By Simon Taylor, founder of Fintech Brainfood
Translated by Jiahua for ChainCatcher
“Imagine another hedging opportunity that could make you 13x in seven months.” That was how former CFTC commissioner and current Kalshi board member Brian Quintenz recently described one Kalshi contract on CNBC.
The contract was tied to how many times the Federal Reserve would cut rates in 2026. In January, the “0 rate cuts” contract traded at $0.06. It has since climbed to $0.82. On the example used in the piece, a $10 purchase in January would now be worth about $137. A comparable correct call expressed through fed funds futures, the instrument professional investors usually use to trade rate expectations, might have produced only a few cents. For businesses that depend on stable rates, that gap matters.
But prediction markets act like a Rorschach test. A former derivatives regulator can call a 13x return a hedge. A corporate treasurer may see insurance. A retail user tapping a phone app may see a 13:1 bet. They are all looking at the same contract, the same market, and the same price.
The appeal of prediction markets comes from their simplicity. A yes-or-no contract can be used to trade almost any question: Will the Fed cut? Will the Knicks repeat? Will Fintech Nerdcon be the best event of 2026? If a “0 cuts” contract trades at $0.82, the market is saying there is an 82% chance that outcome happens. If there are no cuts, the contract settles at $1. Price is probability.
These contracts also look like a strong business. Robinhood said in its Q2 earnings that prediction-market revenue surpassed both crypto and stock trading revenue for the first time. Event contracts brought in $156 million, compared with $129 million for equities and $100 million for crypto. Among trading businesses, only options were larger at $342 million.
Robinhood has long been good at using speculative products to capture consumer attention. If a speculative asset can bring users into the app, it is an effective acquisition tool. The article asks a blunt question in return: without those products, would Robinhood really have reached 28.4 million funded accounts?
There is at least some evidence that a share of users who arrive through prediction markets later build longer-term portfolios. Robinhood Gold subscribers reached 4.8 million, up 39% year over year. Net deposits hit a record $21.7 billion, and platform assets reached $369 billion.
The strongest case for prediction markets, in the author’s view, rests on three points: better hedging tools, more efficient user acquisition, and probability estimates that can outperform polling for people who need an answer in market form.
Still, the picture comes with three major problems. Sports are becoming the biggest category in prediction markets and are directly replacing part of traditional sports betting, which brings state tax revenue into the fight. A large share of contracts reach the market through CFTC self-certification, a process that depends on legal interpretations that can change with political leadership. Most important, prediction platforms are competing for the same attention as sports betting apps, yet the people flowing into these markets may not be covered by the same state or federal consumer protections used in traditional gambling markets, even as problematic betting behavior rises.
Platforms designated as DCMs, or designated contract markets, still have to comply with at least 23 core principles under CFTC oversight, and new requirements are being added. But if the legal disputes drag on, regulators change in two years, and consumers are not just customers but also part of the market’s input data, uncertainty is the only safe conclusion.
The author jokes that the regulatory fate of prediction markets may deserve its own event contract.
The larger point is more serious. The future of prediction markets will not resolve as cleanly as a yes-or-no contract, and the outcome may shape both financial market structure and consumer welfare.
Sports have become the main growth engine
The growth numbers look explosive.
According to the article, prediction-market volume reached $111 billion in Q2 2026, more than the combined full-year totals of 2024 and 2025, a year-over-year jump of 1,795%.
June was the biggest month on record at $52.7 billion, driven mainly by the World Cup. World Cup-linked prediction markets alone generated $17 billion in notional volume.
Prediction Atlas data cited in the piece shows 125 active prediction-market platforms, with Kalshi and Polymarket accounting for 91% of total notional volume.
There were 23 public fundraising rounds in Q2, totaling $1.288 billion. Kalshi raised $1.2 billion at a $22 billion valuation, and by late June the Financial Times reported it was discussing a transaction that could value the company at $40 billion. Intercontinental Exchange, the parent of the New York Stock Exchange, invested $2 billion in Polymarket. Polymarket only began charging fees in January, yet annualized revenue had already passed $1 billion by June.
Sports are doing much of the heavy lifting. On Kalshi, 87% of June volume came from sports. On Robinhood, 97% of open event contracts that month were tied to the World Cup. The tournament itself produced $17 billion in trading volume, and some estimates put the figure at $20 billion. During the competition, prediction markets accounted for about 27% of all U.S. sports betting activity, according to a Predicted report cited by Bloomberg.
In the first half of June, Kalshi and Polymarket together captured 73.5% of new sports-betting app downloads, while DraftKings had 13.7% and FanDuel had 8.9%. During the World Cup, Kalshi added 3 million users, and at peak periods daily trading fees topped $10 million.
The mechanics may differ from legacy sportsbooks, but the user battle is plainly the same. Apptopia data cited in the article shows Kalshi and Polymarket surpassing DraftKings and FanDuel in daily active users during the World Cup.
The author’s conclusion is that prediction markets are not creating a wholly new demand pool. They are pulling from mainstream bettors directly, and they are winning. That naturally leads to a blunt question: are they simply sports betting with a federal license?
Not entirely, the article argues.
On a traditional sportsbook, the platform is the counterparty. It sets odds to protect its margin, and users who win too much may be limited. Customer gains are house losses, so the model depends on a large base of users who lose slowly over time.
An exchange model works differently. The market sets the price, another trader takes the other side, and the platform earns from activity rather than directional outcomes. Betfair proved 25 years ago that this exchange structure can work. What is new today is the mix of regulatory licensing, dollar payment rails, and distribution.
But a platform that does not care who wins also creates a different outcome: the people who usually make money are professional traders and market makers.
In May 2026, the Wall Street Journal analyzed 1.6 million Polymarket accounts and found that just 0.1% of accounts captured 67% of all profits, while more than 70% of accounts lost money. Those advantaged traders, the so-called sharps, profit from casual flow provided by ordinary users.
Kalshi’s own data points in the same direction. For every profitable user, there were 2.9 losing users. Co-founder Luana Lopes Lara responded that even so, a person’s odds of making money on prediction markets were still better than in sports betting or day trading.
That leaves a market structure where retail participation drives growth, retail money often flows to professionals, and consumer protections remain thinner than in adjacent products.
The value goes beyond prediction
The article’s strongest defense of prediction markets is not that they are exciting. It is that they can price economically meaningful uncertainty in a direct way.
Quintenz’s Fed example is striking, but the author sets it aside because of his role on Kalshi’s board and looks at other use cases instead. Large prediction platforms already list markets tied to oil prices, stock prices, Fed policy, elections, inflation, and even “the highest-grossing movie worldwide in 2026.” If a company’s profits depend on any of those variables, the same hedging logic applies.
The odds may also be among the best available estimates of outcome probability. The piece says Federal Reserve staff studied Kalshi’s rate markets this year in a paper titled Kalshi and the Rise of Macro Markets. The study found Kalshi’s forecasts could match or outperform fed funds futures and large professional survey forecasts. Since 2022, ahead of each Fed meeting, it had a perfect record on the single most likely outcome.
The article adds an important caveat. This was a paper by Fed staff, not an official Federal Reserve position, and the authors also warned that market prices are not perfectly unbiased probabilities. Even so, the result stands out.
Today, someone trying to trade a Fed decision will often use fed funds futures or an instrument such as TLT, the long-duration U.S. Treasury ETF. Both carry basis risk. Stephen Sikes, COO of Public, put it plainly on the Tokenized podcast: 「I don’t want to trade TLT because there’s basis risk relative to the ultimate Fed decision. I just want to trade the Fed decision directly.」
That, the author says, is exactly what prediction markets offer.
They can even hedge indirect risks. The Jeffrey, a bar on Manhattan’s Upper East Side, once promised customers free drinks if the Knicks won. To hedge that promotion, the bar bought $5,000 of “Knicks win” contracts. When the Knicks did win, it collected about $8,000 from the contract, enough to cover the drink tab.
This was a real risk, but one too small, too binary, and too time-specific for an insurer to build a custom product around. Prediction markets could do it. The article notes that fewer than 10% of small businesses hedge foreign exchange risk, compared with 92% of Fortune 500 companies. Kalshi has already formally submitted a corporate hedging program to the CFTC, according to the piece.
Exchanges can also price outcomes that traditional sportsbooks or other markets generally avoid, especially outcomes driven by private decisions.
In July 2026, the market for “LeBron James’s next team” generated more than $245 million in combined volume across Kalshi and Polymarket. A regulated sportsbook would not normally build a business around pricing one person’s private decision with confidence. An exchange does not need to model that decision for its own book. The market finds a price on its own. In this case, the market got it wrong, and LeBron ended up in Philadelphia.
Some prediction-market hedges already resemble insurance closely. Spanish club Osasuna paid a €1.2 million premium for about €6 million of protection against relegation risk. The coverage was designed by broker Howden and, according to the article, executed through Kalshi, with quantitative trading firm Susquehanna taking the other side. Osasuna stayed up. Susquehanna made more than $1 million, and the hedge expired worthless for the club. That is what insurance is supposed to look like.
