Prediction markets are moving from niche instruments to embedded components of the broader financial system. As they begin to sit alongside equities, FX, options, and derivatives within established trading environments, the question is no longer whether they can attract users, but whether they can operate at the standard required of mainstream financial infrastructure.
In the first part of this feature, we examined how that transition is unfolding through distribution, liquidity, and market structure. We looked at how firms like Interactive Brokers are integrating event contracts into multi-asset platforms, and how infrastructure providers such as Devexperts position themselves within the stack. We also explored integrity and information dynamics through insights from Polysights.
In this second part, we continue the analysis by expanding into additional dimensions of the space, examining how these markets evolve under greater institutional pressure, where the remaining structural gaps lie, and what will ultimately determine whether prediction markets can move from integration to long-term stability within the financial system.
Why Prediction Markets Resemble Gambling More Than Investing
Tom Higgins, CEO of Gold-i, takes a direct and critical view of prediction markets, arguing that their structure aligns more closely with wagering than financial trading.
“I believe they should be regulated as wagering products,” Higgins told FinanceFeeds. “They are 100% gambling, not investing or speculating.”
That classification, in his view, explains why regulators remain cautious. “They hate them,” Higgins says when asked how international authorities see event contracts. He compares them to binary options, noting that they “require no skill or training” and often lead to “large losses for retail users.”
He also points to political contracts as a potential flashpoint. “Yes, I believe they could be a risk,” Higgins says, warning that visible sentiment in election markets could “sway voters one way or the other.” This, he suggests, raises broader concerns beyond market structure alone.
From a regulatory perspective, Higgins expects overlap with existing frameworks. “Yes, anything similar to prediction markets could be treated the same way,” he says, adding that this may extend to how authorities handled binary options in the past, where outcomes “ended up very badly for so many, with some people even in prison.”
Even outside regulation, he questions the industry’s core narrative. “They amplify narrative volatility,” Higgins says, rejecting the idea that prediction markets improve price discovery in a meaningful way.
He also highlights practical barriers for brokers. “Most of the existing platforms that brokers use for FX do not support prediction markets,” Higgins notes, pointing out that the transaction model itself is fundamentally different from traditional trading infrastructure.
Liquidity, however, is not where he sees the problem. “As the executing exchange or venue makes so much money out of this, it is highly sustainable by them,” he says. Instead, the concern lies with outcomes for participants.
“The losses outweigh the gains for most traders,” Higgins argues. “That does not bode well in the end.”
He also draws a distinction between event contracts and more familiar derivatives. “As the time-horizons are so short, it is really a different animal to short-dated options,” he says, though he acknowledges there may be overlap, particularly between short-term options and longer-dated prediction contracts.
Taken together, Higgins’ view stands apart from more optimistic takes on prediction markets. For him, the core issue is not technology or access, but the underlying structure — and whether it can sustain long-term trust among retail participants.
Why Demand Is Expanding Now
Retail: Simplicity of Event Exposure
Binary contracts reduce complex narratives to tradeable probabilities. Instead of modeling implied volatility or delta exposure, a trader expresses a view on a discrete outcome.
That simplicity scales. Retail traders increasingly trade catalysts rather than long-term valuation. Earnings surprises, rate decisions, election outcomes, ETF approvals, regulatory rulings — these are event-driven narratives. Prediction markets convert narrative into price.
Institutional: Event Risk as a Data Layer
Even when institutions do not trade directly, they monitor event probabilities as signal inputs.
Event contracts compress dispersed information into a single number. That number updates continuously. In a market structure driven by flows and expectations, that probability becomes a real-time sentiment indicator.
Prediction markets therefore serve both as instruments and as data products.
Media And Engagement Economics
Financial media thrives on forward-looking narratives. Probabilities provide a measurable, dynamic framing device.
As probabilities begin appearing in broadcast and digital environments, they normalize event trading as part of mainstream financial discourse.
Regulatory Classification: The Gatekeeper Variable
Prediction markets sit at the intersection of derivatives law and gaming law. Classification determines distribution viability. If categorized and regulated as derivatives:
They can integrate into broker ecosystems
• Institutional liquidity becomes more comfortable
• Marketing constraints become clearer
If treated primarily as wagering:
Distribution becomes geographically fragmented
• Broker partnerships become unlikely
• Banking relationships become fragile
The regulatory trajectory in major jurisdictions will determine whether prediction markets evolve into an asset class or remain a parallel ecosystem.
Election markets in particular function as regulatory stress tests. How regulators treat them signals broader policy intent.
Why Prediction Markets Are a Regulatory Test Case, Not a Niche Product
Daniel Lo, Managing Director and Chief Legal Officer at Acheron Trading, frames prediction markets first and foremost as a regulatory issue rather than a product question.
