PinnedInData And BeyondbyBen Fairbairn·Apr 16, 2025Do not underestimate Linear RegressionA simple but powerful ideaA response icon10A response icon10
PinnedInScience SpectrumbyBen Fairbairn·Mar 11, 2025Democracy, Taskmaster and Simpson’s ParadoxWhat an often-overlooked statistical quirk has to do with the UK’s 1951 general election and the 2024 broadcast of a multi-award winning…A response icon4A response icon4
PinnedInData And BeyondbyBen Fairbairn·Aug 17, 2025Why your ML model may be underpredicting: A Smear Campaign Against Jensen’s InequalityA beautiful and powerful idea undermining regressionsA response icon3A response icon3
PinnedInData And BeyondbyBen Fairbairn·Jul 13, 2025Expect the Non-Expected (and no, I don’t mean the Unexpected)!Why should there be a mean?A response icon6A response icon6
PinnedBen Fairbairn·Jun 14, 2024Monty Hall Simplified & GeneralisedSome thoughts on a classic brain teaser
InScience SpectrumbyBen Fairbairn·3d agoA Strong Opinion on OpenAI’s Mathematical BreakthroughsSit down, calm down and stop catastrophising!A response icon16A response icon16
InThe Quantastic JournalbyBen Fairbairn·Jun 771, 701, 7001, 70001, and 700001 are all prime, but 7000001=197x35533. What’s going on here?Recall that a whole number greater than 1 is prime if the only whole numbers dividing it are 1 and itself. These have played a central role…A response icon4A response icon4
InScience SpectrumbyBen Fairbairn·May 5The 7-Colour Problem with Graphs on the PlaneAnd no, I did not mistype 4 as 7!A response icon2A response icon2
InScience SpectrumbyBen Fairbairn·Mar 14Is 𝜋 > 4?The subtlety of convergent seriesA response icon3A response icon3
