Conviction Voting

Unlike snapshot voting, Conviction Voting is a continuous process where a user's voting power for a proposal grows over time as long as their tokens remain staked on it. This creates a time-weighted signal that measures sustained, rather than momentary, community conviction. Key dynamics include:

• Decay: Conviction decays if a user withdraws support or tokens are moved.
• Patience Premium: Long-term supporters have greater influence.
• Real-time Signal: The aggregated conviction for all proposals is always visible.

Proposals compete for a shared funding pool (e.g., a community treasury). A proposal only passes and receives funds when its total accumulated conviction crosses a dynamic activation threshold. This threshold is often calculated as a function of:

• The total size of the funding pool.
• The total conviction staked across all active proposals.
• A configured alpha or beta parameter that controls the threshold's sensitivity. This ensures funds are only released when there is sufficient, sustained demand relative to available resources.

Many Conviction Voting implementations incorporate quadratic scaling to reduce plutocratic influence. A user's conviction accumulation rate is often proportional to the square root of the number of tokens they stake, not the linear amount. This means:

• Diminishing Returns: One user staking 100 tokens gains less than 10x the conviction of 10 users each staking 1 token.
• Sybil Resistance: It encourages broader participation over concentrated wealth.
• Preference Revelation: More accurately aggregates the intensity of preference across a diverse group.

The lifecycle of a proposal is governed by the flow of conviction. A user can:

• Stake: Add tokens to a proposal to begin accumulating conviction.
• Withdraw: Remove tokens at any time, causing the associated conviction to decay rapidly.
• Switch: Move staked tokens from one proposal to another without a withdrawal penalty in some implementations. This fluidity allows the community signal to continuously evolve in response to new information or proposals, creating a dynamic market of ideas competing for funding.

DAO administrators configure Conviction Voting using key parameters that define its behavior:

• Decay Rate: How quickly conviction dissipates after support is removed (e.g., half-life).
• Max Ratio: The maximum percentage of the funding pool a single proposal can request.
• Spending Limit: The minimum time between funded proposals to ensure treasury sustainability.
• Minimum Stake: The smallest token amount required to participate. These parameters are critical for balancing agility, safety, and long-term treasury health.

While pioneered for retroactive funding and grant allocation, the mechanism is generalizable:

• Prioritization: Ranking features in a product roadmap.
• Delegation: Signaling preference for oracle data providers or validators.
• Parameter Governance: Continuously tuning system variables (e.g., fee rates).
• Content Curation: Ranking or filtering information in decentralized media platforms. Its core utility is in any scenario requiring persistent, weighted sentiment aggregation.

Every conviction voting implementation configures key parameters that define its behavior and security model.

• Decay Rate: Determines how quickly conviction scores decrease over time, affecting proposal velocity.
• Max Ratio: Limits the percentage of the total commons budget a single proposal can request.
• Spending Limit: A function (often a logistic curve) that sets the conviction threshold required for a given funding amount, preventing large, quick withdrawals.
• Minimum Stake: The smallest token amount required to create a proposal, reducing spam.

The core security model relies on time-locked capital to prevent Sybil attacks. A malicious actor cannot cheaply create many identities to sway a vote, as each requires a separate, time-accumulating stake. The conviction score grows logarithmically with time, making long-term commitment more valuable than a large number of short-term votes. This aligns voter incentives with the long-term health of the treasury or commons being governed.

Key attack vectors include:

• Proposal Spam: Flooding the system with proposals to dilute attention. Mitigated by requiring a minimum stake or conviction threshold for a proposal to be fundable.
• Last-Minute Swing Attacks: A whale could deposit a large stake at the last moment. Mitigated by the logarithmic growth curve, which heavily penalizes new stakes and rewards sustained conviction.
• Collusion & Bribery: While still possible, it is made more expensive and detectable due to the transparent, on-chain nature of time-locked commitments.

The system incentivizes a Nash equilibrium where voters allocate their conviction to proposals they genuinely believe in. The game-theoretic design discourages "hot-potato" voting or frequent switching because:

• Moving conviction resets the accumulation timer, incurring an opportunity cost.
• Voters are motivated to research and back proposals early to maximize their influence over time. This leads to more stable, considered funding outcomes rather than reactionary swings.

Unlike one-token-one-vote systems that lock capital for fixed periods, conviction voting allows for asynchronous unbonding. A voter can signal to withdraw their funds, which triggers a decay period (e.g., several days) where their conviction linearly decreases to zero. This prevents sudden, disruptive capital flight while allowing liquidity to exit. It creates a smoother, more predictable ebb and flow of committed capital in the governance system.

Contrasts with other common mechanisms:

• vs. Token Voting (1p1v): More resistant to whale dominance, as sustained conviction from many small holders can outweigh a large, new stake. Does not require locking all capital in a single snapshot.
• vs. Quadratic Voting (QV): Both aim to reduce plutocracy. Conviction voting uses time preference instead of cost curvature to achieve similar goals. QV is better for one-off decisions; conviction voting is designed for continuous resource allocation (like a treasury).

A critical comparison of governance models. Snapshot voting is a common one-token-one-vote system for binary, time-bound decisions.

• Snapshot: Fast, simple polls. Prone to whale dominance and short-term manipulation.
• Conviction Voting: Slower, preference signaling. Rewards long-term commitment and mitigates snap attacks.
• Hybrid Use: Many DAOs use Snapshot for high-level signaling and conviction for continuous budgeting.

An emerging design pattern extending conviction voting across multiple blockchains. This allows a DAO's treasury and governance to be chain-agnostic.

• Technology: Relies on cross-chain messaging protocols (e.g., LayerZero, Axelar) to sync voting states.
• Benefit: Enables unified governance for multi-chain treasuries and ecosystems.
• Complexity: Introduces challenges in finality, security, and gas cost management.

The tunable variables that define a conviction voting system's behavior and security.

• Spending Limit: The maximum percentage of the treasury that can be allocated at once.
• Decay Factor: How quickly conviction dissipates when funds are moved; controls proposal velocity.
• Minimum Conviction Threshold: The required conviction level for a proposal to pass.
• Proposal Duration: The time over which conviction can accumulate.