Election odds manipulation risk flagged as midterm contracts boom

Days before the U.S. midterms, the integrity of the odds is itself a story. Concerns about election odds manipulation grew on Friday after CNBC published data from the Anti-Corruption Data Collective showing how cheaply some political contracts on Kalshi and Polymarket can be pushed around.

What happened

ACDC, a Washington nonprofit, found that 38 of 167 congressional markets with more than $100,000 traded, about 23%, saw price spikes of 30 cents or more. Across about 18,000 active congressional markets on Polymarket International, Polymarket U.S. and Kalshi, 88% could move 10 cents on a single order under $1,000.

Its September briefing went further: in 97% of live congressional markets, $3,500, the maximum personal campaign contribution, would buy a shift of at least five cents. On many Polymarket markets about $1 was enough to move a price five cents, and on Kalshi 63% of markets cost $25 or less to shift by that amount.

CNBC highlighted a Kalshi contract in the Illinois Senate primary where about $8,000 of buying briefly drove one candidate's price from 47 cents to 99 cents before it settled at 52 cents.

Why it matters

Odds increasingly shape perception. Wagers on 2026 congressional races are more than triple those on 2024 contests, ACDC estimates, and Kalshi says viewers outnumber traders five to one on midterm markets. A Brennan Center-funded survey found voters were 10 to 20 percentage points more likely to think a candidate would win when markets favored them.

The platforms push back. Kalshi says its favorites have won roughly nine times out of ten since 2024 and that large trades, such as a $1.25 million bet in the Los Angeles mayoral race, often revert within seconds. It added labels for thinly traded markets and live trade feeds, while Polymarket launched election-specific surveillance. ACDC counters that some moved prices stick for days. Regulators are already circling, as our CFTC market surveillance and Polymarket CFTC probe coverage shows.

Not every spike is manipulation. Prices can jump because a poll lands or because one trader buys a large block and clears out cheaper offers. ACDC itself says it cannot know the motive behind specific wagers, but notes that buying at 99 cents makes little sense as a profit strategy. The researchers' core point is structural: tens of thousands of markets were listed faster than real demand grew, which leaves many order books so thin that a few hundred dollars can tilt a headline probability. That is the gap election odds manipulation would exploit.

What's next

Tuesday's results will offer a live test of whether thin markets mislead. CNBC's investigation into election contracts has the full case studies.

Expect post-election reviews of spikes in close races. If evidence of deliberate election odds manipulation emerges, pressure for position limits on political contracts would rise quickly.

Readers using these markets as a guide should check volume and order-book depth first; a quoted percentage on a market with a few thousand dollars traded says far less than one backed by millions.

That caution applies to every platform, including the markets this site uses for its own forecasts.