How Borda Count Crowns a Loser
Borda count collapses the *ordering* of preferences into *points*, so the size of the margin at each rank — not just the ranking itself — decides the winner, and this lets a candidate who is last against every rival still win overall.
A complete interactive classroom, not just a preview.
Start when you are ready to enter this Stage's 8 scenes and explore, respond, and learn as you go.
Why can Borda count produce a Condorcet loser through preference reversals?
An election method trusted for centuries can elevate the candidate every voter prefers least.
Borda count feels intuitive because it tallies ranked preferences — yet it can crown a Condorcet loser, contradicting the will of a majority.
A concrete 5-voter, 3-candidate ballot profile that shows a Condorcet loser winning the Borda tally, followed by an interactive widget where learners adjust preference intensity to reproduce the reversal.
A precise mechanism — preference intensity is amplified and reversed at the extremes — that explains why positional methods like Borda can defy majority will.
Most learners assume that if a candidate beats every other candidate head-to-head, any reasonable tally should agree — so the surprise is that Borda can disagree.
- Full taxonomy of voting paradoxes beyond the Condorcet loser
- Other positional rules like Plurality or Anti-Plurality
- Strategic voting and manipulation in detail
- Social choice axioms and impossibility theorems beyond a passing reference
- 01A Majority Loser Can Still WinslideQuestion
Open with the paradox: introduce Borda count, define a Condorcet loser, and pose the driving question that the investigation will resolve.
- Borda count awards 2 points for 1st, 1 for 2nd, 0 for 3rd
- A Condorcet loser is beaten by every other candidate head-to-head
- Driving question: How can the 'loser of all matchups' still win the tally?
- 02Commit to a First GuessquizPrediction
Ask the learner to pick which mechanism they think produces the Condorcet loser before any evidence is shown.
- Choose between majority logic, point arithmetic, ballot format, or strategic voting
- Commit before seeing the worked example
- 03A 5-Voter Ballot That Breaks the IntuitionslideEvidence
Display a concrete profile: 5 voters, 3 candidates A, B, C, with ballots arranged so each voter ranks a different candidate last. Tally pairwise winners and Borda points side by side.
- Each candidate is last for some voters, so pairwise margins vary in size
- Show head-to-head tallies: every candidate loses to at least one rival
- Show Borda tally: the candidate with the most mid-rank points wins
- 04Tune the Margins and Watch the Winner FlipinteractiveEvidence
A simulation widget where learners adjust the size of each pairwise margin by moving a slider for how many voters place candidate C last versus last-second. Recompute Borda and pairwise outcomes live.
- Move the 'C-last' voter count and observe Borda winner change
- Watch pairwise loser status update in real time
- See the exact interval where the Condorcet loser wins the Borda tally
- 05Points Capture Intensity, Rankings Capture Only OrderslideExplanation
Explain the mechanism: Borda treats the 2-to-1 margin the same as the 2-to-0 margin, because both award 2 points for first. The size of a pairwise loss disappears into a single rank.
- Borda collapses an ordered list into a sum of fixed rank values
- Pairwise margins are erased: a 4–1 loss and a 2–1 loss both give 0 points to the loser
- A candidate can be 'crushingly last' to many voters and still gain points from other ranks
- 06Apply the Idea to a Changed ElectioninteractiveTransfer
Present a new 7-voter, 4-candidate profile and ask the learner to predict which candidate Borda will crown and whether that candidate is a Condorcet loser, before the widget reveals the answer.
- Use a four-candidate profile so the mid-rank effect amplifies
- Learner predicts winner and loser status before computing
- Widget confirms or overturns the prediction
- 07When Does the Paradox Disappear?slideBoundary
Mark the boundary conditions: with two candidates, Borda equals majority rule, so no Condorcet loser can arise; with near-unanimous rankings, Borda and pairwise agree; the paradox requires genuine preference dispersion across three or more candidates.
- Two-candidate elections: Borda reduces to plurality, no paradox possible
- Highly correlated preferences: Borda and pairwise converge
- Three or more candidates with dispersed rankings: paradox is available
- 08The Answer in One SentenceslideResolution
Resolve the driving question directly: Borda rewards the *shape* of a ranking across the ballot, not the *direction* of pairwise contests, so a candidate who is mildly favored on a few rankings can outscore a candidate who is harshly opposed on one.
- Restate the mechanism: fixed points per rank ignore margin size
- Connect back to the evidence and transfer scenes
- Frame the broader lesson: ordinal methods can violate majority intuition
Discussion threads for a Stage aren't available yet.
This path ends here.
