Recently, I delved into the Futarchy experiment that Optimism conducted last year. To be honest, this was quite interesting and also sparked some thoughts on the direction of DAO governance.

In simple terms, Futarchy is using prediction markets to make governance decisions. At that time, Optimism invested 500k OP tokens, allowing participants to use simulated tokens to predict which projects would see the greatest TVL growth after receiving 100k OP incentives. Sounds ideal, right? Let the wisdom of the market guide resource allocation. But the experiment revealed a bunch of real-world issues.

First and foremost is the metric design. They used USD-denominated TVL to measure project performance. The result? Rocket Pool predicted a growth of $59.4 million, but the actual growth was zero; SuperForm predicted $48.5 million, but it actually declined by $1.2 million. The total TVL of the five projects selected by Futarchy actually dropped by $15.8 million. The reason is simple—ETH price fluctuations directly affected the TVL figures. Participants were essentially betting on the token price rather than evaluating project capabilities. It’s like measuring with a deforming ruler that can’t reflect the true situation.

The user experience was also poor. 41% of participants joined only in the last three days, indicating high learning costs. Each prediction required six on-chain interactions, with an average of only 13.6 transactions per person. Out of 2,262 visitors, only 19% actually participated, and the participation rate among OP governance contributors was just 13.48%. This high friction directly killed market liquidity and prediction accuracy.

Another interesting phenomenon is that Badge Holders, considered OP ecosystem experts, ranked the lowest in trading win rates. Conversely, anonymous accounts engaged in high-frequency trading, like @joanbp, made 406 trades in three days and climbed to the top. What does this imply? The Futarchy results actually measured trading ability, not prediction ability.

The deepest issue is the paradox of Futarchy itself. Prediction is decision-making; collective expectations directly influence outcomes. If everyone is optimistic about a project, it can secure funding and is more likely to succeed; the opposite is also true. This creates a self-fulfilling cycle. Participants must weigh two strategies: follow the hype to ensure funding or bet on obscure projects for higher gains. This dual role makes it hard to make purely predictive judgments.

However, the experiment isn’t a complete failure. It did demonstrate the potential of Futarchy as a governance mechanism, especially in activating community participation. The key issues are metric design and participation barriers. If we can find metrics that are harder to manipulate and better reflect a project’s true operational ability, and if we can lower participation hurdles, Futarchy could become a new paradigm for DAO governance.

This experiment showed me a possibility: future DAO governance doesn’t necessarily have to be purely rational and cold discussions. It could also be a deeply gamified consensus-building process. If we can align the energy of speculators (Degens) with ecosystem-building goals, turning betting into judgment and speculation into co-creation, then Futarchy could truly activate new possibilities for Web3 governance.