Can prediction markets win the competition for perpetual contracts?

Article Author: Prathik Desai

Article Compiled by: Block unicorn

Over the past year, we have spent a significant amount of time reporting on perpetual contract (perps) trading platforms. Their rapid rise is hard to ignore. Perpetual futures allow participants to price events shortly after they occur, providing around-the-clock high leverage and ample liquidity. Existing exchanges have never offered such services due to trading time and trading day restrictions. A team of 11 members has built Hyperliquid into the fastest-growing cryptocurrency exchange with this 24/7 trading concept, achieving nearly $1 billion in annual revenue.

In 2025, the average trading volume of perpetual contracts is seven times that of spot trading. This seems to be a reliable way to build a sustainable business. Thus, the inevitable has happened: others have followed suit.

Last week, the two major prediction markets, Polymarket and Kalshi, announced the launch of perpetual futures and cryptocurrency trading within hours of each other. Just a few months ago, Hyperliquid also announced it would launch event contracts. The integration of perpetual contracts and prediction market platforms is a natural progression. Everyone wants to become an all-encompassing exchange, providing one-stop services that integrate attention, capital, and leverage.

Three weeks ago, Saurabh wrote in a report on X that Hyperliquid's entry into the prediction market would help the exchange dominate the financial sector. But does the reverse hold true? Can the moves by Polymarket and Kalshi also yield similar returns?

Today, I will tell you all about it.

Why Perpetual Contracts Are Important for Prediction Markets

Prediction markets face a stickiness problem. They tend to be cyclical, with trading volumes hitting historic highs during significant events to bet on, as seen during the U.S. presidential elections, Super Bowl season, or Federal Open Market Committee meetings.

During the U.S. presidential election in November 2024, Polymarket's monthly active users peaked at 321,500. Three weeks later, this number dropped by 25% to 245,000.

However, the monthly user count fluctuates due to seasonal factors.

In January 2025, Polymarket's user count peaked at 500,000, then fell below 200,000 in September. This reflects Polymarket's user retention rate.

Dune's user group data shows that since 2024, only 8% to 11% of users in the monthly user group are still trading a year after joining. About 75% of users will churn within 90 days. Users may return to participate in events but do not necessarily find the platform sticky.

But this is only part of the problem.

Prediction markets also freeze funds until the questions are resolved. Perpetual contracts (perps), on the other hand, update event prices every second, attracting attention for longer periods and establishing ongoing user interaction. This is also more beneficial for prediction markets, as traders' trading volumes are larger, leading to higher fee income.

In 2025, the notional trading volume of malicious traders exceeded $60 trillion, while the notional trading volume of precious metals traders was $28 billion.

Therefore, this expansion into adjacent fields for prediction markets has become a natural evolution. Platforms that meet some speculative demand often expand their business into other areas. They either develop related features themselves or acquire other platforms that have those features. We have witnessed this many times: Robinhood expanded from the stock market to the options market, the cryptocurrency market, and eventually into prediction markets (PM). Coinbase acquired Deribit for a record $2.9 billion to enter the derivatives trading space. Binance also expanded from providing spot trading to futures trading, ultimately creating its own native blockchain.

We often see this in traditional fields. A company expands its service offerings, hoping to cross-sell new products to the same batch of customers. This serves two purposes: to increase average revenue per user (ARPU) and to diversify reliance on multiple revenue sources, thereby enhancing the company's ability to withstand market cycle fluctuations.

In the early 1970s, the Chicago Board of Trade's (CBOT) commodity futures revenue was continuously declining. So, they utilized a 4,000-square-foot smoking lounge from their parent company CBOT to establish the Chicago Board Options Exchange (now known as Cboe). Since both required shared infrastructure, they could operate synergistically: risk management, clearing, and a network of professionals who understood derivatives pricing.

However, running a perpetual contract trading platform involves a huge gap between having the capability to implement it and actually doing so.

Perpetual Stacking

Operating a perpetual trading platform involves too many components. Let's start with liquidity.

The Hyperliquid platform processes over 200,000 orders per second through a fully on-chain order book. The trading venue settles daily trading volumes exceeding $6-7 billion, using a bilateral market-making model. Insufficient liquidity can lead to extreme volatility, excessive bid-ask spreads, and high slippage, making it easier for whales to manipulate prices.

Next is the risk engine—the core of any derivatives platform. It tracks every trade and checks the margin requirements for each order. In October 2025, the cryptocurrency market evaporated $19 billion, and the Hyperliquid platform handled billions in settlements without interrupting service.

Additionally, there is a funding rate mechanism that ties traders' prices to the spot price of the underlying asset. This mechanism operates continuously by settling small amounts between long and short positions every few hours.

