Chapter 19: Further Reading - Player Performance Forecasting

Academic Papers

Projection Methodology

  1. "Improving Simple Models with Confidence Profiles" - Baseball Research Journal - Confidence interval construction - Sample size adjustments - Applicable to football

  2. "Age Curves and Decline Patterns in Professional Sports" - Multi-sport aging analysis - Position-specific curves - Methodology for football adaptation

  3. "Forecasting Individual Player Performance" - MIT Sloan - Machine learning approaches - Feature engineering - Evaluation methods

Sports Economics

  1. "Player Valuation in Professional Sports" - Contract evaluation - Projection accuracy importance - Economic implications

Books

Projection Systems

  1. "The Book: Playing the Percentages in Baseball" - Tango, Lichtman, Dolphin - Marcel projection system origin - Regression methodology - Transferable concepts

  2. "Big Data Baseball" - Travis Sawchik - Analytics implementation - Projection in decision-making - Organizational adoption

  3. "Mathletics" - Wayne Winston - Sports analytics fundamentals - Player evaluation - Multi-sport perspective

Statistical Methods

  1. "Bayesian Data Analysis" - Gelman et al. - Hierarchical modeling - Prior selection - Uncertainty quantification

Online Resources

Fantasy Football Projections

  1. Fantasy Points Data - Historical projections archive - Accuracy tracking - https://www.fantasypointsdata.com/

  2. The Podfather (Fantasy Football Analytics) - Projection methodology articles - R code examples - https://fantasyfootballanalytics.net/

  3. ESPN Fantasy - Projection explanations - Historical accuracy - https://www.espn.com/fantasy/football/

Analytics Sites

  1. Football Outsiders - Player metrics explanations - Historical data - https://www.footballoutsiders.com/

  2. PlayerProfiler - Advanced metrics - Prospect evaluation - https://www.playerprofiler.com/

Tools and Libraries

Python

  1. scikit-learn - https://scikit-learn.org - Regression models - Cross-validation - Feature selection

  2. statsmodels - https://www.statsmodels.org - Statistical models - Time series - Confidence intervals

  3. scipy - https://scipy.org - Statistical functions - Optimization - Distribution fitting

Data Sources

  1. nfl_data_py - NFL data access
  2. cfbd - College football data API
  3. Pro Football Reference - Historical stats

Fantasy Industry Resources

Projection Providers

  1. FantasyPros - Consensus projections
  2. ESPN - ESPN projections
  3. Yahoo - Yahoo projections
  4. Rotowire - Expert projections

Accuracy Tracking

  1. Fantasy Pros Accuracy - https://www.fantasypros.com/nfl/accuracy/ - Historical accuracy rankings - Expert comparison - Methodology insights

Research Groups

Academic

  1. CMU Statistics in Sports - Player modeling research - Aging curves - Uncertainty quantification

  2. MIT Sloan Sports Analytics - Annual conference papers - Player evaluation research

Industry

  1. PFF Research - Grading methodology - Projection systems - Data-driven analysis

Key Methodologies to Study

Marcel System

  • Origin in baseball (Tom Tango)
  • 5/4/3 weighting
  • Regression to mean
  • Applicable to football with modifications

PECOTA

  • Comparable players methodology
  • Percentile projections
  • Career trajectory modeling

Steamer/ZiPS

  • Regression techniques
  • Aging adjustments
  • Component projections

Suggested Learning Path

Week 1-2: Regression Foundations

  • Study Marcel methodology
  • Implement basic regression projector
  • Test on historical data

Week 3-4: Position-Specific Models

  • Build QB projector
  • Build RB projector
  • Handle multi-position players

Week 5-6: Aging and Development

  • Research aging curves by position
  • Implement aging adjustments
  • Validate against historical data

Week 7-8: Comparable Players

  • Build similarity finder
  • Integrate with regression
  • Evaluate improvement

Week 9+: Complete System

  • Combine all components
  • Add confidence intervals
  • Deploy and evaluate