This project delivers a fully interactive MLB performance dashboard built using Plotly Dash. It enables users to explore team and player statistics from the Retrosheet dataset (2000–2019) with custom filters, visualisations, and comparisons.
The dashboard empowers users — especially team managers, analysts, and fans — to:
- Track player performance across seasons
- Compare batting vs pitching metrics
- View positional breakdowns and time-series stats
- Make informed decisions on roster and strategy
Alex Conroy
Individual project submitted as part of IFN703 — Interactive Visualisation & Analytics
- Python
- Dash / Plotly for UI and charts
- Pandas / NumPy for data manipulation
- Retrosheet MLB CSVs as data source
✅ Filter by player, team, or position
✅ Dynamic radar plots to compare player skill areas
✅ Bar charts for team-level performance
✅ Time series visualisations for performance trends
✅ Built with modular components for clean code
mlb-dashboard/ ├── data/ # Contains MLB stats (see below) │ ├── Battings.csv │ ├── Pitchings.csv │ ├── 2019combinedroster.txt │ └── TEAM2019.txt ├── notebooks/ │ └── Dashboard.ipynb # Original notebook prototype ├── src/ │ ├── app.py # Dash layout & callbacks │ └── visualisations.py # Reusable Plotly chart functions ├── utils/ │ └── data_processing.py # Merging and cleanup utilities ├── report/ │ └── IFN703_Assignment3_Report.pdf ├── requirements.txt └── README.md
- Source: Retrosheet.org (historical MLB data)
- Manual preprocessing done to extract:
Battings.csv,Pitchings.csv- Team/player mappings (
2019combinedroster.txt,TEAM2019.txt)
These files must be manually placed in the
data/folder.
They are not included in the repo due to size.
- Integrate live data feeds via API
- Add player prediction metrics (e.g. WAR forecast)
- Support mobile layout and responsive resizing
- Add download/export options for filtered stats