Articles and links I've found interesting, curated from my RSS reader. Subscribe
In data visualization, we often focus on which chart is best for a given data set configuration, as if one chart is always better than another for that specific data set. For example, it’s not...
It seems execution beats foresight: Retail traders correctly forecast asset price direction yet lose money. Using 222 million prediction market trades with observable terminal payoffs, we decompose...
Mediation analysis goes beyond asking “does the treatment work?” to ask “how does the treatment work?” Understanding the mechanisms by which an intervention achieves its effect can have important...
Warm temps and patio happy hours are on the horizon, so you know what that means: It’s time to play the IRS’s favorite twisted (and forced) guessing game — we’re going to make you figure out how much...
This is a possible section from the open-source living textbook Better Code, Better Science, which is being released in sections on Substack. The entire book can be accessed here and the Github...
Next week, I’m opening registration for my new course Everyday Energy. In the last essay, I discussed why we’re living through a crisis of human energy. Today, I’d like to share some thoughts on how...
I largely agree with this piece by Maarten Boudry called “The Fallacy Fallacy.” The central insight is quite correct: the kind of fallacy hunting that’s become popular in the era of online forums and...
Tom Hartmann had not planned to become a Software Mechanic. But then, nobody who was a Software Mechanic had planned to become one, because the job hadn’t existed seven years ago, and the people...
Academics are worried about AI replacing them and their work. I think we spend insufficient energy appreciating how much we can expand the frontier of research thanks to this new tool. To make this...
This post comes from many fruitful discussions with my friend Kyle Jensen. It also benefited from helpful comments from Kyle as well. Researchers increasingly use large language models to summarize,...
Before the 1950s, we shipped goods in boxes, crates, and sacks. Dock workers would load and unload each item individually.Then came the shipping container.Containerization came with huge upfront...
Agentic Engineering Patterns > LLMs are restricted by their context limit - how many tokens they can fit in their working memory at any given time. These values have not increased much over the...
[This article was first published on Seascapemodels, and kindly contributed to R-bloggers]. (You can report issue about the content on this page here) Want to share your content on R-bloggers? click...
All correspondence concerning Patchs, Certificates, Photos, Profiles, Etc., for the "GRID" and "All" the associated Lists and Web Pages is to be Addressed to "Ed Hawkins" E-Mail Address:...
T is an experimental, reproducibility- and pipeline-first programming language for declarative data manipulation and statistical analysis. Built on the foundation of Nix, T is designed from the...
Coding agents for data analysis Here's the handout I prepared for my NICAR 2026 workshop "Coding agents for data analysis" - a three hour session aimed at data journalists demonstrating ways that...
Raise a glass to William Sealy Gosset The post How Beer Brewing Revolutionized Modern Statistics appeared first on Nautilus.
Each year at BigML we use machine learning models trained on historical awards data to predict the winners of the Academy Awards. In our previous post, we presented our machine learning predictions...