Scott Alexander’s long-form (15-hours-of-debate) review, concluding zoonosis wins decisively and reporting his own post-debate Bayesian estimate of ~90-10 in favor of zoonosis (a Good Judgment Project/superforecaster sub-study separately reached ~75-25 zoonosis). Its most consequential contribution to this case is a comparison table/figure of six people’s independent Bayesian point-estimates of the same debate evidence — Peter Miller, Saar Wilf/Rootclaim, judges Will van Treuren and Eric Stansifer, Alexander himself, and Daniel Filan (who also watched the debate) — which the piece reports “span twenty-three orders of magnitude,” the figure quoted in this case’s framing. Alexander explicitly excluded Michael Weissman’s analysis from that table (“it would have taken too long to translate his language into mine”). Also links a blank Bayesian-calculator spreadsheet template for readers to submit their own estimates.
relevance_note: the primary source of the case’s framing claims about the debate’s outcome and the “23 orders of magnitude” comparison, and the highest-circulation single write-up of the whole debate.