Prediction markets turn changing election-contract prices into an easy-to-read measure of perceived odds. But an Ipsos poll of U.S. adults found that most respondents oppose betting on election outcomes. For the industry, the debate is about more than forecast accuracy: it also concerns who can trade, which events platforms list and whether bettors could influence the outcomes they wager on.
Ipsos surveyed 4,506 U.S. adults from September 30 to October 5. The results showed that 69% of Democrats and 61% of Republicans opposed using prediction markets to bet on elections. Respondents also opposed betting on military actions and government policy decisions. The poll measured support for the betting activity; it did not establish whether respondents trust market forecasts or explain why they objected.
That distinction complicates a common defense of prediction markets: that they aggregate information and improve forecasts. Someone may find election odds informative while still opposing the idea that people involved in an election, or its outcome, should be able to profit from it. The usefulness of a forecast and the case for allowing bets are separate questions.
How election contracts translate prices into odds
In a common election contract, a winning outcome pays $1, while a losing one pays nothing. If a candidate's contract trades at 60 cents, a buyer would gain 40 cents if the candidate wins and lose 60 cents if they lose. The price is commonly read as implying a 60% chance of victory.
Traders take positions based on their own assessments. Those who believe a candidate's chances are higher than the market price may buy; those who think the odds are lower may sell or take the opposite position. As trading continues, prices reflect participants' views. But a price alone does not show which traders had the greatest influence, or distinguish informed analysis from political preference or speculation.
Kalshi has taken steps to make market activity more visible. In an election-market update published October 6, the company added labels for low-volume markets, increased the visibility of its trading feed and provided scores comparing forecasts with actual results. Kalshi also said most visitors observe markets without trading. The idea that a relatively small group of traders can generate a price for a much larger audience is central to the platform's argument that its markets provide useful information.
Prediction contracts can also serve as hedges. A company exposed to losses from a government decision might buy a contract that pays out if the decision goes against it, potentially offsetting some of that loss—provided the contract accurately matches the company's risk. The other side of the trade may simply be betting on a different outcome, without having the same business exposure.
Yet ordinary voters who check election odds have no say in which events a platform lists. Kalshi decides whether to offer a contract; traders then bet on it, turning the event into a tradable product. Market prices can also feed into news coverage and everyday election discussion. That leaves platforms in a dual role: presenting themselves as sources of public information while operating commercial markets.
Candidate trading raises conflict-of-interest concerns
Kalshi has said it fined and suspended the accounts of four congressional candidates who traded in elections in which they were running. Candidates are both participants in an election and potential bettors on its outcome, creating the possibility that political interests and personal financial incentives overlap. Even without evidence of access to confidential information or an attempt to manipulate an election, profiting from a race in which a candidate is running raises an evident conflict-of-interest concern.
The cases point to two distinct risks. One is that someone with information unavailable to other market participants could trade on it. The other is that a bettor might use their influence to affect the outcome of a contract. For example, a campaign staffer who learns in advance that a candidate will withdraw the next day could trade before the announcement. The price might then more accurately reflect the outcome, but other traders could not obtain that information through ordinary research. If the staffer could also influence the timing of the withdrawal announcement, the issue would extend beyond an information advantage to the ability to affect the event being traded.
Kalshi says its rules bar relevant political insiders and that it screens accounts and investigates suspicious trading. The company also prohibits contracts related to war and death, limiting the events it will list. Where to draw the boundaries remains contested. A broad ban could push trading to platforms that are harder to investigate and hold participants accountable. But the claim that trading capital can improve forecasts does not resolve public objections about who should be allowed to participate or whether political outcomes should be commercialized.
Broader trust in election forecasts will depend in part on whether platforms make their pricing process and past performance open to scrutiny. That includes showing trading activity, comparing the probabilities available at the time with final results, and explaining how investigations and penalties are handled. Whether prediction markets are treated as reliable information sources will depend not only on the value of their odds, but also on how platforms manage information advantages, conflicts of interest and the range of events available to trade.