Editorial notice: This article is for informational purposes and does not constitute financial advice or investment recommendations. Prediction markets are high-risk products, prohibited or restricted in many jurisdictions: Polymarket applies geoblocking in the United States, and regulatory treatment varies by country—in the UK they fall under the Gambling Commission, in the European Union they coexist with the MiCA framework for crypto-assets, and in markets like Australia or Singapore, online gambling is heavily restricted. Volume, user, and fee figures correspond to publications from May and June 2026 (data as of June 22, 2026), originate from third parties, and change daily; several are pending cross-validation. CleanSky does not receive commissions or referral payments from any of the cited platforms.
Polymarket's "World Cup Winner" market is nearing $3 billion in volume for a single event, even as the platform's user base has shrunk for two consecutive months. Between March and May 2026, active participants dropped from approximately 780,000 to about 650,000, highlighting the paradox of the 2026 World Cup: the money moved is breaking records just as fewer people are betting. The convenient narrative is that "crypto betting is exploding with football." The uncomfortable one, and the one this article defends, is that record volume with declining users does not describe a mass of fans betting, but rather a market concentrating in fewer, more sophisticated hands—bots, arbitrageurs, and institutionals. We are going to break down the World Cup figures on Polymarket and Kalshi, explain why volume and users are diverging, look at who really wins when 84% of traders lose money, and verify that the rest of the on-chain betting ecosystem (Azuro, Overtime, SX Bet) is not benefiting from the boom: the money is concentrating among the big two.
How much has the World Cup actually moved on Polymarket and Kalshi?
The figures for the summer of 2026 are the highest ever recorded by prediction markets for a sporting event. The protagonist is a single Polymarket market, the "World Cup Winner"—a prediction market is a contract where you bet on whether an event will occur or not, and you collect $1 per share if you are correct. As of June 15, that market had already accumulated about $2.34 billion (CryptoTimes), and six days later it was nearing $3 billion (BanklessTimes, June 21): it is the same market on two different dates, not two different metrics. These are magnitudes that, before 2025, only appeared during U.S. presidential elections.
Kalshi, the regulated competitor in the United States (operating under a license from the CFTC—the Commodity Futures Trading Commission, the U.S. derivatives regulator—and settling in dollars off-chain), moved around $344 million ($M) in World Cup contracts and posted a record weekly volume: the partial total at Saturday's close was around $7.49 billion, but the week closed near $8.5 billion, according to CBS Sports. Between the two platforms, the cumulative lifetime volume was around $150 billion as early as April 2026, according to tallies from The Block and Yahoo Finance.
The headline writes itself: the World Cup has turned prediction markets into the most visible showcase for crypto finance in 2026. The problem is that the headline only stays on the surface. Volume is how much money changes hands; it doesn't say how many different people are moving it or if they are winning. And that is where the crack begins.
Why are volumes rising while users are falling?
The divergence is the heart of the story. According to data collected by The Block and CNBC in June 2026, Polymarket's monthly volume in its domestic segment (Polymarket US) was around $1.16 billion—a figure well below the platform's global volume, which approached $7.1 billion in May—while the user base shrank for the second month in a row: the "participants" metric reported by CNBC fell from about 780,000 to about 650,000 from March to May 2026, a contraction of nearly 17%. This should not be confused with The Block's "active traders" tracker, which uses a narrower definition and placed the figure around 242,000 in June. CNBC had already reported on June 10 that May's monthly volume was down compared to the previous month despite the World Cup buzz.
A healthy activity metric rises in both dimensions simultaneously: more people and more money. When money goes up and people go down, the arithmetic conclusion is inevitable: volume per active user skyrockets. You don't need a sophisticated theory to see it. If 650,000 accounts move more than 780,000, each remaining account is moving much more. And moving much more per account is not the behavior of a fan betting $20 on their national team to win. It is the behavior of operators recycling capital dozens of times a day.
This fits a known pattern in crypto: the volume metric is easy to inflate and difficult to interpret. A bot that buys and sells the same position a hundred times generates a hundred times the volume of a human who buys once and waits. In a market where World Cup contract prices move every time a goal is scored, the opportunity to recycle capital is continuous—and those exploiting it are not retail investors.
Who really wins: retail, bots, or institutionals?
