Table of Contents
Researcher
Related Projects
Key Takeaways
HIP-3 removes the technical barrier to launching new perpetual futures markets and enables a demand driven model of market creation, turning DEXs from CEX competitors in a PvP dynamic into PvE style expansion paths that extend into non crypto assets and real world data.
The market is transitioning from narrative driven growth to cash flow driven, sustainability oriented valuation, and only the small set of projects with real revenue that flows back to the token, such as Hyperliquid and Pump.fun, are likely to control the next cycle.
Prediction markets convert what used to be private or illicit betting activity into public, on chain data and time series data of collective expectations, creating real time probability signals and alt data that financial institutions, data vendors, and AI models can use as an economic mechanism for aggregating information and estimating probabilities.
Regulation has produced a split regime in which prediction markets move toward institutionalization in the West and suppression in Asia, creating a major short term constraint but opening the path for prediction markets to evolve into infrastructure for turning collective beliefs into data and markets that produce information.
1. How HIP-3 Opens a New PvE-Style Growth Regime
The exchange business model is shifting.
Centralized exchanges (CEXs) still maintain structural advantages based on institutional trust: fiat on- and off-ramps, custody, and regulatory access. This makes them the natural entry point for institutional capital and provides stability in liquidity and operational reliability. But the same regulatory obligations, internal controls, and custody infrastructure create high fixed costs. As a result, CEXs are slower to experiment and slower in decision-making, which limits their pace of innovation.
Decentralized exchanges (DEXs), by contrast, have grown through incentive structures. They coordinate rewards among LPs, traders, and builders natively on-chain. But launching a new exchange or market previously required a team to build matching engines, margin and liquidation systems, and oracles from scratch. This created a meaningful technical barrier to entry.
HIP-3 removes this barrier.
Hyperliquid now allows anyone who stakes 500,000 HYPE to deploy their own perpetual futures market using the same CLOB engine, margin logic, and liquidation system that the main exchange uses. The technical burden of building an exchange disappears. Market creation becomes a standardized on-chain deployment process that requires capital and a credible oracle, not a full engineering team. The barrier shifts from technical capability to capital and oracle design.
This change does more than improve efficiency; it shifts where innovation happens.
Builders can now experiment with liquidity structures, fee design, oracle definitions, and leverage limits without needing to recreate the backend. The challenge becomes identifying a demand surface—some number people want to speculate on—and anchoring it to a trustworthy oracle. In effect, markets can now be created from three components: Market + Oracle + Demand.
This expands the scope of what can be listed.
As Ventuals founder Alvin Hsia describes, the “fat head” will consist of asset classes already well-covered in traditional finance—index products, FX, commodities. The “chunky middle” includes private equity, real-world datasets, and commodity indices. And the “long tail” reaches into niche signals like localized real-estate prices, product premiums, or cultural trend indices. Traditional finance cannot easily commoditize these data points, but an on-chain settlement system can. HIP-3 effectively enables a demand-driven model of market creation.
Source: X (@alvinhsia)
This transforms DEXs from CEX competitors into something structurally different.
Rather than fighting for a fixed pool of crypto-native liquidity (a PvP dynamic), HIP-3 allows DEXs to expand into non-crypto assets and real-world data. This brings in new flows, new users, and new forms of demand—a PvE dynamic where the market size increases instead of being redistributed. It also deepens protocol-level revenue.
A clear example is Hyperliquid’s XYZ100 market, which passed $1.3B in cumulative volume within three weeks of launch, showing how quickly new asset classes can scale when the infrastructure is standardized.
In short, CEXs continue to provide stability and regulatory access, but HIP-3-based perpetual DEXs gain an advantage in speed, experimentation, and asset expansion. These are not substitutes; they are distinct growth paths.
The competitive edge in exchanges will shift from backend engineering to market design and user experience, and leadership will be defined by which protocol can turn that into sustainable value.
2. From Narrative-Driven Valuations to Cash-Flow-Driven Valuations
The 2025 market is fundamentally different from previous cycles.
The broad liquidity environment that lifted every asset together is gone. Capital now flows selectively. Prices reflect actual performance more than narrative, and projects that fail to generate revenue are naturally pushed out. Most altcoins still have not recovered their 2021 highs, while protocols with clear revenue continue to show relative strength even during market pullbacks.
The arrival of institutional capital is solidifying this shift.
Tradfi frameworks are being applied directly to crypto. Revenue, net income, fee generation, user activity, and profit distribution are becoming the primary metrics for project evaluation. The market is moving away from valuation by story or expected growth. Only projects with real revenue that flows back to the token are earning higher market valuations.
In this context, Uniswap’s recent proposal to activate the fee switch is symbolic. A flagship DeFi protocol is explicitly choosing to link cash flows with token value, signaling that fundamentals (not narratives) are now at the center of market pricing.
A clear group of outperformers has emerged.
Hyperliquid (HYPE) and Pump.fun (PUMP) are leading examples.
Hyperliquid is the largest perp DEX by volume, open interest, and trader count, and is positioning HyperEVM as a backend infrastructure layer (“House of All Finance”). As of November 2025, cumulative trading volume stands at $3.1 trillion, total users at 810,000, and open interest at $9 billion. Notably, Hyperliquid allocates 99% of perpetual fee revenue to HYPE buybacks, directly linking protocol cash flows to token value. Total buybacks have reached 34.4M HYPE (~$1.3B), roughly 10% of circulating supply. HIP-3 expansion, USDH stablecoin adoption, and protocol-level revenue growth remain key drivers going forward.
