sloansportsconference.com/

Notable player projection papers: - Various years include papers on player valuation, aging curves, and projection methodology - Research competition submissions often address projection-related questions

SABR Analytics Conference

While baseball-focused, methodology papers often apply to basketball: - Projection system comparisons - Aging curve methodology - Uncertainty quantification

For Beginners

  1. Oliver (2004) - Basketball on Paper
  2. Silver (2012) - The Signal and the Noise
  3. CARMELO documentation
  4. James et al. (2021) - Statistical Learning (Chapters 1-6)

For Intermediate Practitioners

  1. Tango et al. (2007) - The Book (applicable chapters)
  2. Gelman et al. (2013) - Bayesian Data Analysis (Chapters 1-5)
  3. Academic papers on aging curves
  4. FiveThirtyEight methodology articles

For Advanced Analysts

  1. Complete Gelman et al. (2013)
  2. Research on functional data analysis for aging
  3. Advanced machine learning texts
  4. Original research paper development

Staying Current

Newsletters and Updates

  • The Athletic NBA coverage (subscription)
  • FiveThirtyEight sports section
  • Cleaning the Glass newsletter
  • Academic journal alerts (JQAS, JSE)

Social Media Follows

  • Active basketball analytics community on Twitter/X
  • NBA team analytics staff members
  • Sports analytics researchers

Conference Attendance

  • MIT Sloan Sports Analytics Conference (March)
  • SABR Analytics Conference
  • Local sports analytics meetups