Free Course · 5 lessons · ~3 hours
AI for Sport Analysts
A hands-on course on using AI in sport analytics — performance, tactical, scouting, and broadcast. For practitioners who already know their sport and want to know what AI changes about their work.
Who this is for
You should take this course if you work in sport analytics — as a performance analyst at a club, a scout in any sport, a data scientist in a federation, an analytics lead at a broadcaster, or a sport-tech operator. You already know your sport. You have probably written more "weekend report" CSVs than you can remember. The question this course answers is: what does AI actually change about that work?
You should also take it if you work in AI and want to understand the sport-analytics application area properly. Sport is one of the most data-rich applied domains in the world; the patterns that work here transfer to many other places.
You do not need to be a data scientist. The exercises use AI tools through a browser. There is no code, no installs, no model fine-tuning. We assume you understand basic statistics (mean, ratio, sample size); you do not need anything more advanced than that.
What you will learn
The shape of sport analytics in 2026
The four domains — performance, tactical, scouting, fan and integrity. Where the field has been, where it is now, what AI changes and what it does not.
Building your sport analytics assistant
The system prompt design for sport. Setting it up to handle sport-specific terminology, athlete privacy, and the difference between description and prediction.
Working with real sport data
Event data, tracking data, biometric data. Feeding it to the bot, asking good questions, avoiding the small-sample traps that sport analytics is famous for.
From data to narrative
The translation problem. Same analysis, different audience — coach, scout, broadcaster, board. The bot as a writing partner, not a decision-maker.
Where this leads, what it doesn't replace
Athlete data ethics in practice. The role of sport-specific judgement. The frontier: video, computer vision, biometrics, multi-modal. A use policy you write for your work.
What you will need
- About three hours of total time. Each lesson is 25 to 45 minutes.
- A browser, and an account with either Claude.ai or ChatGPT. The free tier is enough.
- For Lesson 3, a CSV file you can paste in. We will point at free, public sport datasets (StatsBomb open data, FBRef, World Athletics tables) so you have something concrete to try. You can also use your own data if you have it.
- Knowledge of your own sport. The course is sport-agnostic; you bring the sport.
How this fits with the full programme
This is a taster of Path E — AI for Sports, the Tier 3 specialisation of the Integrated AI Program. Path E is 60 ECTS across ten courses, covering performance modelling and injury prediction, computer vision for sport, tactical AI, scouting and player valuation, fan and broadcast AI, officiating and integrity, esports analytics, ethics and governance, and an applied capstone. Path E launches with the January 2027 cohort; applications open from November 2026.