AS '26

Agentic Sport
Analytics

The Evidence-Contract Field Guide: automated sport analytics, measured on pickleball and rugby league

Read the Book

By Mehran Mozaffari · August 2026

An Inflection Point

In 2025, sport analysts were experimenting with AI. In 2026, they’re rebuilding their entire analysis workflow around AI agents.

The Five Verbs of Sport Analytics

34

Chapters covering the full agentic sport analytics pipeline

600+

Figures, screenshots, and comparison diagrams

About the Book

Agentic Sport Analytics covers the full pipeline from raw footage to a defended coaching decision: the evidence contract, capture and calibration, detection and tracking, trajectory recovery, event data, ratings and expected value, vision-language models, and the agent harness that binds them.

Written with AI assistance using an agentic publishing pipeline. The author directed, reviewed, and approved all content. Technical review by Claude Opus 4.6 and Gemini 3.1 Pro. Free for personal use under a personal use license.

The book includes 34 chapters, 600+ figures, and a running laboratory of 20 experiments measured on two real sports — pickleball as the fixed-camera baseline, rugby league as the moving-camera stress test. Whether you’re a coach who needs defensible numbers, an analyst building the pipeline, or an engineer shipping sport AI, this book provides the evidence discipline and practical workflows you need.

Mehran Mozaffari

“No run, no dossier. Every number that reaches a coach must trace back to an observation someone can re-run.”

Mehran Mozaffari →

Author, Sydney, Australia · AI Operator & Applied AI Builder

What the book covers:

  • Agent platforms (Claude Code, OpenCode, Gemini CLI)
  • Design-for-agents harnesses & skill files
  • Design-as-code workflows & design systems
  • Motion & video design (Remotion, Hyperframes)
  • MCP server integrations (Figma, Miro)
  • Multi-agent team orchestration case studies
Browse all chapters →
AI coding agents are now operating design tools, generating visual artifacts, and closing the loop from design intent to production output.

Watching

Watching 01

The watching phase of the sport analytics pipeline.

Tagging

Tagging 02

The tagging phase of the sport analytics pipeline.

Modelling

Modelling 03

The modelling phase of the sport analytics pipeline.

Interpreting

Interpreting 04

The interpreting phase of the sport analytics pipeline.

Acting

Acting 05

The acting phase of the sport analytics pipeline.

Appendices

Reference materials

Reference

The Claims Register

Every measured claim, E-ID, and honest [verify] label — the book's evidence contract in one place

READ →

Reference

The Book as a System

Dual-sport navigation: the pickleball practitioner track and the rugby league track

READ →

Reference

The Frontier

What 2026-27 changes in sport foundation vision, edge hardware, and the agentic twin

READ →
ection>
AS '26

Agentic Sport Analytics

The Evidence-Contract Field Guide: automated sport analytics, measured on pickleball and rugby league. By Mehran Mozaffari. Research Edition, August 2026.

Chapter 1

Why This Book Exists: The Five Verbs of Sport Analytics

Chapter 2

The Evidence Contract & Data Provenance

Chapter 3

Sport Rules as Formal Systems

Chapter 4

Capture: Cameras, Lenses, Shutter, Placement

Chapter 5

Data Engineering for Sport Video

Chapter 6

Calibration I: Homography, Intrinsics, Distortion

Chapter 7

Calibration II: Broadcast Dynamics, GMC, and Per-Frame H_t

Chapter 8

Detection: YOLO, RF-DETR, and the AGPL Decision

Chapter 9

Tracking & Identity: Metrics, ReID, and Role Priors

Chapter 10

Segmentation & Foundation Models: SAM 2/3, DINOv3

Chapter 11

The Body in Motion: 2D Keypoints to 3D Biomechanics

Chapter 12

Smashing the Ball Wall: Spatio-Temporal Trajectory Recovery

Chapter 13

Identity: Who Is Who

Chapter 14

Multi-Camera Geometry, Line Calls, 3D Reconstruction

Chapter 15

Audio & Multimodal Cues: The Free Sensor

Chapter 16

Video Understanding: Action Recognition, Spatio-Temporal

Chapter 17

Event Data & the Common Representation

Chapter 18

Annotation: The Ground-Truth Workflow

Chapter 19

Automatic Eventing: State Machines, Confidence, Review Queues

Chapter 20

Statistics for Sport Practitioners

Chapter 21

Rating Systems: DUPR, ELO, Glicko, and Skill

Chapter 22

Expected Value: xG, VAEP, EPV, and Their Sport Transplants

Chapter 23

Tactical ML: Graphs, Equivariance, and Honest Forecasting

Chapter 24

Simulation & Counterfactuals: The Honest Rebuild

Chapter 25

Where Vision-Language Models Help, and Where They Lie

Chapter 26

From Numbers to Narrative: Reports, Scouting, Coach UX

Chapter 27

The Agent Harness for Sport Analytics

Chapter 28

The Live Coaching Cockpit: Real-Time Systems, Honestly Measured

Chapter 29

Practice Design & Interventions: The Acting Loop

Chapter 30

Sensors & Hardware: Wearables, Smart Courts, the Fusion Spine

Chapter 31

Deployment, Licensing, Rights & Ethics

Chapter 32

The Laboratory: Reproducing the Book's Claims

Chapter 33

The Frontier: What's Changing in 2025-2026

Chapter 34

The Book as a System: How to Use It (Human + Agent)

Get the book & stay updated

Free for personal use. Download the HTML version or read online. Get notified about new chapters, tools coverage, and edition updates.

CC BY-NC-ND 4.0 · Personal use free, commercial requires permission · Written with AI assistance