Beer is a beverage with a rich history and an incredible diversity of styles. Recently, I analyzed a dataset of over 60,000 beer recipes to visualize how different styles compare in terms of alcohol percent and bitterness (International Bitterness Units or IBU).
The data is from thousands of homebrewers, so it is a bit noisy. I took the steps below to clean up the dataset. See the bottom of this post for the full Jupyter notebook and data source.
- I kept only the top 80 or so recipes by usage.
- I capped the IBU at 150. From what I’ve read, people can’t really taste bitterness above ~120 IBU.
- I removed recipes with outlier ABV or IBU values.
The resulting image shows a wide range of beer styles. For example, Imperial IPAs stand out for their high ABV and IBU, while others like Berliner Weisse or American Light Lager are much lighter and less bitter. The visual nicely conveys the landscape of beer styles and their characteristics.
Key takeaways:
- High ABV & IBU: Styles like Double IPA and Russian Imperial Stout are both strong and bitter.
- Low ABV & IBU: Wheat beers and light lagers are milder and more approachable.
- Diversity: The spread of points highlights the creativity and variety in brewing.
Sources
The figure was made using Python’s Plotly library and Figma.
Data Source:
Jupyter Notebook: