OKX Ventures Research Report: Understanding the Landscape of "Prediction Markets" in One Article

Core Viewpoint

2025-11-26 23:36:53

Collection

Compliance has become a growth anchor, regulatory qualitative assessments will determine the market ceiling, and the industry is gradually differentiating...

Author: OKX Ventures

In 2025, as regulations become clearer and institutional funds accelerate their inflow, the crypto prediction market is transitioning from a marginal experiment to an important tool for information pricing—reflecting the on-chain collective intelligence's real-time betting and risk management of future uncertainties. From macroeconomic indicators to technological innovations, the probability judgments that once relied on single channels may now be rapidly and accurately repriced by on-chain funds. This change is affecting the way we understand information and market signals.

In simple terms, the prediction market in the crypto industry is a platform where users can bet on future events using cryptocurrencies, such as predicting whether a certain coin's price will rise or fall, or who will win a match. If they guess correctly, they can make money; if they guess wrong, they will lose.

OKX Ventures will continue to focus on this sector, outlining the overall landscape of "prediction markets." Our research report is for learning and communication purposes only and does not constitute investment advice.

I. Origins and Development of Prediction Markets

As an information mechanism that aggregates "collective intelligence," prediction markets are entering a new stage driven by Web3 technology. After 2000, this industry has continuously evolved through academic innovation, regulatory games, and shifts in technological paradigms, presenting a new pattern of rapid growth and structural transformation by 2025.

The modern model of prediction markets originated from the Iowa Electronic Markets (IEM) in 1988, which first proposed the idea of "price as probability," aggregating participants' divergent judgments on event outcomes through trading financial contracts. In the 1988 U.S. presidential election, IEM achieved accurate predictions under a small-scale operation. Subsequent studies showed that IEM's election prediction accuracy was higher than 74% compared to public opinion polls between 1988 and 2004, and it could demonstrate predictive advantages 100 days before the election. In 1993, the CFTC's exemption for IEM established a policy foundation for event-based contract markets, allowing academic experiments to venture into commercialization.

Starting in the 2000s, a wave of commercialization in prediction markets emerged, but it faced multiple challenges from gambling/financial regulations. Representative platforms like Betfair and Intrade grew rapidly but were also limited or even shut down due to their gambling nature and compliance difficulties. Although PredictIt initially received a "no-action letter," it was later withdrawn and embroiled in litigation, highlighting the extreme difficulty of institutionalizing external compliance markets. In contrast, internal prediction markets like Inkling avoided regulation and became important tools for corporate decision-making.

In the 2020s, the industry has undergone two major structural transformations. First, Kalshi received CFTC approval, becoming a compliant "event contract" exchange, marking formal regulatory recognition, but the 2023 election market case still highlighted the sensitivity of topics and regulatory boundaries. Second, the intervention of Web3 technology allowed prediction markets to achieve trustless clearing and settlement through blockchain smart contracts, significantly enhancing censorship resistance and lowering compliance operational thresholds. In October 2025, ICE, the parent company of the New York Stock Exchange, plans to invest approximately $2 billion in the decentralized platform Polymarket, which is seen by the industry as a recognition of the Web3 model by top financial infrastructures, signaling a paradigm shift for the entire industry.

Prediction markets (information markets, decision markets, event derivatives) allow participants to bet with funds, with market prices directly reflecting the probabilities of events occurring. Their theoretical foundation includes the efficient market hypothesis and the principle of collective intelligence: the former posits that market prices can be approximated as probabilities, while the latter emphasizes that collective market decisions outperform individual ones under diverse, independent, and decentralized participation mechanisms.

Taking the probability of interest rate cuts as an example, we compare professional data with prediction market data.

Tools like CME FedWatch derive probabilities from derivative prices, representing institutional expectations but are influenced by tool design and funding structures. In contrast, open prediction markets like Polymarket and Kalshi determine probabilities directly through bets from all participants, theoretically offering greater transparency and democracy; the higher the platform's activity, the more comprehensive the reflected information.

Overall, since the Iowa Electronic Markets (IEM) validated the price as probability in 1988, prediction markets have undergone a long regulatory tug-of-war until 2020, when Kalshi's compliance and the intervention of Web3 technology marked the beginning of a paradigm shift characterized by institutionalization and decentralization. Driven by the resonance between Web3 and traditional financial backers, prediction markets are currently in a window of industry explosion and mechanism evolution. The parallel of decentralization and institutionalization will reshape the global financial information aggregation and price discovery system.

