← All work

Earlier data work

Premier League performance analysis

Match data 2023/24 – 2025/26 · Built 2025–2026

I picked football because every result is already a structured experiment in tactics, luck and sample size. Three seasons of Premier League fixtures (from football-data.co.uk) gave me enough volume to build a credible league table from scratch.

The pipeline

  • Load and clean multi-season CSVs, then merge seasons consistently.
  • EDA: home vs away splits, goal trends, and the correlation between shots on target and goals.
  • Team-level leaderboards and a chart library, including the biggest margin wins and clean sheets.
Horizontal bar chart of the ten biggest margin wins over three seasons
Top 10 biggest margin wins.
Horizontal bar chart of the ten teams with the most clean sheets over three seasons
Clean-sheet leaderboard: defensive signal beyond raw points.

Let’s talk

Hiring for a backend role?

Graduate & junior roles from June 2027, part-time roles now.

Get in touch