Ranked choice voting produces better outcomes than plurality voting

Whether ranked-choice voting produces better outcomes than plurality voting turns on several distinct criteria: representation of preferences, participation, campaign quality, administrative performance, and legitimacy. The comparison therefore cannot be resolved merely by asking whether ranked ballots are more expressive; it must also consider which winner-selection principle is preferred, what electoral arrangement RCV replaces, and whether implementation distributes burdens fairly. The central policy question is whether RCV's capacity to preserve backup preferences and consolidate rounds generally outweighs its risks of ballot exhaustion, unequal error, counterintuitive outcomes, and added administrative demands. RCV's strongest comparative advantage is that it records backup preferences and transfers votes after eliminations, reducing the spoiler role that similar candidates can play under plurality voting. Under plurality, voters who prefer a less competitive candidate may have to choose between expressing that preference and influencing the contest among leading alternatives; RCV permits both choices on one ballot by transferring support when the first choice is eliminated. That mechanism improves the ballot's ability to register contingent preferences and makes vote splitting less consequential, although it does not by itself guarantee selection of the candidate with the broadest head-to-head support. RCV also offers a comparatively clear institutional benefit when it replaces a separate runoff, because an instant runoff can combine the initial choice and later-round decision in one election. This structure can eliminate a second trip to the polls, but its practical advantage depends on local rules, equipment, certification procedures, and the kind of runoff or nomination process being replaced. The benefit is consequently more defensible as a comparison with two-round arrangements than as proof that RCV is administratively superior to every one-round plurality election. The claim that ranked ballots are inherently unusable is contradicted by large-scale ballot-record research finding that most voters successfully cast usable RCV ballots. That research distinguishes harmless skips or duplicate rankings from mistakes that actually invalidate a vote, preventing raw irregularity counts from overstating disenfranchisement. Successful ballot use does not eliminate distributional concerns, but it establishes that additional ranking is generally administrable rather than intrinsically prohibitive. RCV may further improve campaign incentives because candidates can benefit from becoming the second or third choice of rival candidates' supporters, and comparative survey evidence found lower perceived negativity in preferential-voting jurisdictions. Survey-based research also associates RCV elections with greater candidate contact and participation among younger voters, supporting the possibility that broader outreach and a more expressive ballot can encourage engagement. These observational findings support plausible campaign and mobilization benefits, but they do not establish that the voting rule alone caused them. Finally, institutional reviews and election-official research indicate that RCV can be administered successfully across real jurisdictions, even though doing so requires preparation and resources (see Figure 1). Operational feasibility strengthens the case that RCV is a workable alternative to plurality, although feasibility alone does not prove superior electoral outcomes. The strongest challenge to RCV's legitimacy claim is ballot exhaustion: the eventual winner may hold a majority only of ballots still active in the decisive round, not of all ballots originally cast. A study of four local RCV elections found that substantial shares of ballots ceased to count before the decisive round because voters ranked too few candidates or did not rank either finalist. RCV therefore supports a precise claim of majority among continuing ballots, but a broader assertion that every winner commands majority support among all participating voters can be misleading. RCV's additional ballot task can also create unequal risks of confusion, incomplete ranking, and invalid voting across demographic groups. San Francisco studies identified demographic disparities in ranking patterns and overvotes, including higher precinct-level overvote rates in communities with larger African American, Latino, elderly, less-educated, and low-income populations. Separate peer-reviewed research reports that understanding of RCV is unevenly distributed and that confusion can reduce effective use of rankings and political participation. These findings do not show that most voters fail, but they establish an equity concern that aggregate ballot-validity rates can obscure. Instant-runoff RCV can also eliminate a candidate who would defeat every rival head-to-head, so transferred preferences do not necessarily identify the broadest compromise winner. Analysis of cast-vote records from Alaska's 2022 special U.S. House election concluded that Mary Peltola won under RCV even though reconstructed preferences indicated that Nick Begich would have defeated each opponent in a pairwise contest. The case does not uniquely