Pokémon: The Artometrics of the World's Biggest Franchise

A data analysis of 1,028 Pokémon entries across nine generations, drawn from the TidyTuesday 2025-04-01 dataset (sourced from PokéAPI). Built as part of the Artometrics content series — data-forward analysis of creative industries.

What's in this repo

• pokemon.qmd — Quarto markdown file containing all R code, analysis, and write-up
• artometrics.css — custom HTML theme (same across all Artometrics reports)
• art-head.html — header injection for fonts and nav
• art-body.html — body injection for site chrome
• charts/chart1_stat_heatmap.png — median base stat by primary type (heatmap)
• charts/chart2_generation_ridgeline.png — base experience distributions by generation (ridgeline)
• charts/chart3_size_scatter.png — height vs. weight across all Pokémon, colored by primary type (scatter)

What the analysis covers

Three charts examining the design infrastructure underneath the world's highest-grossing media franchise:

1. Stat Identity by Type — Does each type actually play differently? Median base stats across all 18 primary types reveal intentional design specialization, from Dragon's generalist elite profile to Bug's across-the-board undertuning.

2. Generation Creep — Has Game Freak been inflating Pokémon power over time? Ridgeline distributions by generation show a stable median but a rising ceiling — each generation adds extreme outliers at the top end without dramatically shifting the average.

3. Built Different — What does physical size reveal about design intent? A height vs. weight scatter on log scale exposes the implicit size grammar of the franchise, and the outliers (Cosmoem at 999.9 kg, Wailord at 14.5 m) that deliberately break it.

Data source

Data Science Learning Community. (2025). TidyTuesday: A weekly social data project. Week 13, 2025 — Pokémon. https://tidytues.day

Original data sourced from PokéAPI: https://pokeapi.co

Tools

• R / Quarto
• tidyverse · ggplot2 · ggtext · ggridges · ggrepel · scales

Disclosure

This analysis was researched, written, and produced in active collaboration with Claude AI (Anthropic). The research questions, editorial framing, and brand vision are the author's own. Code, analysis, and prose were developed through a directed human–AI partnership. All data is publicly available via TidyTuesday and PokéAPI.