II. Market Growth and Platform Strategy Analysis

(1) Macroeconomic Market Overview: Dual Oligopoly and Capital Inflow

Catalyzed by the 2024 U.S. election and the influx of institutional capital, the prediction market is experiencing structural explosions in 2025. The market has shifted from being driven by single events to ongoing financial trading activities, with both capital depth and user base reaching historical highs.

Transaction Volume Peaks: The market has experienced two significant peaks. The first was driven by the U.S. election from October to November 2024, with weekly transaction volumes approaching $2 billion. The second explosion began in July 2025 and reached a historical high in October 2025, with weekly transaction volumes exceeding $2.5 billion, surpassing the peak during the election period.

User Growth: Market activity and transaction volume are highly positively correlated. In October 2025, the total number of weekly active users across the market surpassed 225,000, indicating a continuous and genuine influx of new users.

Capital Depth (Open Interest OI): OI (Open Interest) represents the real capital locked in the market. During the 2024 election, the total market OI peaked at nearly $800 million, then declined due to settlements. Entering the second half of 2025, the total market OI has steadily rebounded and stabilized in the range of $500 million to $600 million. This marks a departure from mere short-term speculation, establishing a capital base composed of institutions and long-term participants.

Competitive Landscape: The market exhibits a dual oligopoly between Polymarket and Kalshi. In 2024, Polymarket held about 90% of the absolute dominant share. However, by October 2025, the compliant platform Kalshi's market share had climbed to nearly 60%, surpassing Polymarket in total volume. Despite the concentration at the top, second-tier platforms like Opinion, Limitless, and Myriad still captured the remaining market and gained stable liquidity during specific time windows (such as token issuance periods).

(2) Head Platform Data Breakdown

1. Polymarket: From Election Windfall to Multi-category Retention.

As the leader in decentralized prediction markets, Polymarket's data exhibits strong "event explosiveness" and has maintained high user retention through category expansion after the election.

In terms of user data, active wallets reached a historical peak of nearly 80,000 by the end of 2024. Although there was a decline after the election, by October 2025, it remained stable above 60,000, significantly higher than early 2024 levels. Monthly active traders (MAU) peaked at 450,000 in January 2025, and even after the election heat subsided, it maintained over 260,000 active users, demonstrating strong long-tail stickiness. Daily active users (DAU) reached a single-day peak of 58,000 on October 19, 2025, growing from an early 10,000 to 58,000, an increase of nearly six times.

In terms of transaction volume, the historical cumulative transaction amount has exceeded $18.1 billion, with the monthly peak occurring during the November 2024 election period, reaching $2.63 billion, approximately 1,000 times higher than early data from December 2020. After the election, monthly transaction volumes fell to about $1.9 billion, a decrease of about 30%-40%, but still far above 2023 levels. In terms of growth rates, the year-on-year growth in July 2024 reached 5,270%, and in October 2025, the year-on-year growth soared to 26,000%, mainly influenced by the low base from the same period last year; on a month-on-month basis, transaction volumes also achieved explosive growth of about 750% from October to November 2024.

In terms of market structure and categories, during the peak of the 2024 election, "political/economic" transactions accounted for over 60%, with weekly transaction volumes once exceeding $1 billion; by 2025, the trading focus gradually shifted towards "sports" and "crypto assets," with contracts related to the Super Bowl reaching about $1.1 billion in trading volume, and Bitcoin prediction markets, such as "Bitcoin price in 2025," exceeding $15.5 million in trading volume. Meanwhile, market supply is also expanding, with the number of newly established prediction markets in April 2025 exceeding 7,000, setting a historical high.

2. Kalshi: Exponential Growth Driven by Compliance Channels

Kalshi demonstrated the strongest growth momentum in 2025, leveraging its compliance advantage to tap into Web2 channels, achieving multiple-fold increases in all core metrics.

In terms of trading and scale, from the end of 2024 to early 2025, Kalshi's trading volume and transaction count exhibited explosive growth, with cumulative transaction amounts exceeding $10 billion and cumulative transaction counts surpassing 40 million, with an average transaction price of about $250–300, indicating clear retail user characteristics. Weekly nominal transaction amounts surged from $150 million to $200 million in Q4 2024 to over $850 million in Q2 and Q3 2025.

In terms of market share and ranking, by October 2025, Kalshi's weekly transaction volume accounted for 55%-60% of the entire market, officially replacing Polymarket as the most liquid prediction market platform. In terms of total contribution, from September to October 2025, the total weekly transaction amount across the industry exceeded $1.5 billion, with Kalshi contributing $800 million to $900 million in a single week, more than five times the growth compared to 2024.

In terms of positions and user structure, open interest surged from less than $50 million at the end of 2024 to over $200 million in Q3 2025; monthly active users (MAU) expanded from 80,000–100,000 at the end of 2024 to over 400,000 in mid to late 2025. The market structure is highly concentrated, with sports accounting for about 45% and politics about 30%, contributing over 75% of the open interest, while economic-related transactions account for about 10%. On the supply side, the number of effective markets expanded from about 300 to over 1,200 by October 2025.

(3) Emerging/Vertical Platform Data Performance

In addition to the two giants, platforms like Opinion Lab, Myriad, and Limitless have also produced noteworthy data under specific incentives or vertical scenarios.

Opinion Lab: Explosive Growth After Mainnet Launch

The daily fees on the Opinion platform saw an explosive surge on October 25, exceeding $200,000 in a single day, rapidly pushing cumulative transaction fees from nearly zero to about $320,000, and by November, cumulative transaction fees had surpassed $600,000. On the launch day (October 25, the day the mainnet incentive was activated), nominal transaction volume, transaction counts, and user numbers all reached daily peaks, with Opinion accounting for about 15%-20% of the nominal trading share in the entire market that day. In terms of funding, on October 30, the platform's total locked value (TVL) peaked at $50 million, indicating that large amounts of capital briefly stayed after the incentive activities.

Myriad: Media Traffic Conversion and Retention

In terms of user conversion, its registered users exceeded 513,000, but active trading users were only 30,000 (USDC traders), resulting in a registration-to-trading user ratio of about 17:1, reflecting the conversion funnel from content readers to financial traders, with approximately 30,000 active wallets. In terms of trading scale, cumulative trading volume reached $12 million (USDC), with a peak daily transaction amount close to $2 million during the explosive period from September to October 2025, and weekly peaks exceeding $6 million. In terms of capital retention, total locked value (TVL) approached $800,000 in mid-October, showing an overall increase of about 300% since August, while open interest (OI) peaked at $500,000 in early October, with a month-on-month growth of about 2.5 times. In terms of revenue capability, cumulative fee income was about $400,000, with a peak daily fee of about $6,000.

Limitless: Data Fluctuations Under High-frequency Incentives

In terms of trading volume, a surge occurred from August to September 2025, with trading volume increasing by about 25 times, and by mid-October, trading volume exceeded $100 million in just half a month, with cumulative totals surpassing $500 million. The first season of incentives attracted about 34,000 active traders, completing 750,000 transactions. Despite the cumulative trading volume reaching $500 million, the peak total locked value (TVL) was just slightly above $1 million, indicating that users primarily engaged in ultra-short-term high-frequency trading. After the airdrop ended, the 24-hour trading volume quickly dropped by 34.7% to $7.56 million, with data gradually returning to normal.

(4) Core Data Horizontal Comparison Table

From different indicators:

From a GTM perspective:

The growth quality brought by GTM strategies presents a clear hierarchy: structural growth quality from channel integration (like Kalshi) is the highest; followed by cyclical explosions driven by events (like Polymarket); then high-friction growth from content conversion (like Myriad); while volume growth driven by airdrop expectations declines rapidly;

Compliance is the "underlying anchor" for growth: The evolution of Polymarket (event-driven → compliance pain → institutional compliance) proves this point. Although it exploded due to the 2024 election (450,000 MAU), it missed the U.S. market for three years due to early compliance issues. The intention to invest by ICE in 2025 and the acquisition of a licensed exchange mark that top players must ultimately return to a "compliance" path.

III. Industry Structure Analysis

(1) Track Mapping

1. Infrastructure is gradually becoming modular, evolving from "integrated platforms" (like early Augur) to "composable Lego modules":

Infrastructure protocols like Azuro, UMA, and Gnosis Omen have significantly lowered the technical barriers for new projects, supporting multi-chain deployment and plug-in front ends, enriching the general track and providing entrepreneurs with plug-and-play product foundations, fostering more long-tail innovative scenarios, such as AI-driven and community-driven applications. Gnosis (Omen) / CTF provides an industry-standard asset issuance framework, while Azuro Protocol offers plug-and-play betting middleware, enabling new projects to focus on front-end experiences without needing to build liquidity pools and odds engines from scratch.

This modularization has greatly lowered the startup threshold for new applications, promoting the prosperity of numerous niche tracks, including sports and gamification, bringing more innovation and diversified development possibilities to the entire prediction market ecosystem.

2. Vertical Track Opportunities Abound: Sports and Creator Economy Become Growth Points

In addition to general-purpose and leading platforms, projects in tracks like sports, music, pre-IPO, knowledge verification, social gaming, AI prediction, and Telegram Bots are rapidly emerging. Viral growth mechanisms such as social media/UGC, creator revenue sharing, and community interaction have become common features of new projects, enhancing user participation and diffusion speed.

The window for general-purpose platforms is closing, and new project opportunities are gradually concentrating towards verticalization. Data maps show that sports betting and creator/social economy are currently the two fastest-growing vertical tracks. Sports betting, as an independent track, has already spawned dedicated public chains (like SX Network), middleware (Azuro), proprietary protocols (Overtime), and even phenomenal applications like Football.fun, which reached a TVL of $10 million in just two weeks.

In the creator and social economy track, project models are no longer about "creating markets themselves," but rather driving growth through "empowering KOLs." Platforms like Melee offer a 20% revenue share to creators, while Index.fun provides 30% creator earnings, transforming prediction markets from mere information tools into tools for creators and opinion leaders to monetize their influence, further expanding the diversity and application scenarios of the ecosystem.

3. AI is a Standard for New Projects: Automated Creation and Settlement

AI demonstrates immense potential in market generation, event analysis, content production, settlement, sub-gaming, and risk control scenarios. Combined with AI agents or Copilots, it can assist in information archiving, automatic setting of odds, and even predicting outcomes, thereby enhancing user experience and operational efficiency. Artificial intelligence is transitioning from an auxiliary tool to a core product positioning for some new platforms, addressing the fundamental pain points of cost and efficiency in traditional prediction markets. For example, OpinionLabs and BuzzingApp both position "AI agents" as core, achieving zero-cost creation, infinite supply, and automated settlement, fundamentally disrupting Polymarket's reliance on UMA and Kalshi's dependence on manual compliance.

4. Interactive Layer Reconstruction: "Bots" and "Aggregators" Become Key Entry Points

Front-end tools and Telegram bots are significantly reducing user experience friction for underlying protocols, thus giving rise to new interactive layer opportunities. For instance, Bots like Flipr and Noise simplify complex prediction trading operations from cumbersome web interfaces to one-click orders within tweets or group chats, becoming key GTM strategies to reach socially-driven markets; simultaneously, aggregators like Flipr and XO Market address the liquidity fragmentation of leading platforms (Polymarket, Kalshi) and add features like leverage and stop-loss that native platforms lack, precisely meeting the needs of professional traders.

5. Prediction Markets as Part of DeFi Lego: Composable Infrastructure

Prediction markets are gradually evolving from isolated platforms to true "DeFi Lego." On one hand, some projects (like Index.fun) are starting to package different prediction outcomes (based on Gnosis CTF standards) into tradable "creator indices"; on the other hand, this data (like Kalshi integrating Pyth) is provided to other DeFi protocols (like Gondor) through oracles, becoming new source data for on-chain derivatives, insurance, or lending protocols.

6. Prediction Markets Can Be Viewed as Financial Derivatives: Convergence from "Event Prediction" to "High-frequency Trading"

Product design is rapidly converging from "predicting events" to "financial derivatives." For example, Limitless's 30-minute ultra-high-frequency contracts are used as volatility trading tools, while Flipr's 5x leverage transforms it into futures, and Touchmarket's price range predictions evolve into structured options, indicating that the entire track is quickly developing from its initial purpose of information aggregation to a new branch of high-frequency DeFi derivatives trading.

7. Middleware Opportunities Are Worth Noting, Such as Data and Cross-chain Solutions

Middleware and data tools are forming a complete prediction market infrastructure, significantly enhancing trading efficiency and user experience. For example, Polysights provides services similar to a "Bloomberg terminal for prediction markets," aggregating decentralized data from platforms like Polymarket, Kalshi, and Limitless, tracking smart money movements and generating advanced arbitrage signals; in the betting and sports track, Azuro has demonstrated the opportunity of "backend as a service," packaging liquidity pools, odds engines, and sports data oracles (Chainlink), allowing new projects to focus on user experience and GTM promotion.

At the same time, cross-chain liquidity and order aggregation have become key, with markets distributed across Base (Limitless), Polygon (Polymarket), and Solana (Melee). Flipr aggregates order flows, routes odds, and solves cross-chain settlements, playing the role of "the 1inch of prediction markets"; additionally, the opportunities for GTM and interactive layer middleware lie in embedded tools, such as Flipr and Noise, injecting trading functions (like one-click orders, leverage) into Telegram Bots, X platform tweets, or content wallets, capturing traffic and trading behavior at the moment users express trading intent.

(2) Comparison of Leading Projects

(3) Regional Analysis of Track Investments

1. North American Market: Compliance and Data-Driven

Compliance Finance Dominates (Kalshi): Kalshi leads the U.S. compliant market with its CFTC license, monopolizing regulated macro data trading such as economic indicators and Federal Reserve interest rate decisions, with its share once surpassing Polymarket to reach 60%.
Political and Event Diversion (Dual Oligopoly): High-profile political events are contested between Polymarket (global funds) and Kalshi (U.S. compliant funds), forming clear regional and funding attribute distinctions.
Avoiding Platform Competition (Extremely High GTM): Barriers for leading platforms have formed, and blindly establishing new compliant exchanges is too costly and unlikely to succeed; the idea of directly challenging Kalshi/Polymarket should be abandoned.
Mining "Auxiliary Layers" (Tool Opportunities): The real opportunity lies in serving giants, such as developing front-end analytical tools for Robinhood/Kalshi channels (Polysights model) or becoming institutional compliant data distributors.

2. Global/Offshore Market: Sports as a Necessity and Product Shortage

Sports as a Necessity (Largest Vertical): Sports is the most certain growth track globally, with 92% of Kalshi's recent growth coming from sports, and high TVL of on-chain protocols like Azuro/Overtime validating its offshore necessity.
Middleware Strategy (Shovel Business): Investing in liquidity and odds middleware like Azuro is the best strategy, providing underlying infrastructure for all front-end sports betting apps, achieving "plug-and-play."
Solving "Parlay" Pain Points (Missing Piece): Leading platforms lack "parlay" functionality due to compliance avoidance; platforms that can achieve this function through decentralized means will create a significant siphoning effect on highly sticky betting users.
Crypto-native Short-term Betting (Limitless): High-frequency crypto price volatility predictions targeting DeFi traders (like Limitless model) represent another necessary niche market, meeting short-term speculative demands.

3. Asian Market: Social Fission and Long-tail Micro

Mobile-first and Social-driven (Market Characteristics): The Asian market is an undeveloped blind spot, with user preferences for mobile operations, social sharing, and high-frequency interactions, contrasting sharply with the institutional styles of Europe and America.
Abandoning Grand Political Narratives (Localization): Asian users lack interest in and are sensitive to European and American macro politics, and the strategy of replicating Polymarket's political betting should be completely abandoned.
Telegram Bots Reign (Best Entry Point): Building lightweight entry points based on Telegram bots (like Flipr, okbet) significantly lowers entry barriers, perfectly aligning with Asian users' social and payment habits.
Micro-entertainment Topics (Long-tail Strategy): Focusing on long-tail micro-events like esports, KOL gossip, and meme prices, utilizing Melee-style "creator revenue sharing" mechanisms to stimulate KOL self-promotion and fission.

IV. Analysis of Regulatory Agency Status

(1) Current Status of U.S. Regulatory System

At the federal level, the regulatory status can be referenced in the following diagram:

State and local (State Level) regulatory status:

1. The "Federal vs. State" jurisdictional battle in the U.S. is the most heated battlefield, and its outcome will have a decisive impact on the industry.

Kalshi and the CFTC argue that the Commodity Exchange Act has exclusive federal jurisdiction, claiming that its "event contracts" belong to financial derivatives and should take precedence over state laws; meanwhile, at least eight state gaming commissions, including New York, argue that this is merely regulatory arbitrage, a "backdoor" to evade state gaming licenses and taxes. This dispute is not only a matter of legal text interpretation but also a struggle for regulatory power and financial interests. The ruling in the Kalshi case will be a watershed: if Kalshi wins, it will clear the biggest obstacle for federal financial regulation; if it loses, all prediction markets may be forced to accept fragmented gaming regulation led by 51 states, with compliance costs that are unimaginable.

2. Case: Kalshi vs. NYGC Litigation

The conflict between Kalshi and the New York State Gaming Commission arises from the core legal contradiction between the Commodity Exchange Act (federal law) and state gaming laws. The incident began when the New York State Gaming Commission issued a cease-and-desist letter to Kalshi, accusing it of offering unauthorized sports betting. Kalshi subsequently filed a lawsuit in federal court, claiming that federal law has exclusive jurisdiction over its event contracts and emphasizing that federal jurisdiction should take precedence over state gaming laws to avoid the regulatory chaos brought by "51 different and conflicting state laws." The complexity of the case is further increased by tribal litigation, involving Kalshi's partner Robinhood, with California tribes also filing lawsuits and citing the Indian Gaming Regulatory Act (IGRA), expanding the jurisdictional issue to a three-dimensional federal, state, and tribal level, significantly increasing the legal and regulatory uncertainty of the case.

Current regulatory status in various countries:

The "borderless" or "decentralized" characteristics relied upon by prediction markets (especially DeFi platforms) are viewed as ineffective by regulators. Regulatory agencies in various countries are actively employing means such as geographic blocking, payment channel blocking, and ISP bans to enforce their domestic jurisdiction.

Globally (outside the U.S.), the regulatory trend for prediction markets is clearly tightening. Regulatory agencies are not letting them go unchecked due to their "innovative" nature; instead, they are rapidly incorporating them into existing regulatory frameworks (either financial or gambling) and firmly rejecting overseas operators attempting to evade domestic licenses.

Frankly speaking, the current regulatory situation is extremely chaotic, and it is nearly impossible to unify in the short term; different classifications will lead to entirely different compliance requirements and legal consequences. First, if classified as commodities or derivatives, the regulatory agency would be the federal CFTC (Commodity Futures Trading Commission), whose logic is that the market is used to hedge risks of future events (such as interest rates, political events), functioning similarly to futures, serving price discovery and risk management; Kalshi is striving to develop along this path and has already obtained CFTC authorization.

Second, if classified as gambling or betting, the regulatory agencies include various state gaming commissions (such as the New York State Gaming Commission) and international gaming institutions (such as France's ANJ and the UK's UKGC), whose logic is that the market is used to "bet" on future event outcomes, primarily for entertainment rather than economic risk management; real-world cases include New York State accusing Kalshi of offering unauthorized sports betting and countries like Belgium directly banning Polymarket.

Finally, if classified as securities, the regulatory agency would be the federal SEC (Securities and Exchange Commission), primarily targeting DeFi platforms, where expectations of platform profitability depend on the continuous efforts of the core team or the underlying assets being linked to stock prices, potentially triggering the Howey Test and being viewed as investment contracts; real evidence includes the SEC's ongoing scrutiny of token issuance in DeFi prediction markets.

(2) Mechanism Analysis: Business Model, Trading Mechanism, and Oracles

Business Model Analysis

Prediction markets are evolving from purely B2C transaction fee monetization to a B2B DaaS model that extracts data value, ultimately moving towards a hybrid model embedded in media and creator ecosystems to minimize customer acquisition costs.

Kalshi's Compliance Revenue Strategy: The platform relies on regulatory licenses as a moat, adopting a high-fee centralized model to directly capture B2C transaction revenue, aiming to dominate the compliant financial market trusted by institutions and conservative traders.
Polymarket's Share Priority Strategy: The project utilizes capital support to maintain extremely low fees to prioritize market share and data assets, building barriers through network effects and focusing on extracting long-term data value from B2B DaaS.
Limitless's Protocol Value Strategy: The platform adopts a DeFi-native model, directly empowering protocol tokens through an economic flywheel of fee buybacks, relying on token incentive mechanisms to compete for high-frequency liquidity in the crypto space.
Cultivate Labs' B2B Stability Strategy: This model avoids speculative markets, focusing on providing SaaS solutions for large enterprises and governments to assist decision-making, locking in predictable recurring revenue through solid customer relationships.
Potential and Challenges of Hybrid Models: Embedded models represented by Melee and Myriad have the largest potential market size by reaching general internet users, but still need to address the challenge of converting entertainment-based predictions into sustainable economic models.

