This is a fascinating and important set of data which shows us where things are going, using OpenAI as a canary in the coal mine. The chatbot era is over, and agentic systems are coming to tasks beyond engineering. And skills show promise as a way to standardize AI use in firms.
contributed by Andy on Jun 30, 2026 in curves, things that go up
Last night's big wins for Mamdani-backed candidates were driven by young college grads, often at odds with the party's traditional working-class & minority base. In NY-13, the less Hispanic and more college-educated a district, the more likely it was to vote for the challenger.
contributed by Andy on Jun 30, 2026 in decision surface, gradient
Presented at #ASCO26: Among patients with previously treated metastatic pancreatic ductal adenocarcinoma, the RAS(ON) inhibitor daraxonrasib led to significantly longer overall survival and progression-free survival than chemotherapy. Full phase 3 RASolute 302 trial results:
— @NEJM
contributed by Andy on Jun 9, 2026 in curves, things that go down, standing ovations
what a fantastic chart from @kevinmuir showing how unusual this earnings season has been
contributed by Andy on May 14, 2026 in spaghetti, things that go up, outlier
At the same time, most great coaches tend to outperform win probability model expectations So being simply average in spots with >75% win probability may point to something
contributed by Andy on Sep 8, 2025 in VORP, things that go up
contributed by Andy on Mar 16, 2025 in group chat, sad
videoI just pushed a new paper to arXiv. I realized that a lot of my previous work on robust losses and nerf-y things was dancing around something simpler: a slight tweak to the classic Box-Cox power transform that makes it much more useful and stable. It's this f(x, λ) here:
contributed by Andy on Feb 18, 2025 in curves, future fields medalists
Something happened in year 774 that was so powerful we can measure the radioactivity in old tree rings. Best guess is a solar flare, 10 times more powerful than the Carrington Event in 1859 that zapped telegraph wires. Worth considering the effect of the next one. Details:
contributed by Andy on Feb 14, 2025 in trees, eek
My textbook now includes an updated figure from Carter & McCullough, 2014. It's now generated from the raw data. It's a striking image. The uncorrected effect size is d = 0.62 based on 198 studies. Now, large replications later, we know the true effect size is 0. What a waste.
— @lakens
contributed by Andy on Jun 12, 2023 in hard to parse, acronyms, replication
Los Angeles compared to other cities
contributed by Andy on May 29, 2023 in small multiples, things that go up, bloomberg