Chapter 9: Further Reading

Overview

This curated collection of resources extends your understanding of Expected Threat (xT), ball progression metrics, and possession value frameworks. Resources span the original research, implementation guides, and advanced extensions.

Foundational Papers

xT Development

  1. "Introducing Expected Threat (xT)" - Karun Singh (2019) - Original blog post introducing the xT concept - Clear explanation with visualizations - Implementation details - Link: karun.in/blog/expected-threat.html

  2. "Expected Threat" - StatsBomb Research - Commercial application of xT concepts - Integration with other metrics - Professional implementation considerations - Link: statsbomb.com/articles

VAEP Framework

  1. "Actions Speak Louder than Goals: Valuing Player Actions in Soccer" - Decroos et al. (2019) - Machine learning approach to action valuation - Comprehensive framework for all actions - Includes defensive value - Published in KDD '19 proceedings

  2. "Valuing On-the-Ball Actions in Soccer" - Decroos et al. (2019) - VAEP technical implementation details - Feature engineering for action prediction - Comparison with other approaches - KU Leuven DTAI publications

EPV Framework

  1. "Decomposing the Immeasurable Sport: A Deep Learning Expected Possession Value Framework for Soccer" - Fernández & Bornn (2018) - Deep learning possession valuation - Continuous surface approach - State-of-the-art accuracy - MIT Sloan Sports Analytics Conference

  2. "A Framework for the Fine-Grained Evaluation of the Instantaneous Expected Value of Soccer Possessions" - Fernández et al. (2019) - EPV technical details - Off-ball value incorporation - Real-time valuation capabilities - Academic paper with methodology

Progressive Metrics Research

Passing Analysis

  1. "Not All Passes Are Created Equal: Objectively Measuring the Risk and Reward of Passes in Soccer" - Power et al. (2017) - Pass risk-reward framework - Foundation for progressive pass concepts - MIT Sloan Sports Analytics Conference

  2. "A Data-Driven Method for Evaluating Creative and Destructive Plays in Soccer" - Routley & Schulte (2015) - Play value assessment - Chain analysis approach - Conference paper on action sequences

Ball Progression

  1. "Measuring Ball Progression in Football" - Various analysts - Multiple blog posts on progression metrics - Different definitions and implementations - FBref, StatsBomb, and Wyscout methodologies

  2. "Progressive Passes: What They Mean and How to Use Them" - American Soccer Analysis

    • MLS-focused progressive analysis
    • Practical interpretation guide
    • Position-specific benchmarks

Books and Textbooks

Soccer Analytics

  1. "Soccermatics" - David Sumpter (2016)

    • Mathematical modeling in soccer
    • Pitch control and passing models
    • Accessible academic approach

  2. "The Numbers Game" - Anderson & Sally (2013)

    • Statistical thinking in soccer
    • Foundation for advanced concepts
    • Historical context

Machine Learning for Sports

  1. "Hands-On Machine Learning with Scikit-Learn and TensorFlow" - Aurélien Géron

    • ML implementation techniques
    • Applicable to VAEP-style models
    • Practical Python examples

  2. "Pattern Recognition and Machine Learning" - Christopher Bishop

    • Theoretical ML foundations
    • Value iteration and dynamic programming
    • Essential for advanced implementations

Online Resources

Tutorials and Guides

  1. Friends of Tracking YouTube Channel

    • Video tutorials on xT implementation
    • Code walkthroughs
    • Academic presentations
    • Link: youtube.com/friendsoftracking

  2. McKay Johns Analytics Tutorials

    • Practical implementation guides
    • Progressive pass visualization
    • StatsBomb data tutorials
    • Link: youtube.com/mckayjohns

  3. FC Python Tutorials

    • Python-based soccer analytics
    • xT calculation examples
    • Beginner-friendly approach
    • Link: fcpython.com

Data Visualization

  1. mplsoccer Documentation

    • Python visualization library
    • Pitch plots and heatmaps
    • xT grid visualization
    • Link: mplsoccer.readthedocs.io

  2. Plotly Soccer Visualizations

    • Interactive xT surfaces
    • Dashboard creation
    • Web-based display

Industry Blogs and Analysis

Analytics Providers

  1. StatsBomb Articles

    • Professional xT applications
    • Player and team analysis
    • Methodology insights
    • Link: statsbomb.com/articles

  2. FBref Progressive Stats Guide

    • Detailed metric definitions
    • League-wide data
    • Position benchmarks
    • Link: fbref.com

  3. Wyscout Academy

    • Professional scouting integration
    • Video and data combination
    • Industry applications
    • Link: wyscout.com

Independent Analysts

  1. Karun Singh's Blog

    • Original xT creator
    • Technical implementations
    • Model refinements
    • Link: karun.in/blog

  2. American Soccer Analysis

    • MLS analytics application
    • xT and progression metrics
    • Accessible explanations
    • Link: americansocceranalysis.com

  3. The Athletic - Soccer Analytics

    • Premium analysis content
    • xT applications in journalism
    • Player evaluation features
    • Link: theathletic.com

  4. Statsbomb Conference Presentations

    • Annual research presentations
    • Industry practitioner talks
    • Video archives available

