A prediction-market percentage beside a football team’s name expresses a price-based estimate of a particular result. It does not tell you what share of the public supports that team. The number comes from trading in event outcomes, rather than answers to a survey. Understanding whose view it captures means looking at the people placing orders, the question being traded, and the time the figure was recorded.
On 7 January 2026, Dow Jones announced a deal to carry Polymarket prediction data across publications including The Wall Street Journal, Barron’s, and MarketWatch. The plans covered digital data displays and selected print placements. Such arrangements put market estimates within reach of people who follow the news without trading themselves.
Trading activity is not a headcount
On crypto-backed prediction markets, participants trade digital tokens representing event outcomes. A token records a position on whether a specified result will happen. It is different from a survey response: it can be bought, sold, and held in varying quantities. Alongside prices, reporting may give a total for trading volume, which describes activity in those positions.
Polymarket’s World Cup market had recorded more than US$1.8 billion in cumulative trading volume, according to AlphaWire in a report dated 15 June 2026. That reported total describes trading activity accumulated by that point; it cannot establish how many different people participated or how many supported any team.
A position can change hands repeatedly, and one participant can make multiple trades. Counting trading activity therefore answers a different question from counting participants. Volume also differs from fresh deposits or the amount still committed to unresolved outcomes. Without a reporting period and a clear calculation method, a large total cannot support a precise comparison with another market’s activity.
How a price becomes a percentage
In a standard yes-or-no market, tokens represent the two possible outcomes. On Polymarket, each pair is backed by one unit of its dollar-denominated currency, Polymarket USD (pUSD). That backing is called collateral. pUSD is separate from the outcome tokens. When a market settles to yes or no, the corresponding token redeems for 1 pUSD, and the opposite token has no redemption value. A token’s price as a proportion of that fixed redemption amount is read as its implied probability: an estimate derived from prices.
The displayed number is not necessarily the price of the most recent trade. Polymarket’s pricing guidance says it normally uses the midpoint between the highest price a buyer is offering and the lowest price a seller is asking. Those offers are called the bid and ask. The midpoint sits halfway between them, so it summarises the available prices without counting people’s answers.
There is an exception: when the gap between those offers exceeds ten US cents, the displayed figure uses the last traded price. A midpoint need not be a price anyone has paid; the last traded price records an actual transaction. Under the midpoint method, new offers can change the displayed percentage before a trade is completed. The time of the underlying quote can also differ from the time a story appears.
Orders can cover different quantities of tokens, and participants can place more than one. The price consequently has no straightforward translation into a percentage of people who agree. It also provides no demographic description of those people. Treating it as the opinion of football supporters in England would require information that the price itself does not contain.
The event wording sets another boundary. Winning a match, reaching a final, and winning a tournament are different outcomes. Each market’s rules specify which result counts. A percentage has to travel with that wording and its timestamp. A rising figure records a change in the quoted estimate; by itself, it cannot establish what caused the movement or guarantee the eventual result.
What a fan survey measures
Before the 2026 World Cup, YouGov found that 46% of adults in England said they would support England. Respondents were asked about their attitudes towards teams, including whether they would support them. The percentage describes a stated preference within a defined population and period. It does not give England a 46% chance of winning the tournament.
Supporting a side and judging its prospects are separate questions, even when they concern the same competition. Nor should the result be widened to all British adults or described as the view of only dedicated football fans. The population named in the research is part of the findings.
Therefore, a support survey and a market forecast can differ significantly without contradicting each other. One records an answer about support; the other expresses a traded estimate about an outcome. To interpret a news percentage, identify its question, measurement, and date. Even matching questions would not turn a share of respondents into a market probability.
