What if Clutch markets were resolved by the community?

Every prediction market eventually comes down to one question: who decides what actually happened?

If a market asks “Will France win the 2026 World Cup?”, resolving it is easy. But reality isn’t always that clean.

Sources can disagree. Announcements can be ambiguous. A condition can technically be met in an unexpected way. Sometimes, determining whether a market should resolve YES or NO requires actual research and interpretation.

Today, resolving markets is ultimately Clutch’s responsibility. But we’re exploring a different idea: what if Clutch players could resolve markets themselves?

Not through a simple majority vote, but through a system where trusted members of the community put their own ⚡ at stake behind the answer they believe is correct.

The basic idea

When a market closes, instead of Clutch deciding the outcome, it becomes available for resolution by trusted members of the community.

Importantly, you cannot resolve a market you participated in. If you bet YES or NO, you have a financial interest in the result. You can report a suspicious resolution, but you can’t be the person deciding its outcome.

An eligible resolver could then propose:

The resolver isn’t simply voting YES. They’re saying: “I believe YES is the correct resolution, and I’m willing to risk 1,000 of my own ⚡ on it.”

This proposal would then start a challenge period.

No challenge? The market resolves.

Suppose the challenge period lasts two hours. During those two hours, other trusted resolvers who didn’t participate in the market can review the proposed result and its sources.

If nobody contests it: the proposed result becomes final.

The resolver gets their 1,000 ⚡ back, plus a small reward for doing the work of resolving the market.

That’s the optimistic part of the system: if someone provides a well-sourced answer and nobody disagrees, we don’t need another layer of voting or review.

But what if someone thinks it’s wrong?

That’s where things get more interesting. Imagine this market:

The market closes. Alex, a trusted resolver who didn’t bet on the market, proposes:

But another resolver, Sarah, checks the source. The post Alex referenced was actually announcing a new single, not an album. Sarah can contest the resolution:

Now both sides have something to lose. At this point, the market would be escalated to Clutch for a final review.

If Sarah is correct, she gets her stake back and receives a reward from Alex’s lost stake. Alex loses the ⚡ he put behind the incorrect resolution. If Alex was correct, the opposite happens.

In other words, finding an incorrect resolution becomes valuable.

Why not just let everyone vote?

Because prediction markets create conflicts of interest. Imagine 10,000 players voted YES and only 2,000 voted NO.

If we simply asked the community to vote on the resolution, the YES side would have a massive incentive to vote YES again — regardless of what actually happened.

That’s why we’re thinking about separating two roles:

• Players predict the outcome.
• Resolvers determine what happened.

A resolver therefore couldn’t have participated in the market they’re resolving. And rather than counting votes, we’d ask resolvers to put something at stake behind their answer.

Why stake ⚡?

Because a resolution should have consequences. If submitting an incorrect resolution costs nothing, there is very little preventing someone from trolling, guessing or intentionally trying to manipulate a market.

With staking, the incentives change.

• If you’re correct: you get your ⚡ back and earn a resolution reward.
• If you intentionally submit a fake result: someone else can catch it, contest it and take part of your stake.

This creates an interesting dynamic: the system doesn’t only punish bad resolutions — it rewards people for finding them. The safest strategy becomes simply doing good research and submitting the correct answer.

Becoming a trusted resolver

We also wouldn’t open resolution to every newly created account. Resolvers could be selected based on things such as:

• account age and activity;
• history on Clutch;
• previous resolution accuracy;
• fraud and multi-account checks;
• enough ⚡ to cover the required stake.

Over time, users could build a resolver reputation. Consistently providing good resolutions would increase that reputation. Repeatedly getting successfully contested could decrease it or eventually remove access to community resolution entirely.

Bigger markets could require bigger stakes

Not every market carries the same risk. Resolving a small market incorrectly might affect a few thousand ⚡. Resolving one of the biggest markets on Clutch incorrectly could affect millions.

So the amount required to resolve a market could potentially depend on its size:

• A normal market might require 500 ⚡.
• A larger market: 2,000 ⚡.
• An exceptionally large or sensitive market: 10,000 ⚡ + a longer challenge period.

For the largest markets, we could still require a final Clutch verification. The goal isn’t decentralization at all costs. It’s creating a system where trying to manipulate a resolution is simply not worth it.

What happens if nobody wants to resolve a market?

Some markets take five seconds to verify. Others require digging through three websites, translating an announcement and figuring out whether a very specific condition was actually satisfied. Those shouldn’t pay the same reward.

One possibility we’re considering is a resolution bounty that increases over time. A market could start at:

• Resolve this market: +50 ⚡
• If nobody touches it after 30 minutes: +100 ⚡
• After an hour: +200 ⚡

Eventually, doing the research becomes worth someone’s time. If nobody resolves it after a maximum period, it simply falls back to the Clutch team.

What the full system could look like

1. A market closes.
2. Trusted players who did not bet on it can resolve it.
3. Someone stakes ⚡, chooses YES or NO and provides their evidence.
4. A challenge period begins.
5. Nobody contests → the result becomes final.
6. Someone contests → both sides have ⚡ at stake → Clutch arbitrates.

The correct resolver is rewarded. The incorrect resolver is penalized. And Clutch only needs to intervene when the community genuinely disagrees.

Why we’re interested in this

Clutch is fundamentally a game about collectively predicting what happens in the world. We think there’s something pretty powerful about letting the same community help determine what actually happened.

It could make resolutions faster, more transparent and more robust while giving our most active players a completely new role inside Clutch.

Instead of a black box saying “Market resolved YES,” you could see:

And because every resolver has something to lose, the system isn’t based purely on trust. It’s based on incentives, transparency and accountability.

This is still a concept we’re exploring, and the exact numbers, eligibility requirements, rewards and challenge periods aren’t decided yet. So we’re especially interested in what you think.

New to how markets work on Clutch? See how Clutch works.