The Truth About the Prediction Market: Why Most New Projects Won't Survive Past 2026

Prediction markets have recently become a hot entrepreneurial trend in the crypto space, but the question is—does this track really live up to the friendly image we imagine?

The Market-Making Trap in Order Book Models

Many people see the innovation in the Yes+No=1 formula and think that prediction markets have solved liquidity issues. But in reality, market-making in prediction markets is far more challenging than in AMM markets.

There are fundamental differences between the two market-making models. In traditional AMMs (like Uniswap), you only need to deposit two assets into the liquidity pool, and the formula x*y=k automatically handles price changes. You just collect trading fees periodically. However, prediction markets use an order book model, which means you must place buy and sell orders at specific price points.

This sounds simple, but in practice, there are four possible outcomes:

First: All orders remain unfilled, earning only platform subsidies
Second: Both sides are filled, locking in spread profits
Third: One side is filled, resulting in a unidirectional position with uncertain risk
Fourth: After a partial fill, the market price swings sharply, leading to floating losses

The key issue is that returns are often quite fixed—the spread is limited, and subsidies have an upper bound—but the risks are unlimited. In prediction markets, a wrong position can wipe out your entire stake.

Price Volatility More Crazy Than Crypto Markets

If market-making alone were complicated but manageable, that would be acceptable. But prediction markets are even more “unethical” than crypto trading markets.

First, the underlying assets are event contracts, each with a settlement date. The Yes+No=1 design ensures that only one option will ultimately be worth 1 dollar, while others are worth zero. This means the market’s funds will inevitably end in a unidirectional trend—forcing market makers to face unavoidable inventory risk.

Second, conventional trading markets have continuous volatility, giving market makers time to adjust. But prediction market prices change in jumps—an unexpected news event can cause the price to jump from 0.5 directly to 0.9 or 0.1, leaving market makers unable to react in time.

Third, prediction markets are filled with players who are close to the source of information or already possess it. They are not here to predict; they are here to harvest. Ordinary market makers are naturally at an informational disadvantage and can only serve as liquidity providers for these players to cash out.

The Harsh Truth About Liquidity Subsidies

Because market-making is so difficult, leading platforms have to pour money into maintaining liquidity.

One exchange invested about $10 million in subsidies, paying over $50,000 daily at its peak. On average, this means that for every $100 traded, about $0.025 in subsidies are needed. Another top platform has already invested at least $9 million and has signed market-making agreements with top Wall Street market makers to stabilize liquidity.

This not only reflects the difficulty of market-making but also exposes the true nature of the track—without continuous capital subsidies, liquidity cannot be built. And only a few platforms possess such financial strength.

The Survival Dilemma of New Projects

Currently, many prediction market projects are emerging, but the reality is: the competitive moat of top platforms is much deeper than expected.

In terms of funding, top platforms have secured hundreds of millions of dollars, ensuring ample resources. In terms of compliance, they have obtained regulatory licenses first, with stable partnerships at a higher level. In terms of user base, they have gained first-mover advantages and brand recognition.

What can new projects rely on to compete head-to-head? What can they spend on? While some projects backed by big capital may survive, they are the minority.

Industry insiders have predicted that the prediction market track will develop rapidly, but 90% of products will eventually be ignored and gradually disappear by the end of 2026. This may sound harsh, but it’s not just alarmist talk.

Conclusion: Focus on the Leaders, Don’t Spread Your Bets

Prediction markets do represent a promising direction, but their track logic is not as inclusive as AMMs, where multiple participants can thrive together. Instead, the high-cost barrier to liquidity ensures that the industry will tend toward centralization—top projects with sufficient capital, regulatory compliance, and established brands will continue to widen the gap.

Rather than diversifying your bets across various new projects in hopes of finding a “dark horse,” it’s better to focus directly on the established leading platforms. The future of this track may well be a game of a few oligopolists.