Condorcet Voting Paradox: Why No Voting System Is Perfect

simulator intermediate ~10 min
Loading simulation...
Different methods → different winners

With 15 voters and 3 candidates, the winner often depends on which voting method is used. Plurality, runoff, Borda count, and Condorcet comparison can each select a different candidate from the same set of ballots — demonstrating Arrow's impossibility theorem.

Formula

Borda score for candidate i = Σ_voters (rank position points)
Condorcet winner: candidate C where C >majority D for all D ≠ C

Democracy's Mathematical Problem

In 1785, the Marquis de Condorcet discovered something troubling: when three or more candidates compete, majority preferences can form a cycle. Voters might prefer Alice over Bob, Bob over Carol, and Carol over Alice — each by majority vote. There is no 'will of the people' in such cases; collective preferences are irrational even when every individual voter is perfectly rational. This paradox lies at the heart of social choice theory.

Four Voting Methods, Four Answers

This simulation lets you compare four common voting methods on the same set of ballots. Plurality counts only first-choice votes. Instant runoff eliminates the last-place candidate and redistributes their votes. Borda count assigns points by ranking position. Condorcet comparison checks every pairwise matchup. The striking result: these methods often produce different winners from identical voter preferences. The 'right' outcome depends on which method you choose.

Arrow's Impossibility Theorem

In 1951, Kenneth Arrow proved mathematically that no ranked voting system can satisfy a small set of reasonable fairness criteria simultaneously. Every system must violate at least one of: treating all voters equally, respecting unanimous preferences, or being immune to irrelevant alternatives. This Nobel Prize-winning theorem means the search for a 'perfect' voting system is not just difficult — it is mathematically impossible.

Strategic Voting Is Inevitable

The Gibbard-Satterthwaite theorem (1973) adds another blow: any non-dictatorial voting system with three or more candidates can be manipulated by strategic voting. A voter can sometimes get a better outcome by lying about their preferences — voting for a less-preferred candidate to block a worse one. This is not a bug in specific voting systems; it is a fundamental property of collective decision-making with ranked preferences.

FAQ

What is the Condorcet paradox?

The Condorcet paradox (1785) shows that majority preferences can be cyclic: voters may prefer A over B, B over C, and C over A. This means there is no candidate that a majority prefers to all others. It demonstrates that individual rationality does not guarantee collective rationality.

What is Arrow's impossibility theorem?

Arrow's theorem (1951) proves that no ranked voting system for three or more candidates can simultaneously satisfy all of: unrestricted domain, non-dictatorship, Pareto efficiency, and independence of irrelevant alternatives. In simple terms, every voting system has fundamental flaws — there is no perfect way to aggregate individual preferences into a group decision.

What is the difference between plurality and Borda count?

Plurality (first-past-the-post) only counts each voter's top choice — the candidate with the most first-place votes wins. Borda count assigns points based on ranking position (last place = 0, first place = n−1). Borda rewards broad appeal; plurality rewards intense support from a faction. They can easily produce different winners.

Can strategic voting be prevented?

No. The Gibbard-Satterthwaite theorem proves that any non-dictatorial voting system with three or more candidates is susceptible to strategic (insincere) voting. Some systems are more resistant than others, but no ranked system is strategy-proof.

Sources

Embed

<iframe src="https://homo-deus.com/lab/sociology/voting-paradox/embed" width="100%" height="400" frameborder="0"></iframe>
View source on GitHub