election_title: "Kissel's five-way example (STAR) — the compromise reaches the runoff and wins"
scenario_description: >-
  The same electorate as …_irv.yaml and …_rr.yaml — the five-candidate field from p.5
  of Adam Kissel's "Can Ranked-Choice Voting Work? A Conservative Approach" — on a 0-5
  score ballot instead of a ranking. The paper concedes that no form of RCV counts the
  A and B voters' second choices, and floats "an even more complicated point system"
  as a possible fix while dismissing it as voter confusion. That fix is STAR, and it
  is one grid on the ballot: the scoring round reads every ballot, C leads it (3221 to
  A's 2695), and the automatic runoff confirms the head-to-head the paper says is
  never revealed — C over A, 511-489. The scores are the paper's own preference order
  expressed as strengths; the ranked files hold the identical order.
bv_test_id: BV2278
bv_election_id: 8cdkkc
bv_results_url: https://bettervoting.com/8cdkkc/results
voting_method: STAR
num_winners: 1
# Scores: A 2695, B 1884, C 3221, D 1557, E 45  -> finalists C and A.
# Runoff: C 511, A 489 — the same head-to-head the pairwise matrix shows.
ballots: |-
  Count:A,B,C,D,E
  306:5,0,3,1,0     # A-partisans — moderate C is their second choice
  300:0,5,3,1,0     # B-partisans — moderate C is their second choice too
  111:3,1,5,0,0     # moderates leaning A
   91:1,3,5,0,0     # moderates leaning B
  183:4,0,2,5,0     # D's voters lean A
    9:1,0,3,4,5     # the <1% candidate the paper says is eliminated first
expected_winners:
  - C

# file: bv2278_8cdkkc_five_way_star.yaml