Predicting the Gold Rush in the Market: Who is Selling the Shovels?

Original author: https://x.com/oliveredge
Original compilation: Shenchao TechFlow

The gold rush of prediction markets has begun. Every founder in the crypto space, fintech entrepreneur, and contrarian believes they have found the secret to success. They are convinced they possess the prediction market platform that can outshine Polymarket and Kalshi. They have raised funds, assembled teams, launched dazzling interfaces, and promised better user experiences, faster settlement speeds, or niche markets overlooked by existing giants.

However, most are destined to fail.

This is not pessimism, but a mathematical outcome. In prediction markets, network effects are incredibly powerful. You need liquidity to attract traders, but at the same time, you need traders to build liquidity. In the crypto-native market, Polymarket already holds a significant advantage in scale. Meanwhile, in the U.S. event market, Kalshi has a regulatory high ground. The cost of shaking the positions of these two players is astonishingly high. From marketing to regulatory compliance to user acquisition, these expenses can accumulate rapidly. Even if new entrants achieve some success, they will only further dilute an already thin market. For platforms that rely on order book depth to survive, this is akin to a death sentence.

The "graveyard" of failed prediction market platforms illustrates this point. Remember the half-dozen markets launched after the 2024 election cycle? That's right, almost no one remembers them.

However, what venture capitalists should really focus on is that the true profit of prediction markets does not lie in operating these markets themselves, but in the infrastructure that supports their operation.

Why Infrastructure is a Superior Investment Choice

Looking back at the history of financial market development, the wealth of the stock market is not all created by exchanges, although some of it certainly is. The real wealth comes from data providers, clearinghouses, trading infrastructure providers, market surveillance systems, and deeper analytical platforms. For example, Bloomberg did not earn billions by competing with the NYSE, but by becoming an indispensable infrastructure.

Prediction markets are developing along the same trajectory, just decades behind. Currently, their infrastructure layer is still in its infancy, fragmented and inefficient, and this is where the real opportunities lie.

Here are several specific areas that venture capitalists should focus on:

Data and Oracle Infrastructure

The core of prediction markets is "real data." They need authoritative data sources to provide key information such as which candidate wins, what the actual GDP data is, or whether a company meets its targets. This seems simple but is actually complex. Different markets require different data sources, and they also need diversified verification and settlement mechanisms to prevent data manipulation.

Oracle networks designed specifically for prediction markets are crucial. These companies are responsible for aggregating data, providing cryptographic proofs, and resolving disputes. As the market expands, a decentralized oracle ecosystem will become unsustainable. The ultimate winner will be the infrastructure provider that all platforms—even competitors—must rely on.

Cross-Market Infrastructure and Aggregation

Current liquidity is dispersed across different platforms. A savvy trader might want to arbitrage between Polymarket, Kalshi, and three other platforms, but there is currently no seamless way to achieve this. If a set of infrastructure could be built to allow traders to view order books across all markets, it would be immensely valuable. Through this system, traders could simultaneously execute hedging operations and manage risk across multiple venues, unlocking tremendous potential value. This is the "Bloomberg Terminal" opportunity in prediction markets: every participant can benefit, and more efficient cross-market operations mean smaller spreads and deeper liquidity.

Analysis and Historical Data

As prediction markets gradually mature, researchers, quantitative analysts, and institutions will want to delve into historical prediction data. They will look for patterns and understand how the market prices events at different times. Someone will establish an authoritative repository of prediction market data, which will be cleaned, standardized, and made queryable. This will become a reference dataset for academic research, institutional analysis, and model building, forming a high-profit and defensible business.

Processing and Settlement

As prediction markets expand and become more complex, their backend systems will also need to be upgraded in sync. More efficient settlement mechanisms, faster data processing capabilities, and improved market infrastructure are all crucial. Companies focused on building middleware will hold immense value. They will connect markets to clearing systems, automate settlement processes, and reduce operational risks. This can be seen as the "pipeline system" that keeps modern markets running smoothly.

Compliance and Risk Management Infrastructure

As prediction markets gradually move towards mainstream acceptance and gain more regulatory clarity, complexity will follow. Infrastructure for managing regulatory reporting will become essential. At the same time, large-scale KYC/AML (Know Your Customer/Anti-Money Laundering) capabilities will become a necessity. Additionally, detecting market manipulation and ensuring compliance across jurisdictions will be key. Such infrastructure may seem "boring," but it is a highly defensible and sticky area. Once embedded in the market system, it is nearly impossible to replace.

Trader-Focused Infrastructure Layout

Another key aspect of prediction markets is the infrastructure support provided to professional traders.

Currently, the users of prediction markets are mainly retail and enthusiasts. However, as the market matures, attracting institutional capital, quantitative traders, and algorithmic traders will lead to a massive shift in demand. These professional traders not only need access to the market but also a complete set of tools that institutional finance takes for granted.

Algorithmic Trading and Trading Bot Infrastructure

Professional traders will want to automate strategies across multiple markets. This requires providing APIs, execution infrastructure, and trading bot frameworks specifically designed for prediction markets. In the future, someone may create a "Zapier" or "Make.com" for the prediction market space, allowing professional users to easily create complex trading strategies. With such tools, they can execute hedging and manage risks without writing code. Furthermore, there may even be companies developing infrastructure specifically for professional quantitative traders, enabling them to efficiently realize these functions.

Portfolio and Risk Management Tools

As traders accumulate positions across multiple prediction markets and platforms, they will need more advanced tools to support their operations. They need to track, manage, and understand their risk exposures. For example, what is the net exposure to political events? How correlated are these positions? What is the optimal hedging strategy? These questions may not trouble retail traders, but for institutions managing millions in prediction market capital, these will become core demands. The first platform to provide institutional-grade portfolio analysis tools will have the opportunity to capture a significant market share of serious funds.

Backtesting and Research Frameworks

Before deploying capital, institutional traders want to backtest strategies based on historical prediction market data. However, currently, this data has not been organized into a format conducive to backtesting, nor are there corresponding tools to support this need. Therefore, companies need to build robust backtesting frameworks, providing clear historical data and realistic simulations of market microstructure. At the same time, these tools need to integrate easily into existing research tools. This type of infrastructure will become a key pillar for the quantitative trading community to enter prediction markets.

Market Microstructure and Intelligence Tools

Professional traders understand that the market is not just about correctly predicting outcomes, but also about a deep understanding of liquidity.

They need to identify market inefficiencies, detect information flow, and precisely time their entries and exits. As prediction markets mature, the demand for real-time market intelligence tools will grow rapidly. Microstructure analysis tools will become particularly important, such as heat maps showing the flow of "smart money," real-time alerts for unusual activities, and tools for identifying mispricings. These features will serve a similar role to Bloomberg Terminals in traditional financial markets, but tailored for prediction markets.

Real-Time Aggregation and One-Click Trading: A Must-Have for Institutional Capital

For professional traders, trading simultaneously across multiple platforms is a basic requirement. In the future, platforms will undoubtedly emerge that can aggregate order books from Polymarket, Kalshi, and other prediction markets in real-time. Through such platforms, traders can view liquidity across all markets in one interface and execute cross-platform trades with one click. This is not only the dream of market makers but also a key infrastructure for the entire prediction market ecosystem to become more efficient.

This trader-focused infrastructure is as important as the market-side infrastructure. These tools are not optional "nice-to-haves," but necessary prerequisites for institutional entry. As institutional capital flows into prediction markets, these tools will become indispensable core elements. Companies building this layer of infrastructure will capture value types that differ from market operators. This value is not only highly defensible but also, to some extent, more scalable.

The Ultimate Question of Valuation: How Much Growth is Left for Prediction Markets?

Recently, the financing dynamics of the two major players in prediction markets have attracted widespread attention. Kalshi recently reached a valuation of $5 billion, while Polymarket achieved a post-money valuation of $9 billion thanks to investment from Intercontinental Exchange, the parent company of the New York Stock Exchange.

This is not a small increase. Just a few months ago, Kalshi's valuation was still at $2 billion, while Polymarket's valuation was only $1.2 billion at the beginning of 2025. In just a few months, these valuations skyrocketed by 2.5 to 7 times.

This raises an unsettling question for venture capitalists: how much upside is left for prediction markets?

Currently, these two companies have reached sufficiently high valuations that future exit multiples are constrained. Assuming Kalshi or Polymarket one day reaches a valuation of $50 billion to $100 billion, from the current base of $5 billion to $9 billion, this is undoubtedly a considerable but not astonishing return.

More importantly, these platforms are gradually becoming potential acquisition targets for traditional financial giants. Exchanges, brokers, and financial institutions have shown strong interest in them. Selling at 2 to 4 times the current valuation to ICE, CME, or other large brokers is entirely possible. But this is not the "power law" investment that venture capitalists pursue for 100x returns.

In contrast, investments in the infrastructure space exhibit a completely different return curve. Whether it’s oracle service providers, analytical platforms, or cross-market execution layers, once they become core infrastructure of the prediction market ecosystem, their returns will cover all platforms, all traders, and all institutions.

The starting valuations for such infrastructure are typically low, but their expansion potential is virtually limitless.

Asymmetry of Risk

In the competitive platform arena, venture capital firms often bet on multiple projects, hoping one will become the next Polymarket. This is a classic "power law" bet: most projects will fail, and even the successful ones may struggle to create significant value due to market fragmentation and liquidity segmentation.

In contrast, the risk curve for infrastructure investments is entirely different. For example, an oracle service provider does not care whether traders use Platform A or Platform B—regardless of which side wins, it can benefit. The value of an analytical platform increases with the number of markets, rather than decreasing. Infrastructure does not need to choose winners; it just needs to be useful across all platforms.

Moreover, infrastructure typically forms strong defensive capabilities through data advantages, network effects, or technological barriers. This is not merely a "burning cash" competition; it is a contest of technological depth and ecological stickiness.

What Does This Mean for Investors and Entrepreneurs?

If you are evaluating a business plan focused on building a new prediction market platform, whether it is pitched on better user experience (UX) or targeting an undeveloped niche market, you need to ask sharper questions:

• How will you solve the liquidity problem?
• How will you achieve profitability in the face of competition from existing giants?
• How many of the numerous competing platforms can realistically succeed?
• More importantly, even if successful, what is the actual likelihood of exit multiples from a base of over $100 million in funding?

If you are focused on opportunities in the infrastructure space, then you are facing a completely different risk and return model. Go build data layers, develop cross-market tools, design settlement mechanisms, create trader analysis tools, and establish smart intelligence platforms. These businesses will grow with the expansion of the entire market, rather than competing with a single competitor. They benefit from the prosperity of the market rather than suffering from its fragmentation. They offer the kind of unconstrained growth potential that venture capital truly seeks.

The ecosystem of prediction markets is still in its early stages, which means there are enormous opportunities. But the real opportunity lies not in replicating what Polymarket has already done, but in building the foundational layer that enables the entire ecosystem to operate more efficiently.

Platforms will battle it out, while infrastructure will continue to expand. ```