Skip to content
Markets & Trends

Prediction Markets, Explained: How A $50bn-A-Month Asset Class Works, Why The FCA Is Rethinking A 2019 Ban, And What Traders Should Actually Understand

Kalshi, Polymarket and Polymarket US traded a combined $50.6bn in July - a monthly record - after a full year in 2025 of $51bn. Bernstein now expects $240bn for 2026 and a trillion by 2030. Cboe has launched regulated binaries on the S&P 500, Robinhood has put event contracts in front of 27 million accounts, and the FCA - which banned the sale of binary options to UK retail in 2019 - has held discussions with the platforms while British users reach them through VPNs, outside any consumer protection. This is the plain-English explainer: what a prediction market actually is, how the price becomes a probability, where the money is really made, why sports is 60% of the volume, what the regulators are fighting over, and the three things anyone trading these should understand before they do.

AlchmAI Editorial14 min read

$50.6bn

Combined July 2026 volume across Kalshi, Polymarket and Polymarket US - a monthly record, and almost the whole of 2025's $51bn

$240bn

Bernstein's estimate for 2026 volume, on a path it projects reaches $1 trillion by 2030

60%+

Share of volume that is sports contracts, with institutional demand for economic and political contracts expected to cut that to 30% by 2030

2019

The year the FCA permanently banned the sale of binary options to UK retail - the rule now under review as British users reach US platforms via VPN

Every few years a corner of the market goes from obscure to enormous fast enough that the people writing about it skip the part where they explain what it is. Prediction markets are having that year. In 2025 the sector traded $51bn in total. In July 2026 alone, Kalshi, Polymarket and Polymarket US traded a combined $50.6bn - a monthly record, and a full year's activity compressed into thirty-one days. Bernstein Research now expects around $240bn for 2026 and projects a trillion dollars of annual volume by 2030. Cboe has launched regulated binaries on the S&P 500. Robinhood has put event contracts in front of 27 million funded accounts. And in Britain, the regulator that banned this product for retail investors in 2019 has started talking to the platforms.

So this is the explainer. It is written for traders, investors and the people who build trading systems - all of whom are increasingly being asked about prediction markets by clients and colleagues - and it assumes no prior knowledge. Where we have a view, we say so.

What A Prediction Market Actually Is

A prediction market is an exchange for contracts that pay out on whether something happens. The simplest contract asks a yes-or-no question - will the Bank of England raise rates in November, will a named team win a match, will a bill pass - and pays a fixed amount, usually $1, if the answer is yes and nothing if it is no. You can buy 'yes' or buy 'no', and you can sell either before the event resolves.

The thing that makes these markets interesting is what the price means. Because the contract pays $1 on yes, a 'yes' trading at 65 cents implies that the market, in aggregate, puts the probability of the event at about 65%. Nobody set that number. It emerged from people with different views, different information and different amounts of conviction buying and selling against each other - which is exactly how a stock price emerges, except that here the thing being priced is a probability rather than a company. That is why economists have liked prediction markets for decades: they aggregate dispersed information into a single, continuously updated, publicly visible estimate, and they tend to do it well.

Where The Money Is Really Made

The romance of prediction markets is the political contract - the election, the central-bank decision, the geopolitical event. The business of prediction markets is sport. Sports contracts currently account for more than 60% of trading volume, which is to say that the fastest-growing corner of financial markets is, in volume terms, a betting exchange with a regulatory wrapper. Bernstein expects institutional demand for economic, business and political contracts to grow and cut sport's share to around 30% by 2030, and there are signs of that - hedgers using rates contracts, funds using macro-event contracts - but the retail sports flow is what has paid for the infrastructure.

  • Kalshi's weekly volume went from around $100m to more than $3bn in a year, and it did more than $30bn in June alone. It is a CFTC-regulated exchange, and its growth was accelerated by a March 2025 partnership that put its markets inside Robinhood's app.
  • Polymarket, the crypto-native platform, hit a record $10.8bn in a month and is formally re-entering the US through its acquisition of QCEX, a CFTC-regulated exchange, having received a CFTC no-action letter in January 2026. Its active traders fell from 733,000 in March to 643,000 in April - the seven-month growth streak ended - before the July record.
  • Robinhood's prediction-markets hub generates around $350m of annual recurring revenue, which is the clearest evidence that this is a real business line rather than a novelty.
  • Cboe Predicts, announced in June, puts yes-or-no contracts on the Mini S&P 500 inside the listed-options framework, cleared through the OCC and distributed by Interactive Brokers and Schwab - the exchange establishment's answer to the platforms.