Yet this trade also exposes an uncomfortable truth. Osasuna could hedge because a highly sophisticated institution stood on the other side. Susquehanna, in turn, had reason to participate because there was broad consumer trading flow in the market.
Retail flow becomes liquidity for hedgers.
That is the central tension in the article. Prediction markets may be building a more efficient financial market, but part of that efficiency comes from less sophisticated consumers.
The piece gives another example from e-commerce. One company hedged inventory risk linked to the World Cup performance of a Latin American local team through a broker on Polymarket, in a trade sized in the hundreds of millions of dollars. By taking the opposite side of its own business exposure, it completed the hedge.
Dragonfly partner Rob Hadick told the author on the Tokenized podcast, 「The possibilities on the institutional side are almost unlimited.」
The article then turns to compute. Prediction markets may even be able to price future GPU rental costs. AI demand for GPUs is enormous, and Meta, Microsoft, Alphabet, and Amazon plan to spend hundreds of billions of dollars on AI capex, with a meaningful share financed through debt. Without a reference price for future GPU earnings, lenders have little basis for pricing credit against those assets.
That is why, on July 14, Kalshi launched a forward curve for GPU rental prices, described in the article as the first public forward price for compute. An H100 GPU rented for more than $8 per hour in 2023. According to the Ornn index, that figure has fallen to about $1.70. If a major input cost can drop 80% in three years and still spike later, a forward curve matters.
CME is developing compute futures, and ICE is working with Ornn on a cash-settled version, but both are still waiting for regulatory approval. Kalshi already listed its product because a licensed exchange can introduce event contracts through self-certification.
Lenders are not yet making loans against expected future GPU prices. But the author argues that if that changes, prediction markets could reshape debt capital market pricing with very little public attention.
These applications remain early. The distance from a platform favored by sports speculators to one that becomes part of the capital markets is still long, and legal and regulatory conflict sits all along that path.
Who regulates prediction markets?
The same contract can become three different things depending on which building it enters. At Cboe, it is a binary option regulated by the SEC. At Kalshi, it is a swap regulated by the CFTC. At a state sports gaming regulator, it may be treated as unlicensed sports betting.
Nearly everyone is fighting over that ambiguity.
The CFTC’s position is clear in the article: event contracts are derivatives. The agency is suing nine states to stop them from shutting down prediction markets and argues that event contracts fall under the CFTC’s exclusive jurisdiction. Since Michael Selig was confirmed as CFTC chair late last year, the agency has pushed that authority more aggressively.
The CFTC has also proposed a broad new framework. Under that approach, generalized sports contracts such as whether a team advances or wins a game could remain because they provide price discovery. Contracts on player injuries, referee decisions, or specific in-game events that run against the public interest would be barred.
Over the past four months, the agency has released more than 500 pages of new rules in an effort to close gaps quickly. It has also made clear that it opposes displaying contracts with American betting odds, even though DraftKings and FanDuel do exactly that. The author dryly wonders whether they wanted regulators to come after them.
States reject the CFTC’s reasoning. From their perspective, sports event contracts are sports betting products.
Kalshi and Polymarket now face at least 20 actions from state regulators, tribal groups, and individuals. More recently, attorneys general from 44 states sent a joint letter to the CFTC arguing that the agency has no authority to regulate sports event contracts.
Days later, New York entered the fight. The article says the New York attorney general sued Kalshi in Manhattan state court, seeking as much as $36 billion. One allegation is that Kalshi allowed 18- to 20-year-olds to trade sports event contracts on the platform. Attorney General Letitia James’s position was direct: prediction markets like Kalshi are betting platforms, full stop.
Traditional exchanges and sports leagues want firmer boundaries as well. CME’s general counsel wrote to the CFTC saying its definition of “gaming contracts” was an astonishing overreach that effectively displaced state sports betting oversight. The NFL also urged the agency to tighten the framework, saying the current approach does not adequately protect game integrity or consumers.
Then there is insider trading.
The author cites his own April essay, The Everywhere Insider: prediction markets were built to find truth, but the people who make the most money are often the ones who know the truth early.
In April 2026, U.S. federal prosecutors charged an Army Special Forces sergeant with using classified information about an operation in Venezuela to bet on Polymarket that Nicolás Maduro would be arrested, earning about $400,000. In May, a Google engineer was accused of using internal data to bet on search trends and making $1.2 million.
In July, Kalshi’s own monitoring system flagged a White House teleprompter operator who had been betting on whether Donald Trump would use certain words in a speech while having access to the script. Kalshi then froze more than $90,000 in that user’s account. Three political candidates also settled with Kalshi after betting on their own election outcomes.
