This project connects WPL scorecards with squad and auction data. The aim is to make questions about spending and performance possible using tables that join consistently. The dataset and the notebook are both public.
My contribution
I defined the analysis questions and reviewed the dataset structure and the source matching. AI tools helped with processing and with building the notebook.
Connect the records
The dataset has nine CSV files, published on Kaggle under the ODC-By licence. Match IDs connect deliveries and scorecards. Season, team and player identifiers connect those records to squads.
The documented scope covers the 2023–2026 seasons. Anyone can download the files and rebuild the joins.
Match names across sources
A player can appear under an abbreviated scorecard name and a full auction name. The dataset uses stable IDs where available and reviewed aliases for the other cases.
The Bangalore-to-Bengaluru team rename is normalised so the franchise stays together in comparisons.
Explain the blanks
An unsold player has no sale price. An unpublished fee is also blank, for a different reason. Those cases are documented instead of being filled with an invented average.
The documentation also explains squad members who did not play and describes checks reconciling runs and wickets with the match data.
The notebook and its charts
The public notebook, "WPL 2023–2026 records and who was worth the money", has 11 charts.
One question is how squad spend compares with results. The answer still has to allow for the small number of teams and seasons.