The Evidence That A Million Americans Died Of COVID

A follow-up defending the 1.2 million US COVID-death figure against commenters' 'died with vs of COVID,' faked-statistics, and treatment-caused-deaths skepticism. The core rebuttal is all-cause excess mortality: ~500-700K excess deaths in each of 2020 and 2021 rules out the idea that these were just normal deaths relabeled; excess deaths tracked reported COVID deaths within ~10%; a back-of-envelope shows that even maximally dishonest incidental-COVID labeling (every senior who died while incidentally positive) could produce under 5% of the observed excess. It can't be a single treatment (ventilators, remdesivir, vaccines) because each was used only in subsets of the pandemic while excess mortality tracked cases throughout, and stitching them all together requires each to be unprecedentedly and equally dangerous yet invisible to studies. The 'I personally don't know anyone who died' objection is dissolved with base rates: 1.2M/340M is ~1/300, the same prevalence as multiple sclerosis (you need ~208 acquaintances for even odds of knowing one), and the ACX survey found 6.5% had a family member die of COVID.

Quality 68 · Strong. Strong band. A clean, correct application of basic epidemiological reasoning (excess mortality, correlation, base rates) to debunk a confused dispute, with good intuition pumps (the 1/300 = MS comparison, the back-of-envelope). Held mid-Strong because it is short, responsive to a comment thread, topical, and the central move (look at excess mortality) is standard rather than novel. A=68.

Claude’s paradigm shift 32 · Slight. Slight. No paradigm shift - it applies well-known excess-mortality reasoning; the 1/300 = MS framing is a nice rhetorical move, not a new idea. B=32.

Real-world impact 2 · Minor. A short rebuttal defending the US COVID-death figure via excess-mortality reasoning; engages a consequential public-health topic but its reach is discourse-internal (correcting skeptical commenters), with no policy or material change → RWI 2.