The Regulatory Fight, In Plain English

The central legal question is whether a contract on whether a team wins is a financial derivative or a bet. In the United States that determines whether it is regulated federally by the CFTC, which oversees derivatives, or by the states, which regulate gambling - and the answer decides who gets the tax revenue and who can ban it. The CFTC has come down firmly on the side of federal jurisdiction: in January 2026 a new chairman withdrew proposed rules restricting prediction markets and Polymarket received its no-action letter, and on 2 April 2026 the agency sued Arizona, Connecticut and Illinois to defend its authority over event contracts. Andreessen Horowitz has argued in the CFTC's favour that state restrictions create a serious barrier to impartial access. Meanwhile the SEC delayed the launch of prediction-market ETFs expected in May - Roundhill, GraniteShares and Bitwise had filed for more than two dozen products tied to elections, recessions and other events - seeking more information on mechanics and disclosures.

Britain's position is different and, in our view, more revealing. The FCA classifies financial prediction-market products as binary options, and its 2019 ban on selling binary options to retail consumers remains in force. The practical result is that British users are reaching Kalshi and Polymarket through VPNs, trading outside any UK consumer protection, in volumes nobody can measure. In its March 2026 Perimeter Report the FCA said it would consider whether further work was needed on access to prediction-market products and whether the regulatory perimeter should be clarified; in September it was reported to have held discussions with the platforms about potentially easing retail restrictions.

“A ban that moves the activity offshore and out of sight is not consumer protection. It is consumer abandonment with a compliance certificate. The FCA appears to have noticed, and that is to its credit.”

Three Things To Understand Before Trading One

  1. 01Resolution risk is the risk. In a stock, the thing you own is unambiguous. In an event contract, someone decides whether the event happened, according to rules written before it did, and edge cases - a delayed announcement, an ambiguous outcome, a data revision - are where the losses live. Read the resolution source and the resolution rules before you read the price.
  2. 02Liquidity is concentrated and the tails are thin. Headline volume is dominated by a small number of very liquid markets. The obscure contract that looks mispriced is usually mispriced because nobody is there to price it, and you will find out when you try to exit.
  3. 03Price is probability only when the market is wide and the participants are informed. In a market moved by one whale or by fans of one team, the price is a measure of who showed up. Treating every contract price as a calibrated forecast is the single most expensive assumption in the asset class.

Why Platform Builders Should Care

We build trading platforms and charting systems, and event contracts are the first genuinely new instrument type to reach retail scale in a decade. They do not fit the tooling that exists. A binary's payoff is a step, not a line; its price is a probability that should be displayed as one, with the time to resolution and the settlement rule alongside; and its risk is resolution risk rather than price risk, which no existing risk panel shows. Platforms that bolt event contracts onto an options chain will produce something technically correct and unusable. The ones that render the contract as what it is - and that surface the disclosures binary payoffs require - will own the retail flow when it comes onshore. With Cboe inside the options framework, Robinhood at $350m of ARR and the FCA reviewing its perimeter, that is a design problem worth solving this year rather than next.

The Bottom Line

Prediction markets are exchanges for contracts that pay out on whether something happens, whose prices function as crowd-sourced probabilities - a genuinely useful mechanism that is also, in volume terms, mostly a sports-betting exchange. They traded $50.6bn in July after $51bn in all of 2025, are heading toward $240bn this year on Bernstein's numbers, and have pulled in Cboe, Robinhood and the CFTC on one side of a jurisdictional fight with the states on the other. Britain bans retail access under a 2019 rule aimed at a different product, which has pushed its users onto US platforms via VPN with no protection at all, and the FCA is now - rightly - reviewing that. For traders the essentials are three: resolution risk is the real risk, liquidity is concentrated, and price is probability only when the market is deep and informed. For the people who build trading systems, this is the first new instrument at retail scale in years, and it needs an interface that treats a probability as a probability. As a trading-platform and AI firm in London, we would rather Britain regulated this well and built it here than watched the volume leave through a VPN.

References & Further Reading

prediction markets explainedKalshi Polymarketbinary optionsFCATrading Platform DevelopmentAI Signal & Stock Rating IntegrationAI Agency UK
Share Email
AI

AlchmAI Editorial

Research and analysis, London

The AlchmAI team writes about the markets, technology and regulation we work with every day. We build trading platforms, real-time charts and AI analysis tools for brokers, prop firms and fintech teams from our office in Mayfair, London. Every article lists its sources. Nothing we publish is investment advice.

This article is general information and commentary. It is not investment advice or a recommendation to buy or sell any investment. Important information

Keep reading

All articles