📊 #OTD 🎂🎂🎂 Happy Minard Day! Charles Joseph Minard's, b. Mar 27, 1781, his big 245th. #OTD #dataviz One way to celebrate is to visit @infowetrust.com wonderful Visual Catalog of the Work of Charles Joseph Minard, bit.ly/3lOXsbR Another is my celebration of Minard Day, 2021 ed, bit.ly/49f7k1t

A dozen year ago, I was alone at home and decided to treat myself to a beer and a detailed exploration of the famous Minard map. I wrote this blog post, which is still one of the most popular on our website all these years later. 📊 chezvoila.com/blog/minard-...

@newyorker.com Thanks for dark mode. Could you fix basic text rendering next?

How far can you get in 60 minutes? European cities offer vastly larger areas reachable by public transit than US cities of comparable population size. Source: lconwell.github.io/lucasconwell...

See also: Blank Space by David W. Marx, which argues that aesthetic fragmentation and other forces have resulted in 21st century cultural stagnation.

Why do people mention which slop bot made something for them? Brand loyalty? Why not say nothing and let people think you used your own knowledge and abilities?

Posit is so embarrassing.

What is with all the "📊 for kids" books lately? Is it just slop bots driving the cost of illustration to zero? Do parents really buy these to read to their kids? Is it a résumé padding exercise for the authors?

"Nothing is lost by adding the second graph." Sure, except for a consistent scale. 📊

The StatsCan website is baffling. Every dataviz product is meticulously documented with authorship, update cadence, underlying datasets, keywords, related outputs... everything but a link to the actual viz! 📊

I wonder whether the categorization of farmers into "large" and "small" in the data used for this visualization 📊 of income inequality in 19th c. France amounts to conditioning on the outcome. Or maybe it's by acreage? freerangestats.info/blog/2026/02...

From the first and second editions of @rmcelreath.bsky.social's Statistical Rethinking. 📊

Am I wrong to think that overlapping bars is just straightforwardly an error? Now area (and thus visual impact) is no longer proportional to the data. Maybe I'm too VDQI-pilled? #dataviz 📊 (Chart from SWD: Before & After)

Well, @newyorker.com... is this "re-signed" or "resigned"? Why leave this at a line break? You'll use the dieresis but not the double hyphen?

Chapter 6 of Casella & Berger is a real page-turner! It's a bit funny to read them introduce the likelihood principle and credible intervals while carefully avoiding the term "Bayesian".

"There is a strong relationship, but it isn't statistically significant." This frequentism stuff has absolutely cooked people's brains.

At first I thought I'd like this better with filled, triangular arrowheads, but upon further reflection I think this style makes it more obvious where the data points are. Is there an unconscious tendency to imagine the data point in the centre of a filled arrowhead instead of at the tip?

@allendowney.bsky.social on Hempel's raven paradox. I like the heatmaps visualizing the joint posterior distributions in each scenario. www.allendowney.com/blog/2025/12...

The archives of the journal Visible Language, which has published interesting articles on #dataviz 📊 in addition to typography, calligraphy, spelling reform, and many other interesting things, are available for free online: journals.uc.edu/index.php/vl...

Can you guess which basic #dataviz errors this graphic features? Why in the world wouldn't the colour mapping be the same across the four donuts? And why not a 100% stacked bar chart instead of donuts? Just like movies about making movies, data graphics about making data graphics are never good.

I'm not proud to tell you that I clicked this Wirecutter article fully believing it was about the 📊 Best Bar Charts.

New paper for anyone working with data: Better data viz - for free, in few clicks. Below, all N = 111, M = 0.04, SD = 0.27. One-sided t-tests vs. 0 yield: t(110) = 1.67, p = .049. Use raincloud plots. Or risk wrong conclusions! Plot w/ @jaspstats.bsky.social today! 🧪 📊 #PsychSciSky #StatsSky 🧵👇

Will Posit's recent transition to hyping slop generation endanger its B Corporation status? Surely you can't be serious about climate targets while shilling this stuff, right?