The platforms have started responding. They have brought in professional surveillance services, required employees to disclose occupational information, and taken enforcement action. On insider trading at least, the article says the control systems appear to be starting to work, and the cost of getting caught is high.
That leaves the courts. Their answers are mixed. In April this year, the U.S. Court of Appeals for the Third Circuit handed Kalshi its first federal appellate win, ruling 2-1 that sports event contracts are swaps and therefore fall under the CFTC’s exclusive authority.
At the trial-court level, though, Kalshi’s arguments under the Commodity Exchange Act have already lost in New York, Maryland, Nevada, Michigan, Massachusetts, Utah, and Washington state, while it has won in Arizona and Tennessee. Several other cases remain unresolved. Minnesota even made trading sports event contracts a felony from Aug. 1, though a federal judge has temporarily blocked the ban.
The appeals continue, and at least one of these cases could eventually land before the U.S. Supreme Court.
The article also notes that states are not driven solely by consumer protection. In 2025, U.S. state regulators and local governments collected a record $18.09 billion in direct tax revenue from commercial sports betting and related industries. If trading volume migrates from sportsbooks to federally regulated prediction exchanges, that money no longer lands in state coffers.
Self-certification is central to the whole dynamic. It allows prediction markets to list new contracts quickly, but it also creates a large gray zone. In practice, Kalshi or Polymarket can list first and let regulators examine later. That helps explain why Kalshi already offers compute forward contracts while CME and ICE are still waiting.
Until the federal-state conflict reaches a final legal answer, the article expects prediction markets to keep expanding through that mechanism. In risk terms, the strategy resembles Uber’s playbook.
If you run a prediction market, the logic is straightforward: courts are unlikely to settle the issue soon, today’s regulator may be friendly, nobody knows what two years from now looks like, and every additional quarter of growth improves your leverage in future negotiations. On that logic, the rational strategy is to keep building, keep taking share, and keep litigating.
The author’s question is who, in that fog, is actually on the consumer’s side.
Faster growth makes consumer-protection gaps harder to ignore
The article argues that if prediction markets end up giving people with betting problems a more efficient way to lose money, the project has failed. Current data does not inspire confidence.
After sports betting legalization, the number of people seeking addiction help rose 61%, according to researchers at the University of California San Diego. Bloomberg data cited in the piece says Kalshi users have lost a net $294 million this year through parlay-like trades. In July, these contracts accounted for 36% of all Kalshi contract volume. During the World Cup final, one popular parlay had an implied probability of only 2.7% at kickoff.
Kalshi allows users to open accounts at 18. The National Council on Problem Gambling has called on prediction platforms to raise the minimum age to 21 and display help information prominently. New York’s lawsuit says 18- to 20-year-olds are already trading on Kalshi. Kalshi itself has acknowledged that most of its users lose money in the end.
What makes this harder is that these platforms market products in ways that closely resemble sports betting, including direct promotion of parlays.
No one buys a parlay for price discovery or for hedging, the author writes. A parlay is a dressed-up high-risk betting ticket. Its economics explain why it is one of the highest-margin products in sports betting. State data cited in the article says that for every $1 wagered, the average single-game bettor loses $0.06, while the average parlay bettor loses $0.19.
Advertising is another concern. The Wall Street Journal and Politico reported that Polymarket created a near-identical clone of its own website using a domain that swapped a lowercase l for an uppercase I, then gave it to college-age content creators who made videos showing how they could get rich through simulated bets. If those trades had occurred in live markets, more than half would have lost money.
To be fair, Kalshi has introduced self-exclusion, deposit limits, and mental health support features. The problem is that these protections are product choices, not legal requirements.
A traditional sportsbook operating in New Jersey must offer safeguards because the state says so. Users must be at least 21, and advertising faces clear limits. A prediction market operating nationwide, by contrast, can decide for itself how much protection to provide.
The reason is simple. CFTC rules were written for wheat farmers and swaps desks, not for retail users building World Cup parlays on a phone at two in the morning.
Stephen Sikes of Public told the author that his firm agrees with the “better market” argument and plans to offer prediction markets, but not sports. 「High-risk gambling products should not sit inside an investment account. We won’t do those products.」
The author says he respects that decision, but he also understands the cost. Speculative products such as crypto and, now, prediction markets sit at the top of the acquisition funnel for companies like Robinhood. Robinhood has 28.4 million funded accounts. Public recently disclosed more than 3 million members, roughly one-tenth of Robinhood’s base.
Many users arrive through speculation, and some stay to use longer-term investment products.