Pump.fun is the leading memecoin trading platform, generating ~$1.1B in cumulative fees. Its buyback program has purchased roughly 830,000 SOL (~$165M), equal to ~10.3% of its circulating supply. Recent acquisitions (the Terminal trading platform), media integrations (streaming features), and a mobile rollout point to further user growth. Together, these protocols represent a new benchmark for revenue-driven token valuation models.
Other projects also show strong revenue momentum.
Aave (AAVE) and Jupiter (JUP) continue to post stable and growing cash flows, while Coinbase (COIN), despite being a listed equity, is benefiting from the increasingly visible path toward a Base chain token. Coinbase has also broadened its revenue mix: subscription and services revenue reached $746.7M in Q3 2025 (+13.9% QoQ), demonstrating diversification beyond trading fees.
Aave’s annual revenue grew from $29.75M (2023) to $99.39M (2025), with net income rising from $12.83M to $49.22M over the same period. Jupiter’s revenue expanded far more dramatically—from $1.42M (2023) to $246.06M (2025)—with tokenholder earnings reaching $127.36M.
This shift is spreading beyond individual dApps into L1 and L2 ecosystems. Technical merit or investor branding alone is no longer enough. Chains with real users, real transactions, and protocol-level revenue are receiving stronger market recognition. The key evaluation metric is becoming the sustainability of economic activity.
In summary, the market is transitioning from narrative-driven growth to cash-flow-driven, sustainability-oriented valuation. This is not a temporary cycle but a structural shift in how investors understand and price crypto assets. Only the small set of projects that can demonstrate real revenue and clear value accrual to the token will control the next cycle, and the 2026 market is likely to reorganize around these performance-backed players.
3. Quantifying Market Expectations Through Prediction Markets
Prediction markets are an experiment in converting what used to be private or illicit betting activity into public, on-chain data. At their core, they quantify the probability that people assign to future events by putting real money behind those beliefs. This makes them not just speculative venues but economic mechanisms for aggregating information and estimating probabilities.
Prediction markets have grown rapidly since 2024.
As of October 2025, weekly notional volume is roughly $2.5B with more than 8M weekly trades. Polymarket accounts for 70–75% of all activity, while Kalshi captures around 20%. Kalshi’s share increased sharply after securing CFTC approval (DCM status) and expanding into sports and political markets. Polymarket continues to grow primarily outside the U.S., scaling across political, economic, and technology categories.
Prediction market data matters because it provides something other indicators do not.
Polls, social-media sentiment, and institutional research are slow, costly, and often lag real shifts in expectations. Prediction markets price these expectations in real time. For example, Polymarket reflected Donald Trump’s improving 2024 election odds significantly earlier than traditional polls.
In effect, prediction markets create time-series data of collective expectations. These curves can be used as real-time probability signals for political, economic, sports, and technology events. Financial institutions, data vendors, and AI models are increasingly viewing these markets as a potential alt-data source for quantifying expectations.
Source: Grayscale Research
From an institutional perspective, prediction markets represent not the “datafication of gambling” but the financialization of uncertainty. Because prediction market prices reflect consensus probabilities, macro traders and hedging desks can use them to manage event risk. Kalshi already offers markets tied to inflation, jobs data, and rate decisions. These products naturally attract hedging interest from participants who want exposure or protection tied to specific macro outcomes.
Polymarket’s data has also been examined in academic research, where it has shown stronger lead-lag behavior than traditional polling. Some data-science projects have proposed using these probability curves as features in quantitative models.
As prediction markets mature, they create a new value chain linking market → oracle → data → applications:
Market: transactions generate probability signals
Oracle: credible mechanisms for resolving outcomes
Data: standardized and cleaned time-series datasets
Applications: consumption by finance, media, research groups, and AI systems
The main barrier today is regulation, and regulatory stance varies by region.
According to Four Pillars analyst Calvin’s report, Asian jurisdictions remain mostly prohibitive: Korea, Singapore, and Thailand classify prediction markets as illegal gambling and block access, with penalties for users as well. Japan and Hong Kong lack clear legal frameworks, making operation effectively impossible.
Western jurisdictions are more flexible.
The U.S. regulates prediction markets as “event contracts” under the CFTC. Kalshi operates legally under a DCM license, and Polymarket aims to reenter the U.S. market through its acquisition of QCX in 2025. The U.K. and parts of the EU are testing regulatory sandboxes that allow limited experimentation with prediction-market-like products. These differences have produced a split regime: institutionalization in the West, suppression in Asia.
This regulatory gap is a major short-term constraint.
But in the long term, prediction markets can evolve into infrastructure for turning collective beliefs into data. If jurisdictions begin to recognize their public-interest value, these markets may move beyond the gambling category and function as networks that capture, evaluate, and price the probability of events in real time.
This creates a system where market prices feed into finance, media, policymaking, and AI. In that sense, prediction markets signal a shift from “markets that interpret information” to markets that produce information, reinforcing a world where price becomes the primary representation of collective expectations.
