I used Python, Plotly, and Figma to make the image. The data is from a publicly available dataset of ~60,000 homebrew recipes.

Analysis description and links to the dataset and Jupyter Notebook are here: https://www.memolli.com/blog/tracking-beer-types/

Posted by Alive-Song3042

23 comments

  1. I have nothing constructive to add, but I think this is neat, and thank you for sharing the Jupyter notebook

  3. Good chart but cask ales (quite literally real ale) has been left in the other category rather than ale itself lol

  5. Cool visualization. You have Munich Helles in “other” instead of “lager” though, should fix that.

    Also there appear to be dots at 1% for barleywine and 0% for imperial stout? Are n/a beers included?

  6. Where is “amber” ? It should be sweeter & lighter than stout and porter, maltier & richer than ale and more balanced than IPA & pale ale.

  8. Belgian tripels and strong ales cap out at around 15%, so why is their upper limit whisker being shown as much lower?

  9. What I took from this data is that “American IPA” is a meaningless label for determining bitterness.

  10. Relative bitterness (compared to sugar content) usually is way more useful. Typical pils tastes oft a lot more bitter than IPA, simply since IPA tends to have a lot of sugar compared to pils.

  11. Great chart, very interesting.

    But several styles in the “other” category belong in Lager or Ale (California Common, Helles, ESB, Mild, etc.) Heck, the Pilsner category could go into Lager, and IPAs could go into Ale.

    Checking out the source, doesn’t seem he has much background in beer, using a recipe site as his data.

  14. Great looking figure! Is there a reason you have the bitterness plots much wider than the alcohol % plots? Both measures are interesting to look at, and it’s harder to see differences in alcohol %. Since they use different units, the width doesn’t need to be proportional to the absolute range.

  15. Wouldn’t experimental beer be anything, the complete range?

  16. American IPA IQR is huge, then there’s also a huge number of outliers.

  17. This is really helpful!

    I always knew I wasn’t a “bitter beer” guy (I’m not really a beer guy in general) but now I can actually find the beer/ale that I do like and find other beer/ales in the same IBU class to try out and go from there.

  18. Nice. Could be done better separating ales from lagers. Itd be nicer if it clarify that ipa and stout are types of ales and lagers in another division. This is kind of mixed up in the chart.

  19. No one actually likes IPA. Anyone who says they do is only trying to fit in or to mimic someone else who they think is cool.

  20. Who would have thought a strong bitter is actually medium bitter.

  22. ||
    ||
    |I never need a drink menu. I got the thrills for the pils. ‘Cause I’m a pilsner man. – Teddy|

  23. ||
    ||
    |I never need a drink menu. I got the thrills for the pils. ‘Cause I’m a pilsner man. – Teddy|

Comments are closed.