Building the entire tech stack is not the main issue; I believe prediction markets can do that. The bigger problem lies in stress-testing this tech stack.

Hyperliquid built all these systems and stress-tested them in real scenarios, such as the 10/10 cryptocurrency liquidation event and the U.S.-Iran war. After the entire system was ready, it launched event contracts through HIP-4. Kalshi and Polymarket, on the other hand, are trying to go against the grain. They operate successful prediction markets that do not require any of the aforementioned systems. Now, they not only have to compete with the very successful Hyperliquid but also with a system that has not been stress-tested and cannot handle the high-frequency activities of perpetual trading, thus competing for market share.

For prediction markets, many adverse factors make it more difficult to expand into perpetuals than the other way around.

Hedging Synergies

On the Hyperliquid platform, the risk engine monitors all your positions across all trading varieties, spot, and upcoming event contracts. Saurabh explained this in his HIP-4 report.

It looks at all your positions indiscriminately. Ultimately, the leverage you use and the margin you retain as cross-collateral determine when you will be liquidated. The combination of positions in spot, futures, prediction markets, or any other market determines how much margin you need to maintain.

But Saurabh, aren’t other blockchains like Ethereum or Solana also composable? Of course. On a general-purpose chain, each application runs its own risk engine in its respective smart contracts. They cannot atomically view each other's states. Therefore, Kamino cannot know what is happening on Pacifica. Aave also cannot know what is happening on Lighter. All applications are smart contracts on their respective chains. Each application or smart contract has its own independent risk engine, and making them aware of each other—creating a universal risk engine—requires massive collaboration.

This universal risk engine solves a core funding issue by optimizing the same funds across multiple trades conducted by traders in the trading venue.

Suppose a trader on the Hyperliquid platform goes long on ETH with 5x leverage. She is concerned about the Federal Reserve's interest rate decision next week, so she buys an event contract for "Federal Reserve Keeps Rates Unchanged" at $0.65. Because both positions are stored in the same margin account using the same risk engine, if the Federal Reserve unexpectedly cuts rates and the ETH price rises, her long position profits, while the loss on the event contract is only her initial investment. If the Federal Reserve keeps rates unchanged, the event contract will pay out, partially offsetting her long position's loss.

This is why prediction market platforms or hedging trading venues cannot merely be additional features. This hedging possibility is precisely the value of HIP-4 on the Hyperliquid platform. Ordinary traders on the platform view prediction markets as insurance against reversals in their existing hedging positions.

Currently, collateral on the Polymarket and Kalshi platforms will be locked until the event is resolved. Therefore, unless they provide a unified risk engine between their real-money trading and prediction market trading venues, they will lose a key factor that keeps traders on the platform. Currently, neither platform has announced the implementation of a cross-margin system between their prediction market trading and real-money trading venues.

The segmentation of prediction market categories and the average trader profile further raise concerns about whether they can replicate successful performance in real-money trading.

Over 80% of Kalshi's total monthly trading volume comes from sports-related trades. This proportion for Polymarket also exceeds 40% in 2025. So, how do you build a sustainable pricing mechanism around these sports events for a paid trading platform? This would exclude a significant portion of traders from participating in paid trading.

Moreover, Kalshi's ordinary traders are retail investors who have never been exposed to cryptocurrencies, funding their prediction market accounts through ACH transfers from their bank accounts. Therefore, even if I assume that there is theoretically a possibility of cross-margin on the Kalshi platform, I doubt whether these traders possess the expertise needed to double down on the platform and use perpetual contracts as a hedging strategy.

What Methods Might Work for Prediction Markets?

If Kalshi and Polymarket announce cross-margining, I believe there is one scenario in which these bets could work. Their institutional partnerships with major brokers and clearinghouses could facilitate high-value, high-frequency trading activities for event contracts and perpetual futures.

This would enable institutional trading desks to view prediction markets as part of a broader risk management toolkit.

Both Kalshi and Polymarket have partnerships that allow them to reach institutional clients.

Kalshi's collaboration with FIS and Tradeweb data, along with Polymarket's trading with the Intercontinental Exchange (ICE), could help retain institutional clients who value using perpetual contracts to hedge their prediction market positions on the same platform.

This remains an elusive goal that requires many factors in prediction markets to align favorably. They need to build stress-tested infrastructure, form partnerships, and demonstrate to clients that their platform can help optimize capital allocation.

But this is a necessary condition for their survival in fierce competition. With distribution channels already occupied by Hyperliquid, they have no choice but to seek maximum opportunities elsewhere.

That's all for today; see you in the next article.