“Prediction markets are derivatives. They always were,” Lo told FinanceFeeds. In his view, the long-running debate around classification has not been about substance, but about control. “It’s been about jurisdictional turf wars between the CFTC and state gaming regulators.” Recent developments, including remarks from CFTC Chairman Michael Selig in January 2026, suggest that this balance may be changing. Lo describes the withdrawal of a proposed ban on political and sports contracts, alongside a commitment to formal rulemaking, as “a real shift in direction,” adding that “the federal floor is now being laid.”
That clarity matters directly for liquidity and institutional participation. “Liquidity providers need a clear rulebook before they commit serious capital,” Lo says, while institutional desks require legal certainty before building infrastructure, particularly around AML and counter-terrorist financing controls. He points to 2025 as a turning point, citing developments such as Polymarket re-entering the US as a CFTC-designated contract market, Robinhood acquiring MIAX’s exchange, and CME partnering with FanDuel. “These are institutions voting with their feet,” he says.
Lo sees prediction markets as part of a wider regulatory reset. “Prediction markets are something of a test case for broader digital asset regulation,” he notes. The fact that the CFTC and SEC are working toward a joint interpretation to define the boundary between commodity and security derivatives is, in his words, “exactly the kind of structural reform the industry has needed for a decade.” He argues that a regulator willing to act within its mandate, rather than relying on enforcement, is critical not just for prediction markets but for crypto more broadly.
Where Lo draws particular attention is surveillance and governance — areas he believes remain underdeveloped. “Any serious market operator needs surveillance infrastructure that mirrors what’s required of traditional derivatives venues,” he says. That includes monitoring “unusual position concentrations ahead of resolutions, coordinated wash trading, and suspicious activity around news events.” He flags material non-public information as a distinct risk. “Unlike equities, the insider universe in prediction markets can include political operatives, athletes, and journalists,” he explains, adding that “most platforms aren’t there yet” in terms of handling that asymmetry.
Governance around contract resolution is another weak point. “Resolution governance needs independence, clear escalation paths, and documented evidence standards,” Lo says. The strongest models, in his view, treat resolution “like an arbitral proceeding,” with predefined criteria, neutral review, and an appeal mechanism.
He contrasts two dominant approaches. Kalshi operates a centralized model, with resolution embedded in CFTC oversight, offering “regulatory accountability and predictability,” but concentrating discretion. Polymarket, by contrast, relies on UMA’s Optimistic Oracle, where outcomes are proposed and challenged on-chain. That model offers transparency, Lo says, but “can introduce volatility in decision-making,” particularly when outcomes are ambiguous or politically sensitive.
He points to recent failures as evidence that the issue is structural. “Resolution criteria are often drafted too loosely at the contract listing stage,” he says, leaving platforms to improvise when disputes arise. The solution, in his view, is straightforward but rarely followed. “Treat resolution design with the same rigor as legal contract drafting itself,” he says — define the oracle, define the evidence threshold, define escalation paths, and make all of it visible before trading begins.
On liquidity, Lo identifies regulation as the main constraint. “The single biggest bottleneck is jurisdictional fragmentation,” he says. A contract that is federally permissible may still trigger state-level enforcement, creating what he calls “asymmetric legal risk.” Until that tension is resolved — whether through federal preemption, court rulings, or legislation — “you will not see serious capital committing to unified, scalable infrastructure.”
A second barrier sits in compliance uncertainty. Institutional intermediaries still lack clarity on obligations around customer classification, reporting, and handling material non-public information. “Many institutions will sit on the sidelines not because they lack the appetite, but because their legal and compliance teams won’t sign off,” Lo says.
Despite those constraints, he is clear that institutions are already moving in. “They already are,” Lo says when asked about institutional entry, pointing again to acquisitions and partnerships across major firms. The question, he argues, is no longer participation but scale. “The question isn’t whether institutions enter, it’s whether the regulatory framework matures fast enough to let them operate at scale without legal exposure.”
Lo also flags insider trading as an unresolved systemic risk. “It’s a genuine systemic risk that the industry is underestimating,” he says. He points to real-world cases, including a trader who profited ahead of the capture of Venezuela’s president and another who correctly predicted Google-related outcomes at high accuracy. While detection methods exist — “size anomalies,” “timing relative to information releases,” and “account clustering” — the deeper issue is definitional. “In a political prediction market, is a campaign staffer trading on internal polling inside information?” he asks. “That legal question isn’t settled.”
Finally, Lo highlights how differently platforms approach contract resolution sources. Kalshi relies on “pre-specified, authoritative, and tamper-resistant” sources such as government data and official feeds, even signing licensing deals with leagues like the NHL. Polymarket, on the other hand, uses UMA’s Optimistic Oracle, where outcomes are proposed and challenged by users. While recent upgrades introduced restrictions and automated checks, Lo notes that the model’s weaknesses were exposed when a large token holder manipulated a vote, resulting in a $7 million loss.