Find the data 🧐🤓 #dataviz #datatoinkratio

Embarrassing. Why not just train a slop generator on all the conference submissions and then skip actually running it?

My struggles with Tableau 📊 seem to boil down to two issues: 1. Too little control over the aesthetic mapping. (E.g. cannot map to position.) 2. Dimensions and measures are treated radically differently (e.g by filtering) in ways that are unintuitive and/or just never what one would want.

Watch the seismic waves from today's magnitude 7.3 Alaska earthquake ripple across seismic stations in North America. More ➡️ loom.ly/kaJp5I0

I'm sorry, but if the image slop generator you used for your slides can't produce a world map with South America in it, I'm not buying your data engineering pipeline slop generator. Pretend tools for pretend work.

I've written an explainer blog post for my "data strips" #dataviz workbench web app, trying alternative 1-D distribution summary plots. rawdatastudies.com/2025/07/05/d...

I wouldn't want to be associated with that creep either, but I think if you open-source your #dataviz 📊 theme, you have to accept the possibility of creeps using it. In any case, I'd never assume that two chart authors were associated just because they use the same style, and you shouldn't either.

Vertical axis trickery in Bigalke et al. (2021), via dynomight.net/blue-light/. 📊

A #dataviz adjacent post from Andrew Gelman about which Venn diagrams can be drawn with rotational symmetry. The same Venn after whom the diagrams are named originated the frequency interpretation of probabilities, incidentally.

Some hard-hitting #dataviz coming out of New York.

Does Posit develop open source data science tools, or just LLM plug-ins? Based on their social media presence, you'd assume it were the latter.

6 old and 7 new 1D #dataviz. Trying Shortest Half with a twist: one break is allowed. Also half-sample mode & count-adaptive outlier thresholds. Here's 1000 random normal points plus two outliers. The green ones are new. Try it at xangregg.github.io/data-strips/

A little trick to set unwanted #ggplot2 theme elements to `element_blank()` en masse. 📊 (Copyable code in alt text.)

Is there nothing new in 📊 dataviz to talk about? Why is the same information endlessly rehashed? Career reasons? (I.e., to generate leads, pad résumés, boost personal brands?) Something else? Are we actually reaching new audiences? (Not a criticism of this post/person in particular.)

Dataviz 📊 freelancers/consultants: What proportion of your time is dedicated to churning out buzzword-laden LinkedIn posts and newsletter emails? 90%? 95%?

Seems like a case for syncing scales across facets. 📊

Calling all the data analysis and visualization done prior to 2022 "limited insights and manual analysis" because it doesn't harness the power of the slop generators is obscene. This stuff can't go the way of "blockchain" soon enough.

Just please don't mess with my scrollbar!

Top 15 ggplot2 extensions, by downloads during the last month. Some surprises here, at least for me. #rstats Code: gist.github.com/carl...

Lollipop charts are a very elegant alternative to bar charts. 📊

Andrew Gelman uses this example in a post about counteracting perceptual distortion in dataviz 📊. It compromises on ease of comparisons to maintain a visual metaphor. I want to see how a 100% stacked bar graph would look, but sadly getting the chart data from the code isn't straightforward.

The story you likely *didn’t* know about John Snow’s cholera map. www.maps.com/snow-mistake... “Maps inform, inspire, and ignite interest in a way that is unlike any other media.” By @jscarto.bsky.social 📊🗺️

"Stratification in Late Traditional Chinese Society" by G. William Skinner and Paul D. Barclay 📊

@brrosenau.bsky.social's alternative to the stacked bar chart (example courtesy of @yan-holtz.bsky.social) makes all bar heights easy to compare, but creates a much worse problem: the height and area scales are now out of sync, fundamentally violating the visual logic of a bar chart. 📊

with every careless AI-booster, we stray further from the light