Every few years, a new prediction market platform arrives with a better interface, a cleaner feed, a faster resolution engine. Users migrate. The old platform loses volume. The new one gains it. And then, a few years later, the cycle repeats.
This is not a product problem. It is a structural one.
Every platform that has existed in this category shares the same shape: a company in the middle, deciding what gets listed, holding the collateral, and keeping the economics. The interface changes. The underlying arrangement does not. And because the arrangement never changes, the failure modes never change either.
The arrangement always breaks the same way
Curated venues face a decision that compounds over time: grow the market selection or protect the integrity of each market. In the early days, both are possible. As the platform scales, they are not.
The operator starts making calls. Some events are too politically sensitive. Some outcomes are too difficult to verify. Some markets attract the wrong kind of attention. Each individual call is defensible. The cumulative effect is a narrowing of what the platform will host - and a growing gap between what the crowd wants to predict and what the platform allows.
Users learn to work around it. They migrate to wherever the market they want exists. Or they stop altogether.
This is not a moderation failure. It is what happens when a single entity holds the power to decide what is real enough to trade on. The constraint is architectural, not editorial.
What changes when the constraint is removed
Halley is the prediction DEX - a non-custodial exchange on BNB Chain where anyone can trade what happens next, open their own market in about a minute, and keep 80% of its fees. Trading is free, positions are tokens in your own wallet, and the core contracts are immutable.
On Halley, listing is permissionless - any wallet, no application, no gatekeeper. The creator opens the market, sets the resolution criteria, and brings the crowd. Markets settle on what the public record confirms.
This is not a philosophical position. It is a technical one. The contracts are deployed and audited.
The result is a platform where the question of what gets listed is not a platform decision. It is a creator decision. And because it is a creator decision made in public, it is accountable in a way that closed-door editorial calls are not.
The cold-start question is the honest one
An open network where anyone can open a market is also a network where a market can open to no one. Books are thin where no crowd has arrived yet - and thin books are a real disadvantage against venues that have spent years building depth.
The answer is not to pretend otherwise. The answer is structural: on Halley, every market is opened by the person who already has the crowd. The creator does not build the market and then go hunting for an audience. The audience and the market arrive together, in the same motion. The argument running in the comments becomes the order book.
This does not repeal cold-start. It changes who is responsible for solving it - and gives that person a direct economic reason to solve it well.
The economics of not switching
The reason users switch platforms is almost always the same: the market they want is somewhere else. On a curated venue, that will keep happening, because curation will keep narrowing.
On an open network, the market you want exists if someone with your crowd has opened it. And because that someone keeps 80% of the resolution fee, they have every reason to.
Where this ends
The prediction market that a user never needs to leave is not one with better UX or faster resolution. It is one where the market they want can always exist - because the infrastructure does not get to decide otherwise.
The rules can't move. The markets can't be closed. The economics go to the people who built the audience.
That is what an open network means in practice.
Make a market.
Predict the return.