Skip to content

Weak ranks — equal ranks allowed

A weak ranking lets you mark two candidates equal — "I like these two the same." (Formally: a weak order, or "order with ties.") Ranked Robin and the other Condorcet methods allow it; RCV-IRV (Hare) does not.**

→ The full contrast: strict vs. weak ranks · the other kind: strict ranks


What it means

On a weak-ranked ballot you may give two (or more) candidates the same rank. You're never forced to invent a preference you don't feel — if two candidates are equally good to you, you say so.

A weak ranked ballot: five candidates on rows, bubble columns 1st through 5th. Andre is marked 1st, Carmen 2nd, David 3rd, and both Blake and Ella are marked 4th — two filled bubbles in one column. The 5th column is empty.

That's the same voter as on the strict ballot, with exactly one mark moved: Ella slides from 5th up to 4th, joining Blake. Two filled bubbles in one column is the whole difference — and the 5th column empties out, because once two candidates share 4th there is no 5th place to give.

In the PrefLib data taxonomy these are the TOC / TOI types (Ties allowed, Complete or Incomplete).

Which ranked methods allow it? Condorcet methods — Ranked Robin, Schulze, Ranked Pairs, Minimax — allow equal ranks and compare candidates head-to-head. Borda and Bucklin usually allow ties too. RCV-IRV (Hare) and STV do not — on those a tie is an overvote.

Why it matters

  • Honest indifference. A voter can express "no preference" between two candidates instead of guessing an order.
  • More expressive, less noise. Forcing a strict order records distinctions voters don't actually feel; equal ranks remove that pressure.
  • It's what people wrongly assume RCV allows. Marking ties is a feature of other ranked methods, not the RCV-IRV that's on US ballots — a common surprise.

Weak ranks still only capture order, not strength — a second choice "could be as good as your favorite or almost as bad as your last choice," and even a weak ballot can't tell the difference. Scores can: scores vs. ranks.

file: weak_ranks.md