Code Libraries and Tools

Python Packages

  1. socceraction

    • VAEP and SPADL implementation
    • Official KU Leuven library
    • Research-quality code
    • Link: github.com/ML-KULeuven/socceraction

  2. mplsoccer

    • Pitch visualization
    • Heatmaps and pass maps
    • xT grid plotting
    • Link: mplsoccer.readthedocs.io

  3. kloppy

    • Multi-provider data loading
    • Event and tracking data
    • Standardized formats
    • Link: github.com/PySport/kloppy

  4. statsbombpy

    • StatsBomb data access
    • Event data parsing
    • Free data availability
    • Link: github.com/statsbomb/statsbombpy

Computational Tools

  1. NumPy

    • Matrix operations for transition matrices
    • Value iteration implementation
    • Efficient numerical computing

  2. SciPy

    • Sparse matrix handling
    • Optimization algorithms
    • Statistical functions

Conference Proceedings

Academic Conferences

  1. MIT Sloan Sports Analytics Conference

    • Annual research presentations
    • Soccer analytics papers
    • xT and EPV original work
    • Link: sloansportsconference.com

  2. KDD Sports Analytics Workshop

    • Machine learning applications
    • VAEP and action valuation papers
    • Data mining approaches

  3. ECML PKDD Sports Analytics

    • European ML conference
    • Soccer-specific workshops
    • Technical implementations

Industry Conferences

  1. StatsBomb Conference

    • Annual industry event
    • Professional applications
    • Video presentations available
    • Link: statsbomb.com/conference

  2. OptaPro Forum

    • Industry research presentations
    • Applied analytics
    • Historical archives

Advanced Topics

Tracking Data Integration

  1. "Wide Open Spaces" - Fernández & Bornn (2018)

    • Space creation quantification
    • Off-ball contribution measurement
    • Tracking data requirements

  2. "Physics-Based Modeling of Pass Probabilities" - Spearman et al. (2017)

    • Pass completion modeling
    • Trajectory and interception
    • Foundation for advanced xT

Dynamic Programming

  1. "Reinforcement Learning: An Introduction" - Sutton & Barto

    • Value iteration theory
    • Markov Decision Processes
    • Foundation for xT mathematics

  2. Dynamic Programming and Optimal Control - Bertsekas

    • Advanced DP theory
    • Computational approaches
    • Optimal policy derivation

Network Integration

  1. "Passing Networks in Soccer" - Peña & Touchette

    • Network analysis of passing
    • Centrality and flow measures
    • Integration with xT

  2. "A Network Science Approach to Football" - Various authors

    • Graph-based analysis
    • Pass xT weighting
    • Structural insights

Position-Specific Applications

Defenders

  1. "The Ball-Playing Center-Back" - Various analytics blogs
    • xT for defender evaluation
    • Progressive defender metrics
    • Scouting applications

Midfielders

  1. "Valuing Midfield Contributions with xT"
    • Role-specific analysis
    • Build-up contribution measurement
    • Comparison frameworks

Full-Backs

  1. "The Modern Full-Back: Measuring Attacking Contribution"
    • Inverted full-back analysis
    • Crossing vs. progression
    • Tactical role evaluation

Data Sources

Free Data

  1. StatsBomb Open Data

    • Event data with coordinates
    • Multiple competitions
    • Sufficient for xT construction
    • Link: github.com/statsbomb/open-data

  2. Wyscout Open Data

    • Event data samples
    • European leagues
    • Research purposes
    • Link: figshare.com (search Wyscout)

Commercial Data

  1. StatsBomb 360

    • Freeze frame data
    • Player positions at events
    • Enhanced xT capabilities

  2. Second Spectrum / SkillCorner

    • Tracking data
    • Required for EPV
    • Premium pricing

Beginner (Weeks 1-4)

  1. Read Karun Singh's original xT blog post
  2. Understand basic grid concept
  3. Implement simple xT calculation
  4. Calculate player xT totals

Intermediate (Weeks 5-12)

  1. Build full xT model with value iteration
  2. Study progressive pass definitions
  3. Implement player progression profiles
  4. Create visualizations

Advanced (Weeks 13+)

  1. Read VAEP paper and implement
  2. Explore EPV concepts
  3. Integrate with tracking data
  4. Build custom valuation frameworks
  5. Apply to real scouting/analysis

Citation Formats

Academic Paper (VAEP):

Decroos, T., Bransen, L., Van Haaren, J., & Davis, J. (2019).
Actions speak louder than goals: Valuing player actions in soccer.
Proceedings of the 25th ACM SIGKDD International Conference on
Knowledge Discovery & Data Mining.

Blog Post (xT):

Singh, K. (2019). Introducing Expected Threat (xT).
Retrieved from https://karun.in/blog/expected-threat.html

Data Source:

StatsBomb. (2023). StatsBomb Open Data. Retrieved from
https://github.com/statsbomb/open-data

Keeping Current

Following Updates

  • Twitter/X: @kaaborern (Karun Singh), @StatsBomb, @deaborae
  • GitHub: Watch socceraction repository
  • Conferences: MIT Sloan, StatsBomb annual events

Research Alerts

  • Google Scholar alerts for "expected threat soccer"
  • arXiv sports analytics section
  • SSRN sports economics papers

Last updated: January 2025. Links and availability may change.