Trading Mechanism Analysis: Liquidity, Matching, and Capital Efficiency

From theoretical models (CLOB, LMSR) to specific platform implementations (Polymarket, Kalshi, etc.), a systematic comparison of their differences in matching mechanisms, architecture, position mechanisms, capital efficiency, and risk characteristics has been conducted.

How to trade off "efficiency" and "decentralization"
Capital Efficiency (CEX Advantage): Models like CLOB (Central Limit Order Book) exhibit extremely high capital efficiency, providing low slippage and deep liquidity. This is a necessary condition to attract professional traders and high trading volumes.
Decentralization (The Soul of DeFi): Models like AMM (Automated Market Maker) are permissionless and censorship-resistant. This is the core value of prediction markets as "global information markets."
Current market leaders have not perfectly achieved both. They have all made compromises:
Kalshi: 100% efficiency, 0% decentralization. (Fully embraces regulation and centralization)
Gnosis/Omen: 100% decentralization, 0% efficiency. (LMSR is too inefficient, and no one uses it)
Polymarket: Hybrid compromise. (Exchanges off-chain matching for efficiency but sacrifices some decentralization and brings regulatory risks)
Limitless: Technical compromise. (Attempts to forcibly achieve on-chain CLOB with L2 speed but still relies on professional market makers)
The holy grail of the industry is to find a model that has the efficiency of CLOB and the passive and decentralized characteristics of AMM.
The old AMM's "clumsiness" lies in its uniform distribution of liquidity across the entire range from 0% to 100%, resulting in 90% of funds "sleeping" in a 50/50 market, leading to extremely low efficiency.
"Concentrated liquidity" (the core concept of Uniswap v3) is the holy grail to solve this contradiction, allowing liquidity providers (LPs) to "smartly" concentrate 100% of their funds in high-probability trading ranges like 40%-60%.
This new model perfectly integrates the two advantages: it achieves capital efficiency and low slippage akin to CLOB while retaining the permissionless, passive "set and forget" decentralized soul of AMM.
Therefore, the future of the market is a "smart AMM" tailored for prediction markets, centered around "concentrated liquidity" and potentially using Uniswap v4's "Hooks" to add advanced features like dynamic fees.

Oracle Mechanism: Trust, Security, and Settlement Efficiency of Outcome Judgments

Authoritarian Epistemology: Centralized Regulatory Model (Kalshi)
Mechanism: Based on legal rulebooks, a single authoritative data source (e.g., labor department reports) is designated by the platform for adjudication.
Source of Trust: Legal contracts and CFTC regulation.
Advantages: High legal certainty, suitable for institutions.
Limitations: Adjudication scope is limited by regulatory permissions (e.g., previously banned political predictions).
Consensus Epistemology: Decentralized Schelling Point Model (Augur, UMA)
Mechanism: Utilizes the Schelling point principle to incentivize token holders to vote on outcomes.
Augur: REP token holders vote, triggering forks in case of disputes (extremely slow and costly).
UMA (Polymarket): Optimistic oracle model. Defaults to passing without disputes; disputes escalate to DVM voting.
Security Assumption: Cost of Corruption (CoC) > Profit from Corruption (PfC).
Limitations: Governance attack risks. In 2025, UMA was manipulated by whales in the Ukrainian market, proving that token voting is vulnerable to attacks under low participation.
Judicial Epistemology: Decentralized Arbitration Court (Kleros)
Mechanism: Randomly selects jurors to adjudicate complex/subjective disputes, similar to an on-chain court.
Advantages: High expressiveness, capable of handling ambiguous/unstructured disputes.
Limitations: High costs, slow speed.
Emerging Mechanism Evolution
Pyth Network (Pull Model): Top institutions directly sign on-chain, high frequency, low cost, suitable for objective price data (like Limitless).
AI + TEE (Opinion/Buzzing): AI Agent automatically captures/adjudicates, with TEE hardware ensuring process trustworthiness, backed by the community. This addresses the adjudication costs in long-tail markets.
AVS Shared Security (EigenLayer): Utilizes re-staked ETH to provide economic security, addressing attack risks due to insufficient market capitalization of single oracle tokens.
Creators as Oracles (Melee): Allows creators to customize rules and adjudicate, adapting to social/entertainment long-tail markets.

