Why Quadratic Funding Needs a Prediction Market Backstop

Pure QF assumes unique human identities. In practice, creating fake accounts (Sybils) is trivial, allowing malicious actors to distort matching fund allocation with minimal capital. This flaw is foundational, not incidental.

• Cost of Attack: As low as ~$1 per Sybil identity on many chains.
• Impact: A project with 100 fake $1 donations can outmatch one with 10 real $100 donations.

Small donors bear zero financial consequence for being wrong. This leads to low-effort, sentiment-driven voting rather than rigorous due diligence. The system incentivizes popularity contests, not optimal capital allocation.

• Data Point: <1% of donors perform meaningful project research.
• Result: Meme projects and well-marketed scams consistently outperform substantive but niche public goods.

Integrating a prediction market (e.g., Polymarket, Augur) forces participants to stake capital on outcomes. This creates a financially incentivized oracle that prices the real-world impact and legitimacy of QF-funded projects, backstopping the naive donor model.

• Mechanism: Markets resolve on verifiable metrics (e.g., "Project X deploys mainnet by date Y").
• Outcome: Sybil attacks become unprofitable as market shorts punish fraudulent projects.

QF amplifies small contributions, making it a prime target for Sybil attacks where a single entity creates many wallets to sway results. Low-cost voting also attracts uninformed participants, diluting signal.

• Current cost to manipulate: Minimal gas fees.
• Impact: Public goods funding is misallocated, undermining the entire Collective's mission.

Integrate a decentralized prediction market (e.g., Polymarket, Augur) where participants can stake on the long-term impact of a funded project. This creates a financial market for project quality.

• Mechanism: Bets resolve on verifiable, on-chain metrics (e.g., contract deployments, user growth).
• Outcome: Generates a price signal that is expensive to manipulate and reflects informed, high-conviction capital.

RetroPGF 4 runs in two phases. Phase 1 uses standard QF for broad, low-friction discovery. Phase 2 takes the top projects and subjects them to a prediction market for final ranking and allocation weighting.

• Phase 1 (QF): Democratic signal, high participation.
• Phase 2 (PM): Capital-efficient truth discovery, high-stakes validation.
• Result: Combines inclusivity of QF with the robustness of a financial oracle.

Prediction markets need unambiguous resolution. An oracle or decentralized court (e.g., UMA's Optimistic Oracle, Kleros) is required to adjudicate impact metrics, preventing market stalemates.

• Role: Resolves subjective claims about project impact into binary outcomes.
• Critical Design: Must be sybil-resistant and economically secure to prevent final-stage attacks.

Successful predictors profit, attracting more sophisticated analysts. This creates a virtuous cycle where the market's accuracy improves over time, making RetroPGF allocations increasingly efficient.

• Incentive: Profit from identifying undervalued public goods.
• Long-term Effect: The prediction market becomes a high-signal discovery layer for the entire Optimism ecosystem, beyond just funding.

Continuing pure QF leads to inevitable governance capture and funding entropy. The system becomes a target for sophisticated attackers, while legitimate builders lose faith. Compare to Gitcoin Grants, which constantly battles Sybil farms.

• Risk: RetroPGF's $1B+ endowment becomes inefficiently deployed.
• Outcome: The "Impact = Profit" thesis for public goods fails its largest experiment.

Without a cost to create fake identities, attackers can manipulate QF rounds by coordinating small donations across thousands of wallets to drain the matching pool.

• Sybil resistance is the primary unsolved challenge for on-chain QF.
• Current solutions like BrightID or Gitcoin Passport add friction but aren't natively crypto-economic.
• A purely social solution fails at the $100M+ matching pool scale where financial incentives dominate.

Use futarchy-style mechanisms where market prices adjudicate the quality of a project or the legitimacy of a donor.

• Polymarket or Manifold-style markets can be used to stake on sybil status.
• Creates a financial disincentive for bad actors: attacking requires betting against the market.
• Shifts the game theory from identity verification to skin-in-the-game verification.

Existing prediction market platforms become decentralized truth oracles for QF rounds.

• Markets resolve on questions like: "Is this donor cluster a sybil?" or "Will this funded project deliver a working product?"
• Liquidity providers and traders are incentivized to uncover fraud for profit.
• Creates a self-funding security layer via trading fees, unlike passive identity verifiers.

Protocols that implement QF must design their smart contracts to consume prediction market outcomes as a core input.

• Matching fund distribution becomes conditional on a market-resolved legitimacy score.
• Enables retroactive funding rounds where markets first predict impact, then funds follow.
• Turns QF from a one-round game into a continuous, market-adjusted process.

QF's core vulnerability is cheap identity forgery, turning it into a capital efficiency contest. A prediction market backstop transforms this into a costly signaling game.

• Sybil cost shifts from ~$0 to the cost of losing a bet.
• Attackers must stake capital on the long-term failure of the project they're trying to fund.
• Creates a crypto-economic firewall without centralized identity checks.

Platforms like Polymarket or Augur provide the necessary decentralized oracle. They answer the binary question: "Did this funded project deliver measurable value?"

• Bettors are incentivized to research and price real outcomes, not sentiment.
• Losing side's stake is slashed to penalize fraudulent funding campaigns.
• This creates a continuous validity bond for every QF round.

The classic QF formula is gamed. A PM-backed system weights contributions by the market's confidence in the project's future success.

• Matching pool is distributed based on a combination of quadratic contribution and outcome market odds.
• Grants with high odds of success receive super-linear matching, grants with low odds are filtered out.
• This aligns Gitcoin Grants mechanics with Venture Capital diligence, but in a decentralized, transparent market.