In social choice theory, evaluating the quality of a voting system often begins with a clean mathematical abstraction: voters possess true, underlying preferences—frequently modeled as ordinal rankings or distances within a metric space—and they report these preferences honestly. Under this sincere preference assumption, researchers can derive elegant performance guarantees. For instance, we can bound an election's distortion—the worst-case ratio between the social cost of an elected winner and that of the socially optimal candidate.
However, this assumption creates a significant gap between theory and reality. In practice, voting systems are games. When participants realize that reporting their true preferences might lead to an undesirable outcome, they act strategically. As explored by Mohammad Abam et al. in Tight Bounds on the Distortion of Randomized and Deterministic Distributed Voting, maintaining low distortion bounds becomes a profound challenge when voters intentionally misrepresent their rankings to secure a more favorable collective outcome. Similarly, Negar Babashah et al. note in Distortion of Multi-Winner Elections on the Line Metric: The Polar Comparison Rule that the theoretical robustness of committee selection rules remains an open question when the assumption of truthful ordinal reporting is discarded.
Mechanics of Strategic Voter Manipulation
Why do voters deviate from sincerity? The intuition is rooted in game-theoretic optimization: a sub-optimal honest ballot is often less useful than a calculated, insincere one.
Tactical Voting and the Lesser Evil
When a voter’s favorite candidate has no realistic chance of winning, voting sincerely can feel like a wasted effort. Instead, voters engage in tactical voting, shifting their support to a "lesser of two evils" who has a viable path to victory. Elliot Anshelevich et al. demonstrate in Approximating Optimal Social Choice under Metric Preferences that this behavior directly threatens the validity of standard distortion models. Even when candidate pools are highly representative of the population—which theoretically optimizes majority rule elections (Yu Cheng et al. in Of the People: Voting Is More Effective with Representative Candidates)—tactical voting for viable candidates can distort the expected efficiency of classic rules like the Borda Count or Plurality (Yu Cheng et al. in On the Distortion of Voting With Multiple Representative Candidates).
Strategic Truncation and Reordering
Even structural constraints designed to simplify voting can induce strategic behavior. In Instant Runoff Voting (IRV), jurisdictions often limit the ballot length, forcing voters to rank only a subset of candidates. While models often assume "sincere truncation"—where voters simply list their top choices in order—the authors of Ballot Length in Instant Runoff Voting demonstrate that ballot limits actually incentivize voters to strategically reorder candidates, a phenomenon that can completely flip election outcomes.
Multi-Issue and Sequential Distortions
The problem compounds when decisions are broken down into parts. In sequential voting, where issues are decided one by one, voters may strategically misrepresent their preferences on early issues to manipulate the voter landscape for later, more critical issues (Lirong Xia et al. in A Dichotomy Theorem on the Existence of Efficient or Neutral Sequential Voting Correspondences). This vulnerability allows agenda-setters to execute "bundling attacks," combining or ordering issues to exploit the predictable strategic adjustments of the electorate (Noga Alon et al. in Bundling Attacks in Judgment Aggregation). Even when designing multi-member districts to combat gerrymandering, models must account for how voters manipulate their rankings to maximize partisan advantage rather than expressing pure preferences (Nikhil Garg et al. in Combatting Gerrymandering with Social Choice: the Design of Multi-member Districts).
Beyond the Ballot Box: Primaries, Proxies, and Committees
Strategic behavior is not confined to standard single-winner elections; it permeates every layer of democratic and representative architecture.
Primary Raiding
In multi-stage electoral systems, strategic behavior can cross party lines. Allan Borodin et al. in Primarily about Primaries highlight the phenomenon of "primary raiding," where voters strategically participate in an opposing party's primary to nominate a weaker, more beatable candidate, undermining the social utility of the primary system.