For protocol developers: Reject "universal oracles," embrace specialization
Financial/Price Markets: Choose first-party data source models like Pyth or Chainlink, pursuing extreme performance and fidelity. Avoid using voting mechanisms to adjudicate high-frequency prices.
Objective/Binary Events: Adopt optimistic models based on EigenLayer AVS, ensuring that "corruption costs" always exceed "corruption profits."
Subjective/Ambiguous Events: Integrate decentralized arbitration systems like Kleros or accept Kalshi's centralized "rulebook" model.
Long-tail/Social Markets: Adopt Melee's "creators as oracle" model, clearly positioning it as a social entertainment product.
For bettors and investors, incorporating oracle risks into core considerations
Primary Risk Control: Before trading, one must read and fully understand the market's adjudication mechanism.
Clarifying Bets: Ask yourself, "Am I betting on the event outcome, or on the voting behavior of a group of anonymous token holders?" The 2025 Polymarket attack proves that the two can be completely at odds.
Risk Pricing: For markets relying on crypto-economic consensus, the risk of oracle manipulation must be factored into one's odds and position models as a core consideration.
From "Truth Machines" to "Verifiable Automation"
The development of prediction market oracles is a history of moving from idealism to pragmatism, and then to specialization and modularization.
The initial "truth machines" (like Augur) failed due to their complexity and inefficiency.
More pragmatic "optimistic models" (like UMA) had fatal flaws due to their isolated economic security and were breached.
Regulated "authoritative models" (like Kalshi) are technically robust but face strict limitations in their application scope.
The future will be neither fully centralized nor purely decentralized, but a combination of hybrid and specialization.
Hybrid: Integrating AI automation (to enhance efficiency), human oversight (to ensure security), and shared economic security (to address economic vulnerabilities), forming a composite system of "human-machine loops."
Specialization: Embracing specialized models like Pyth (focused on prices) and Melee (focused on social), achieving their core value by sacrificing universality.

Ultimately, the "oracle problem" will not be "solved" by a single mechanism but will be "managed" by an evolving ecosystem composed of multiple mechanisms. This ecosystem will provide different trade-offs for different types of "truth" based on the four dimensions of security, speed, cost, and expressiveness.

V. Compliance as a Growth Anchor

2025 marks the paradigm shift of prediction markets from marginal experiments to financial infrastructure. Under the dual catalysis of capital and regulation, the sector has formed a dual oligopoly structure of Kalshi (compliant finance) and Polymarket (offshore DeFi), with business logic transitioning from single transaction fees to B2B data distribution and embedded traffic monetization.

Compliance has become the growth anchor, and regulatory classification will determine the market ceiling. The industry is gradually differentiating into institutional paths relying on licenses and censorship-resistant paths relying on code; infrastructure is also developing in specialization, with matching mechanisms evolving from AMM to CLOB to enhance capital efficiency, and oracles upgrading from simple token voting to a hybrid defense system of "AI automation + AVS shared security" to address governance attack risks.

Looking ahead, the window for general-purpose platforms has closed, and growth will break through towards verticalization and socialization. One end is the deep institutionalization of sports and macro derivatives, while the other end is the "attention financialization" combined with the creator economy, leading to a coexistence pattern of high-frequency compliant data flows and long-tail social betting in the industry's ultimate outcome.

Disclaimer

This article is for reference only. It represents the author's views and does not reflect the position of OKX. This article does not intend to provide (i) investment advice or recommendations; (ii) offers or solicitations to buy, sell, or hold digital assets; (iii) financial, accounting, legal, or tax advice. We do not guarantee the accuracy, completeness, or usefulness of such information. Holding digital assets (including stablecoins and NFTs) involves high risks and may fluctuate significantly. You should carefully consider whether trading or holding digital assets is suitable for you based on your financial situation. Please consult your legal/tax/investment professionals regarding your specific circumstances. You are responsible for understanding and complying with applicable local laws and regulations.