vindicate plurality, which likewise does not guarantee the pairwise winner, but it refutes any general claim that RCV necessarily selects the candidate with widest comparative acceptability. RCV additionally has a counterintuitive mathematical vulnerability: in some configurations, ranking a candidate higher can cause that candidate to lose. The cited mathematical analyses establish the possibility of monotonicity failure rather than its frequency in actual elections, so this objection limits claims of intuitive preference translation more than it demonstrates widespread practical harm. Administration and public reporting are more demanding under RCV than under ordinary plurality tabulation because jurisdictions must address compatible equipment, ballot design, testing, voter education, transfer rules, audits, recounts, certification, and explanation of results. Administrator-focused evidence further documents burdens involving ballot handling, tabulation, communication, and delayed results, while finding that effects on cost and reporting speed vary with state law and existing systems. Any savings from eliminating a separate runoff must therefore be weighed against the transition and continuing costs of administering a more complex tabulation rule. The effect of RCV on participation is conditional rather than generally positive: some research reports mobilization associations, while comparative mayoral analysis finds no consistent turnout increase and possible declines in some contexts. The mayoral evidence is itself difficult to interpret causally because adoption may coincide with other institutional changes and city characteristics. RCV may consequently compare favorably with a separate low-participation runoff yet be neutral or unfavorable relative to a single salient plurality election; the baseline institution is indispensable to the turnout judgment. Campaign civility is likewise a plausible but unsettled benefit: earlier comparative survey evidence found less perceived negativity under preferential voting, whereas a later peer-reviewed reassessment found that the conclusion depends on the cases and measures used. The evidence supports an incentive for candidates to seek later-choice support, but not a demonstrated general reduction in negative campaigning. Implementation quality is a major boundary condition because evidence that most voters cast usable ballots coexists with evidence of unequal confusion and error, while administrator studies identify consequential variation in preparation and capacity. Ballot layout, permitted ranking depth, voting equipment, education, and tabulation capacity affect rejection risk and administrative performance, so results from one jurisdiction need not transfer unchanged to another. Well-designed implementation can mitigate RCV's cognitive and administrative costs, but the evidence does not support treating those costs as either inevitable or negligible. The judgment also depends on the chosen definition of a better winner: RCV prioritizes sequential elimination and transferred rankings, plurality prioritizes first choices, and a Condorcet standard prioritizes head-to-head support. RCV reduces plurality's vote-splitting problem, yet exhaustion and Condorcet failure show that it exchanges rather than eliminates all winner-selection pathologies. The principal evidentiary limitation is not the absence of either side, but the difficulty of isolating voting-rule effects from election type, jurisdictional context, implementation choices, and contemporaneous institutional changes. The bundle does not supply a single common metric that aggregates preference representation, turnout, civility, equity, cost, reporting speed, and legitimacy, so the overall judgment necessarily depends partly on normative weighting. Evidence concerning turnout and civility remains causally limited or conflicting, while evidence about rare mathematical failures identifies possibility more clearly than prevalence. The supplied structural classification also flags unresolved conflict-of-interest classifications, which weakens confidence in source independence even though the substantive evidence spans peer-reviewed analyses, data studies, and institutional reports. More credible causal comparisons across matched jurisdictions, longer-term studies of adaptation, subgroup analyses under different ballot and equipment designs, and standardized measurements of campaign conduct and legitimacy would most directly narrow the remaining uncertainty. On balance, the evidence supports a qualified rather than general verdict: RCV offers real advantages over plurality in preserving backup preferences, reducing spoiler pressure, and replacing separate runoffs, but it has not been shown to improve participation, campaign quality, representation, and legitimacy across contexts. Confidence in that conditional assessment is high because multiple source types converge on both RCV's workable expressive benefits and its implementation-dependent costs, while also documenting exhaustion, equity risks, and possible winner-selection failures. The dominant substantive uncertainty is contextual heterogeneity—especially what system RCV replaces and how well it is implemented—supplemented by unresolved conflict-of-interest classifications in the source bundle.