Reporting true ties

One line: a true tie is when two candidates are genuinely equal at a step — same score, or an equal split in the runoff. This page is about how a tie shows up in the reports; the mechanics of breaking one (the full ladder, lot order, imported tieBreakType) are the canonical STAR Tie-Breaking deep-dive.

→ Hub: STAR Reporting · How the LH engine reports · How BetterVoting reports · imported tie data: tie-break JSON.

---

How a tie reads in the LH report

The engine marks the tied rows and says so in words, then resolves them by the tie-break ladder. A runoff that splits evenly:

Automatic Runoff Round
 The candidate preferred in the most head-to-head matchups wins.
   Choco         -- 1 -- Tied for first place
   Vanilla       -- 1 -- Tied for first place
 There's a two-way tie for first.

— here broken by the Scoring Round (Vanilla had the higher total), exactly the ladder's first rung. (Worked end-to-end, with ties in both rounds: Tie-Breaking — the Ice Cream example.)

Reporting detail: on an exact runoff tie the one-line runoff-percentage summary (Voters with a preference: …) is suppressed — a 50/50 split has no "winner vs other" to state, so the tie-break chain explains the result instead. The percentage line returns only for a decided runoff.

How a tie reads in BetterVoting

BetterVoting records how a tie was settled in its result data as tieBreakType, and shows the resolved order on the result card. (For how an imported BetterVoting tie-break maps into a YAML election's lot order, see tie-break JSON.)

Why ties feel common in tiny examples

Teaching elections use the fewest ballots that make a point, and small electorates tie easily (a 1–1 runoff needs just two voters). That's a feature for showing the tie-break ladder — and a reason real public elections, with thousands of ballots, almost never reach the lot-drawing floor. (→ Tie-Breaking — quick rules.)