Here is the data that no promotional material puts on the front page. According to an analysis of on-chain data (authored by Andrey Sergeenkov covering about 2.5 million wallets on Dune Analytics) amplified by Bloomberg on April 28, 2026, and confirmed by The Defiant, around 84% of Polymarket traders lose money, and the accounts that win consistently "appear to be bots": that same analysis estimated that bots averaged about 89 trades per active day compared to ~2.2 for non-bots, a speed incompatible with a human. It is the classic structure of a zero-sum game (what one wins, another loses, minus the house fees), not that of a market where the majority benefits.
The value is captured by two groups. On one hand, algorithmic operators who arbitrage price differences between Polymarket, Kalshi, and traditional bookmakers, providing liquidity in exchange for the spread. On the other, the platform itself via fees. In March 2026, Polymarket moved from charging practically zero to a fee model that, according to Yahoo Finance, generated more than a million dollars in revenue per day. According to The Defiant and crypto.news, Polymarket concentrated around 96.8% of all on-chain fees in the prediction sector—acting as the "toll booth" for this DeFi niche—with some weeks seeing about $7.1 million in fees.
Translated: the World Cup generates a river of money, and that river passes through two cash registers—house fees and bot margins. The fan who enters to bet on the final provides liquidity to the system and, statistically, loses it. The "democratizing" narrative of DeFi—anyone participates on equal terms—is fulfilled in form (anyone can open a wallet and bet) and fails in substance (structural odds favor those with speed, capital, and code).
It is worth clarifying exactly what a bot does here, because "bot" sounds like fraud and it isn't. A prediction market trades in probabilities: a contract at $0.30 implies a 30% chance of the event. When a goal is scored, that probability moves in seconds. A human cannot re-quote at that speed; an algorithm can. The winning operators buy when the price lags behind new information and sell the imbalance—the same market-making function as in any stock exchange, compressed to the scale of a match. It's not cheating: it's a technical advantage, and it's what makes the image of the home bettor competing "on equal footing" meaningless.
| Platform / Metric | Figure (May-Jun 2026) | Source and Date |
|---|---|---|
| Polymarket "World Cup Winner" (same market, Jun 15 → Jun 21) | ~$2.34B → ~$3.00B | CryptoTimes / BanklessTimes |
| Polymarket, monthly volume (Polymarket US) | ~$1.16B | The Block / CNBC, Jun |
| Polymarket, active participants (Mar → May) | ~780K → ~650K | CNBC, Jun |
| Kalshi, World Cup contracts | ~$344M | CBS Sports, Jun 17 |
| Kalshi, weekly volume record (close) | ~$8.50B | CBS Sports, Jun |
| Polymarket traders who lose money | ~84% | On-chain analysis via Bloomberg, Apr 28 |
| Polymarket share of sector on-chain fees | ~96.8% | The Defiant / crypto.news |
| Azuro, total value locked (TVL) | ~$1M | DefiLlama, Jun |
The table tells the paradox at a glance: billions up, user base shrinking, a single platform taking almost all the fees, and a tier-2 reduced to symbolic locked value. It is not a market in democratic expansion. It is a market in concentration.
What is happening with the rest of the on-chain betting platforms?
If the World Cup boom were a tide that lifts all boats, we would expect to see second-tier decentralized betting platforms grow as well. The data shows the opposite. According to the prediction markets category on DefiLlama (an aggregator that measures TVL—total value locked—in each protocol), tier-2 platforms are marginalized:
- Azuro—a liquidity protocol for on-chain betting—carries a total value locked of around one million dollars, a fraction of what Polymarket moves in a single hour of the World Cup.
- Overtime (formerly Thales)—decentralized sports betting—maintains activity, but at a marginal scale compared to the big two.
- SX Bet—a peer-to-peer betting platform on its own chain—survives in a niche without a hint of the World Cup volume.
Money is not being distributed: it is migrating. Liquidity begets liquidity, and in prediction markets, liquidity is almost everything—betting against an empty order book means bad prices and high slippage. Polymarket and Kalshi have the depth; the tier-2s do not. An arbitrage bot operates where there is volume to hide its size and spreads to exploit: the big two. Thus, the World Cup effect reinforces the oligopoly instead of diluting it, and DeFi's promise of an open and plural ecosystem clashes with market physics: the winner takes almost all.
