$431 million has been wagered on congressional prediction market odds. Two weeks before midterms, most of those odds can still be moved for $3,500 or less.
Prediction market odds backing congressional candidates can be shifted five cents for $3,500 or less in 96% of markets with live order books. The $431 million wagered across Kalshi, Polymarket and Polymarket US covers 28,845 markets, up sharply from 2024 and from 2020. Kalshi says $3,500 is a threshold, not proof of manipulation.
- How Cheap It Is to Move Election Odds
- Thin Order Books Underpin Every Market
- Kalshi Transparency Warnings Do Not Fix Data Access
- Wallet Concentration and Wash Trading
Prediction market odds can still be shifted five cents for $3,500 or less in 96% of congressional markets with live order books, according to the Anti-Corruption Data Collective. Researchers expected prices to harden as Election Day approached, but thin trading kept most of them loose. ACDC published the findings on November 2, covering 28,845 markets across Kalshi, Polymarket and Polymarket US.
Michelle Kendler-Kretsch, the ACDC research fellow who led the project, said the team was surprised by the result in September. She told Gambling Insider that markets lacked enough volume for another trader to counter an unusual order quickly. Her team deliberately studied the hardest market to move in each of the 470 congressional races, so the figures do not rely on obscure contracts.
How Cheap It Is to Move Election Odds
At least 66% of live congressional markets can move five cents for less than $100. Only 244 markets across all of them required more than $10,000. Even among the priciest contracts in each of the 470 races, 90% could be shifted five cents for under $25,000.
ACDC framed the finding as a depth problem rather than proof that anyone rigged a race. Ben Head, chief operating officer at UK-licensed bookmaker BetGoodwin, said the $3,500 figure simply shows how little money sits in most congressional contracts. When a few hundred dollars has been traded, the visible price reflects the last trade rather than a crowd consensus.
Matt Cepulis, founder and CEO of 3betmedia, argued the economics suit whoever wants cheap publicity. He spent nine years in cryptocurrency marketing and said a five-cent jump becomes a chart, a chart becomes a post saying momentum is shifting, and the post reaches far more people than ever inspect the order book. He called the screenshot the actual product.
Prediction market odds remain thin two weeks out
The ACDC figures show 89,453 unique wallets trading across 7,397 markets. A typical congressional market attracted just 48 bettors, and only 32 markets drew more than 1,000 participants. That concentration is the story: the top 1% of Polymarket wallets accounted for 75% of congressional volume, up from 68% in August.
Seven thousand markets cannot sustain 89,000 separate participants, so the order books behind headline candidates are still fragile. Kendler-Kretsch said the same five-cent jump that looks like a chart can be repeated by anyone who wants a candidate mentioned. The trade costs a few thousand dollars and the exposure is not confined to the order book.
Kalshi Transparency Warnings Do Not Fix Data Access
Kalshi added low-volume warnings and forecasting accuracy measures in October, and ACDC said 90% of congressional markets fall below the $10,000 volume threshold those flags use. Six percentage points more markets were flagged than in September. The labels help newcomers see that prices may be unreliable, but they cannot tell researchers which accounts are moving the market.
AGBrief covers prediction markets alongside broader gaming news.
Kendler-Kretsch said she still wants the underlying trading records. She asked Kalshi why an operator might list thousands of low-volume contracts at all, and pointed out that screenshots circulate long after traders restore the prices. A poll carries sample size, fieldwork dates and methodology; a prediction market figure often does none of that.
Wallet Concentration and Wash Trading
ACDC found 79% of congressional trades involved heavily funded accounts or algorithmic traders. In suspicious activity, researchers spotted 591 round-trip transactions across 22 Polymarket markets, executed by 544 wallets. The pattern was simple: buy a contract priced at 95 cents or higher, sell almost everything back within six hours, and keep a small profit.
Those transactions moved $618,395 and produced about $34 in combined profit. The average trade was $1,046 and finished in nine minutes, with nearly half completed in under five. One example involved Robert Lebovics, an outsider seeking the Republican Senate nomination in New Jersey. His implied chance of winning was about 3%, yet more than 100 suspicious trades worth roughly $250,000 passed through his market.
ACDC has not established wash trading or money laundering. Its concern is that near-certain outcomes offer a convenient route for inflating volume. Kimi Yu of AskSpinner drew a line between moving a price and changing a vote: the evidence supports the first, not the second. The Brennan Center survey cited in the report found respondents were 10 to 20 percentage points more likely to believe a market-favoured candidate would win, regardless of polling.
Frequently Asked Questions
What does the $3,500 figure mean?
It is the estimated cost to shift five cents on the odds of a congressional candidate. ACDC found that 96% of live markets with order books could be moved for $3,500 or less, days before midterms. The calculation measures market depth, not evidence that anyone manipulated an election.
How much was wagered on congressional markets?
Americans have wagered $431 million across Kalshi, Polymarket and Polymarket US, covering 28,845 markets. That is nearly three times the 2024 congressional election volume. Polymarket US has 61 times as many markets as the last congressional cycle, although the money is far from evenly distributed.
Why do most prediction markets stay thin?
Most congressional markets have little trading volume because they are numerous and not always interesting to general bettors. ACDC found the top 1% of Polymarket wallets drove 75% of congressional volume, while a typical market had only 48 bettors. Thin books let a trader shift prices without needing much capital.
Do Kalshi warnings make prediction market odds safe?
Kalshi added low-volume warnings and forecasting accuracy measures in October, but ACDC found 90% of congressional markets still fall below the $10,000 volume threshold. The labels help new users, while researchers remain blocked from the trading records needed to audit who placed the orders.
Can a few thousand dollars change voters?
Evidence points to the opposite. Kimi Yu of AskSpinner said price movement does not reliably move the ballot box, even if screenshots travel widely. ACDC found that $25.5 million was wagered on outcomes involving candidates who might learn about campaign activity in advance, but the report does not claim voters changed their minds.
What will researchers check after voting?
Once ballots are counted, ACDC will review markets where prices contradicted polling and try to identify the trades behind the moves. The results will not show why traders bet, but Kalshi and Polymarket hold the data. Until those records become available, outside researchers cannot test election odds claims.
This article has been thoroughly researched and reviewed by the CasinoBait editorial team to ensure accuracy and relevance for Asian casino players.
