Skip to content

Edelman's "Myth of the Condorcet Winner," counted — BV2173, live

The tabulatable evidence behind The strongest case against the Condorcet winner. Paul H. Edelman (Vanderbilt, mathematics and law; 22 Supreme Court Economic Review 207 (2015)) argues the Condorcet winner shouldn't be privileged — via a "cancellation" profile that goes back to Condorcet himself (aimed, then, at Borda) and via the no-show/join-consistency theorems. Unlike the FairVote article, this one's math is right — which makes it the perfect steelman: on its own centerpiece profile, every majoritarian count (Ranked Robin, RCV-IRV, STAR's runoff) elects Ada and every positional count (Borda, score sum, Choose-One) elects Ben. Who's "right" is a 240-year-old argument, live.

▶ Live on BetterVoting: vote · results ↗ (election gmfv4c, BV2173) — four races on the same 81 voters: STAR, Ranked Robin, RCV-IRV, Choose-One.

Read the reader-friendly pages (edelman_condorcet_myth_pages/); the .yaml beside each is the tabulatable source.

Page (read this) What it shows src
BV2173 — the 81-voter cancellation profile Ada is the Condorcet winner 41–40; remove the two cyclic blocs and the rest say Ben 28–20 (Borda: 109–101). STAR shows both counts in one method — scoring round Ben 257–233, runoff Ada 41–40. RR/IRV → Ada; Plurality → Ben .yaml
The perfect "Condorcet component," alone 30 voters, a perfectly symmetric cycle: scores tie 70/70/70, pairwise ties 20–10 cyclically, the engine's tiebreak ladder exhausts and flags "[Lot-decided tie — rare]". LH-only (BV would resolve at random — not freezable) .yaml

Concept hubs: Condorcet efficiency · cycle resolution · up: method_comparisons — same ballots, different methods

file: README.md