There is a category nuance. Polymarket is fully on-chain (settling on the Polygon network with USDC); Kalshi is a traditional regulated market that does not live on the blockchain. That the two leaders belong to different technical worlds and yet still starve native DeFi alternatives of oxygen says something unflattering: the user does not reward decentralization; they reward liquidity and usability. The on-chain tier-2 is competing with one hand tied behind its back.
Is the World Cup effect the "killer app" for DeFi or just activity without adoption?
Since the 2024 U.S. election campaign, when Polymarket prices were cited as a barometer in mainstream media, prediction markets have been sold as the use case that would finally take DeFi out of the speculative niche: real utility (aggregating information about the future), mainstream demand (sports, politics), and a stamp of legitimacy. The World Cup is the stress test for that promise, and the numbers leave it ambiguous: the "World Cup Winner" volume is nearing $3 billion (BanklessTimes, June 21) and the infrastructure handles peaks without breaking, but the participant base fell 17% from March to May and the on-chain tier-2 captures a symbolic TVL of ~$1M (Azuro, DefiLlama, June). A healthy "killer app" grows the pie and distributes it; this one grows the volume and concentrates it.
The distinction is the difference between activity and adoption. Activity measures transactions; adoption measures people who return and thrive. The 2026 World Cup provides a masterclass in record activity without corresponding adoption: 650,000 participants moving more money than the 780,000 from two months prior. For a sector that measures itself by user numbers and its democratizing narrative, it is a signal to watch, not to celebrate.
Is volume a vanity metric when users are falling?
Volume is every market's favorite metric because it is large, goes up easily, and sounds like success. It is also the most manipulable. In centralized exchanges, there is a well-documented history of volume inflated through wash trading (buying and selling against oneself to simulate activity). In prediction markets, you don't need to go to that extreme: it is enough for a handful of algorithmic operators to recycle capital at high frequency for volume to completely decouple from the number of real participants.
That is why the drop in users is the data point that matters, and why any "record volume" headline should be read with the follow-up question: how many different people are moving it? If the answer is "fewer than last month," the record describes concentration, not growth. For the fan tempted by the World Cup final, the practical lesson is direct: you are entering a market where, according to available data, the vast majority loses and consistent winners operate at a speed you cannot match. It is not a rigged casino, but the house and the professionals have a structural advantage—exactly like in any other zero-sum market.
There is an additional twist. The change in the fee model in March 2026—from free to over a million dollars a day—could have contributed to the exit of marginal users: when betting is no longer free, the casual bettor with thin margins is the first to leave, while the professional who already factors the fee into their math stays. Monetization and the drop in users could be two sides of the same coin: Polymarket is optimizing revenue, and revenue comes from those who stay.
What are the lessons from this World Cup for prediction markets?
The first lesson is methodological: always separate volume from users. The 2026 World Cup will be remembered for the billion-dollar headlines, but the data point that defines the real health of the sector is the decline in the active base for the second consecutive month. Record volume with shrinking users is the signature of a market maturing toward professionalized oligopoly, not expanding toward the mainstream.
The second is structural: in prediction markets, liquidity always wins. Polymarket and Kalshi accumulate the depth, the bots accumulate the margin, the house accumulates the fees, and the open DeFi ecosystem—Azuro, Overtime, SX Bet—is left watching with symbolic locked value. The boom does not democratize; it concentrates. To understand the mechanics of who extracts value, we have covered in detail why copying Polymarket whales usually makes you lose money, how different platforms compare in Polymarket vs. other prediction markets, and how Hyperliquid is trying to enter the niche with HIP-4. If you are starting from scratch, what is DeFi provides the framework to understand why the democratizing promise and the oligopolistic reality coexist.
The third is about expectations: just because a sector breaks records for money moved does not mean it is gaining users or that its participants are making money. The 2026 World Cup is the magnifying glass that reveals the real structure of the market—and the real structure is less democratic and more concentrated than the volume suggests.
Focus where it matters: CleanSky does not operate derivatives, predictions, or trading. With our application, you can track your portfolio and wallets, compare lending protocols and crypto cards, and maintain control of your positions—without promises of winning bets. The "record volume" narrative is better understood when you watch your own numbers, not third-party headlines.