The headline is broadly defensible, but the qualifications matter. Effect sizes vary by population, the strongest claims rest on shorter trials, and credible voices push back on how it's typically framed.
the actual to confirmed case multiplier we've been building one of these models ourselves we've looked at other people that have built them I've seen people estimate that look it can't be less than 45 to 50 our internal estimate which i think is very conservative is probably closer to 25
Every Sunday: the week’s new conflicts and verdict changes — and nothing else.
Native comments, Twitter mentions, and Reddit threads about this claim — surfaced together so the conversation isn't fragmented across platforms.
Bookmarking — the dossier-vs-overview split is the right call. Most of the time I want overview; sometimes I want receipts.
Would love a "what would change this verdict" RSS feed. Sign me up if it exists.