Chilling Effects
Read the original on Astral Codex Ten →
Summary
Investigative essay (flagged 'extremely confused') on the claim that ~8.5% of deaths are cold-related, prompted by the global-warming-saves-cold-deaths argument. Surfaces real puzzles: the minimum-mortality-temperature sits ~80th percentile of each region's own temperature distribution (so Kuwaitis 'die of cold' below 35C), implying a winter/flu component distinct from a cold/cardiovascular one; the Africa (flu+HIV) and Greenland (hunting-season + returning-visitor disease) anomalies. Concludes the naive 'warming will net-save lives' view is probably wrong, but without a firm resolution.
Why this score
Quality 65 · Strong. Strong (lower): a good, honest investigation that generates real insight (the 80th-percentile MMT, the winter-vs-cold split) but explicitly ends inconclusive; a notch below Diseasonality, which lands a cleaner model.
Claude’s paradigm shift 45 · Moderate. Moderate-Notable: surfaces a genuinely under-appreciated framing (relative-temperature MMT) but doesn't resolve it into a durable model.
Real-world impact 1 · Negligible. Within-blog reasoning on a policy-relevant question; negligible material effect. =1.