That creates a difficult question. If doing the right thing means surrendering the strongest growth engine to rivals that refuse to restrain themselves, how many boards will actually choose the right thing?
The article is skeptical that companies will impose those limits voluntarily. The rational business choice is still to push ahead. When everyone fights, ordinary people often take the loss.
How prediction markets could become better markets
The piece does not call for a ban. The data value is too large and the hedging uses are too real, the author argues. An outright ban would likely push retail users offshore to venues with even weaker protections.
Instead, he outlines four steps.
Build protections into the market itself
That would include cooling-off periods, position limits based on verified income, affordability checks before large trades, and incentives that reward forecasting quality rather than wager size. The author says he proposed similar ideas in October last year and does not believe they would destroy the market. Gaming companies learned long ago that friction can be a feature. Still, as these protections are added, a “market” starts to look more obviously like a gambling product.
Match risk warnings to the product’s actual behavior
If the National Council on Problem Gambling says the age floor should be 21 and help information should be displayed, then at least sports and parlay products should carry disclosures, age limits, and advertising rules closer to sports betting than brokerage accounts. The article also asks whether the industry can stop pushing parlays so aggressively, or at minimum attach stronger disclosures.
Create a path from bettor to investor
Robinhood already has the product stack for this, the author argues. Gold has 4.8 million subscribers, the platform offers retirement account incentives, and net deposits are at record levels. If speculation can bring users in, the same machine should be able to move some of them toward saving and long-term investing. There is already some evidence that prediction-market users later build portfolios, though the evidence is thin. The article argues that companies should measure that transition, publish the data, and place it in earnings reports alongside parlay revenue.
Separate sports event contracts and attach specific consumer protections
There are real sports hedging needs, the author says, but they are niche. Most sports event contracts are still competing directly with sports betting products. If so, a separate sports-related tax and a common set of position limits and consumer safeguards would make more sense than relying on voluntary platform choices.
If turning speculative impulse into financial health sounds idealistic, the article points to a British precedent from 70 years ago.
Putting lottery mechanics inside a savings product
In April 1956, the U.K. faced two problems at once: inflation and weak savings.
Chancellor Harold Macmillan wanted to pull more money out of circulation, but higher rates alone were not enough to persuade ordinary people to save. So on Budget Day he launched Premium Bonds.
The principal was not at risk. Instead of paying a fixed interest rate, the government pooled the interest and awarded tax-free prizes through a monthly draw. The draw was run by a machine called ERNIE, designed by an engineer who had worked on codebreaking at Bletchley Park.
The opposition denounced it as a “squalid raffle.” The public did not care. On the first day alone, Britons bought £5 million of Premium Bonds.
Seventy years later, Premium Bonds have become the most popular savings product in Britain, with more than 22 million holders. A single monthly draw now pays out about £447 million, or roughly $590 million, in tax-free prizes.
The lesson, in the author’s view, is that if society is going to live with more speculation, it can at least try to channel that impulse in a way that does not wreck household finances.
The U.S. legalized its own version in 2014 through prize-linked savings, but almost nobody scaled it. In a year when U.S. inflation once reached 4.2% and nearly every household felt the pressure of rising prices, turning speculative desire into saving is not some nostalgic policy relic. The mechanism is already there.
That connects back to the article’s opening contract. The event contract that offered a 13x return was, at its core, pricing inflation in real time. Quintenz’s “hedging opportunity” was looking at the same number Macmillan was trying to address in 1956.
A disappointed father of prediction markets
The article closes on a personal note. The author says he is always drawn to a thing’s potential.
Crypto is growing up. Stablecoins are here. Tokenization is here. And now there is another gifted but difficult child in the room: prediction markets.
Across his career, including digital assets, he says he has not seen another market product with this much potential. But the standoff among U.S. states, the CFTC, and prediction-market platforms has not solved what matters most: consumer protection.
He wrote in October last year that the industry probably had 18 to 24 months to build enough safeguards. If not, a backlash could arrive and regulators might overcorrect, destroying the genuinely valuable parts of prediction markets along with the harmful ones.
Ten months have now passed. New York is seeking up to $36 billion. Attorneys general from 44 states want to overturn the CFTC’s approach. The backlash, he says, arrived sooner than expected, and the U.S. midterm elections have not even begun.
His bottom line is clear. Prediction markets may be better markets, but for most ordinary people they are not yet friendlier than other high-risk betting products.
He still believes that can change. The companies that manage to survive, grow, and push their products in a healthier direction are preparing for the next era. That era may not arrive until something breaks and the courts finally produce a definitive answer.
His hope is that exchanges, states, and regulators wake up sooner and decide to do a better job before then. Tomorrow is another day.