Skip to content
JM.
All work

Data · Analytics

Draft

Baseball Analytics

Exploring player performance, ballpark factors, and baseball statistics through software and data analysis.

Contents

Baseball is a game of systems and data. This project applies the same engineering habits used on infrastructure — careful modelling, reproducible pipelines, honest measurement — to the statistics of the game.

Overview

An exploratory analytics project built around publicly available baseball data, looking at how player performance interacts with ballpark environments and how the game's statistics evolve over time.

Problem statement

Raw baseball statistics are abundant, but context is scarce. The same batting line means something different in a hitter-friendly park than in a pitcher's park, and comparing players across eras requires careful normalization.

Constraints

  • Use only publicly available, appropriately licensed data sources.
  • Pipelines must be reproducible end-to-end from raw data.
  • Visualizations should be readable by fans, not just analysts.

Architecture and approach

Pipeline (planned)
raw public data ──▶ ingest & clean ──▶ model (park factors, rates)
                                             │
                                             ▼
                                   explore & visualize

Technical decisions and tradeoffs

Placeholder — to be written

Modelling choices (e.g. how park factors are estimated and regressed) will be documented as they are made.

Implementation details

Placeholder — to be written

Results and impact

Placeholder — to be written

Findings will be written up here once analyses are complete.

Lessons learned

Placeholder — to be written

Python · SQL · Notebook-based exploration · Data visualization