Strategic Delegation
In proxy voting and liquid democracy, the sincere assumption dictates that voters delegate their power to the proxy closest to them in the preference space. However, the authors of Representative Proxy Voting show that voters may deliberately select a more distant proxy to strategically manipulate the final outcome or influence Condorcet consistency. Similarly, in ranked delegation frameworks, voters optimize their rankings based on a delegate's perceived likelihood of success and systemic influence rather than pure, sincere trust (Markus Brill et al. in Liquid Democracy with Ranked Delegations).
Institutional Constraints and Coalitions
Even when looking at real-world voting data to infer true ideologies, strategic behavior obscures the truth. For example, Keyon Vafa et al. in Text-Based Ideal Points explain that legislative voting records are heavily distorted by external institutional constraints like party-line pressure and reciprocal logrolling agreements, making it difficult to isolate a lawmaker's genuine ideological stance from strategic compliance.
When representative bodies are formed via sortition (random selection), models assume members will vote according to their true preferences; yet external incentives like lobbying can quickly distort this representativeness (Reshef Meir et al. in Representative Committees of Peers). Furthermore, when groups of voters form coalitions to manipulate multiwinner shortlists, their strategic choices depend heavily on whether they hold optimistic or pessimistic attitudes toward tie-breaking rules (Andrzej Kaczmarczyk et al. in On coalitional manipulation for multiwinner elections: shortlisting).
The Strategic Behavior of Candidates
An election is a two-sided game: as voters optimize their ballots, candidates optimize their entry and positioning. The assumption that candidates enter a race purely based on fixed ideological motivations ignores the mechanics of strategic candidacy.
Candidates frequently decide to run or withdraw solely to alter the voting equilibrium and manipulate who ultimately wins. As shown by Maria Polukarov et al. in Convergence to Equilibria in Strategic Candidacy, these candidacy games are highly unstable, though stability can be recovered when voters hold grudges against strategic dropouts. Furthermore, when models incorporate realistic candidate traits—such as "laziness" (a reluctance to run if it doesn't change the winner)—the computational complexity of finding a stable equilibrium skyrockets to NP-complete (Svetlana Obraztsova et al. in Strategic Candidacy Games with Lazy Candidates).
Svetlana Obraztsova et al. emphasize in Strategic candidacy with reluctance that ignoring the interplay between strategic candidate entry and tactical voting leaves a major blind spot in our understanding of electoral stability. For instance, while Instant Runoff Voting is mathematically shown to possess a "moderating effect" by establishing exclusion zones that block extremists, the open question remains how competitive candidate entry and tactical voting alter this moderation in practice (Kiran Tomlinson et al. in The Moderating Effect of Instant Runoff Voting).
Modeling Hybrid Realities
To build robust voting rules, modern social choice theory is moving away from the binary of purely sincere or purely strategic populations. Instead, researchers model hybrid environments.
[ Voter Population ]
│
┌─────────────────┴─────────────────┐
▼ ▼
[ Sincere / Principled ] [ Strategic / Tactical ]
• Always vote true preference • Vote for the "lesser evil"
• Provide baseline stability • Reorder limited ballots
• Strategic abstention
By introducing "principled" or "truth-biased" voters alongside "lazy" or fully strategic agents, frameworks can map out realistic Nash equilibria (Edith Elkind et al. in Equilibria of Plurality Voting: Lazy and Truth-biased Voters). Interestingly, accounting for natural behavioral patterns—such as strategic abstention due to alienation or indifference—can paradoxically reveal that real-world tactical deviations sometimes lower the worst-case social cost, shifting the theoretical distortion bounds in unexpected ways (Mohammad Ghodsi et al. in On the Distortion Value of the Elections with Abstention; Mohammad Ghodsi et al. in On the Distortion Value of the Elections with Abstention).
Ultimately, designing resilient mechanisms requires moving past the sincere preference assumption and directly embedding game-theoretic optimization into the architecture of social choice.
Key Foundational Work
A Dichotomy Theorem on the Existence of Efficient or Neutral Sequential Voting Correspondences (2009) — contributed 1 concept elements to this direction.
Stackelberg Voting Games: Computational Aspects and Paradoxes (2010) — contributed 1 concept elements to this direction.