Polymarket traders: Democrats 92% to win the House, Republicans 35% to keep the Senate

01 · The oddsRepublicans are now the underdogs in the Senate

In January, traders on Polymarket expected Republicans to keep the Senate. They put the chance at about two in three. Now they put it at 35%. They also give Democrats a 92% chance of winning the House.

Polymarket is a betting site for real events, like elections and football games. You buy shares in an outcome. Each share pays $1 if it happens and nothing if it doesn't. So a share's price is the crowd's guess at the odds: 35 cents means about a 35% chance. These prices now sit next to the polls in news coverage, and Newsweek ran the Senate price beside two forecast models in September.

The site settles every bet on Polygon, a public blockchain, so every trade lands on-chain. Each trader shows up there as a wallet, an account on the chain. Bitquery indexes Polygon, the same data we used to measure how much of that chain this one site uses. We went through every trade on the site's midterm markets since September 2025. We tracked the Senate odds day by day, looked at who moved them, and checked them against the site's other Senate markets. One limit up front: we only see the on-chain side. Polymarket's US app and its rival Kalshi keep their books off-chain.

02 · The slideHow the Senate odds flipped

The odds didn't fall in one go. They dropped in steps, and the market got much busier along the way.

A year ago the Senate market was a ghost town. Last October only 82 wallets traded it, and the price said 73% for the GOP. By January a few hundred wallets were in, and the GOP sat near two in three.

The first big drop came the week of the November 2025 votes, when Democrats won the races for governor in New Jersey and Virginia. The next came in March. Texas held its primary on 3 March, and Democrats picked James Talarico outright while the GOP race went to a runoff. Within two weeks the GOP had slipped below even odds for the first time.

The party bounced back to 59% by late June. It didn't last. The price bled out over the summer, then fell hard in mid-September, losing 13 points in nine days.

That drop came right after Nate Silver's forecast model moved. In the week before, it took Democrats from about even to 59%, on new polls in Texas, Ohio and Michigan, Newsweek reported. The dates line up. The data can't tell us which one moved the other.

03 · Who moved itOne whale against the crowd

On Polymarket the price moves when someone pays more than the last trade, or sells for less. So who did the buying? We sorted every wallet that traded the Senate market over the past five weeks by how much it traded.

Every group bought more Democratic shares than it sold. That holds for wallets that bet less than $100 and for those that bet tens of thousands. Add them all up and the crowd went long the Democrats by about 210,000 shares.

Every share someone buys, someone else sells. So if the crowd is long, somebody is short. Here that somebody is one account. Its name on the site is lomneivan, and it opened this March. It is the biggest whale in the Senate markets, and it is all-in on the GOP. It has a net $1.03 million riding on Republicans, in the market on Senate control and in six state races. Two thirds of it sits in South Carolina, Kansas and Ohio.

In South Carolina, the account was on one side of nine in ten dollars traded, and in Kansas 58%. If Republicans win every race it backs, the account collects about $1.78 million. At this week's prices its bags are worth a little less than it paid.

04 · The seatsThe seats that decide the Senate

That whale's money sits in the races that decide control. Republicans hold the Senate 53 to 47. This year 35 seats are up, with special votes in Ohio and Florida to fill seats left open last year. Democrats need a net gain of four. A tie goes to the GOP, because JD Vance, as vice president, breaks it.

The site runs a market on every race. Here is each one, with the price in the first week of October and a month before.

Traders back someone other than the Republican in five seats the GOP holds today: North Carolina, Alaska, Texas, Ohio and Maine. North Carolina's seat is open, and Democrats are heavy picks there. Democrats also lead in every seat they already hold. The closest of those is Michigan, at 72%. If every one of those leads held, Democrats would end up with 52 seats.

Maine went the other way. The Democratic side there fell from 70% to 58% in a month, the biggest move toward the GOP of any close race. After those five, the closest GOP seats are Iowa, Kansas and Nebraska. In Nebraska the man taking on the GOP is an independent, Dan Osborn, at about one in four.

05 · TexasPolymarket's Texas Senate odds: from long shot to favourite

No Senate race has drawn more money. Texas has traded $1.2 million from about 3,700 wallets, more than any other Senate seat on the site. It also tells the whole story in one market, and its full trade history sits in our Polymarket dataset.

Last November traders gave the Democrat about one chance in five. The price climbed after the March primary, when James Talarico won the nomination outright. When Ken Paxton, the state's attorney general, beat Senator John Cornyn in the May runoff, traders first moved toward the GOP, from 45% to 38% in two days. Then the Democrat climbed back through the summer and reached even odds in August.

The big jump came in mid-September, from 51% to 62% in eight days, the same stretch in which the main Senate price fell hardest. Talarico now trades at 64%. A month ago the race was a coin flip.

Two other races drew big money. In Alaska, Mary Peltola, a Democrat who used to sit in the House, led Senator Dan Sullivan in the August primary and trades at 72%. Maine has drawn $1.1 million, second only to Texas, and it is the closest of the five seats leaning away from the GOP.

06 · Three readingsDo Polymarket's other markets agree?

Add up all those race markets and they should match the main Senate price. The site also runs two more Senate markets that should agree. One asks how many seats the GOP will hold. The other bets on both chambers at once.

Two of them do. The seats market gives the GOP a 34% chance of 50 seats or more, which means control with Vance's vote. The Balance of Power market, which bets on the House and Senate together, says 35%.

The race markets are the odd one out. Add them up one by one and they give Democrats a 65% chance of 51 seats, the same as the main price. But they give the GOP only 27% of reaching 50. The two don't add up to 100% because some seats could go to independents. More than half of that gap is Nebraska. If Osborn wins and Democrats fall short of a majority, neither party gets control without his vote.

One more catch: races tend to swing together in a big year, and adding them one by one misses that. So treat the race markets as a rough check. They back the Democratic side of the main price.

07 · The HouseDemocrats are 92% to win the House

The House is less of a contest. Its odds drifted toward Democrats for most of the year, as the first chart shows. A separate market gives Hakeem Jeffries, the House Democratic leader, an 84% chance of becoming the next Speaker.

The Balance of Power market makes a clean sweep for Democrats the most likely result, at 64%. A split, with the GOP keeping the Senate and Democrats taking the House, comes next at 28%. The House market has also drawn 2.7 times as much money as the Senate one.

08 · The moneyWhere the $103.4 million went

All of these odds rest on $103.4 million of bets.

Control of Congress took $18.8 million in all. The single biggest market was the race for governor of California, at $14.6 million. For scale, Polymarket's short up-or-down crypto bets traded more than ten times as much in their busiest month, March 2026. Those markets keep trading for a short while after their clock stops, and some traders race to bet in that gap.

The money is top-heavy. Most wallets bet pocket change, and a few hundred whales make the market.

Most of the wallets, 59% of them, bet less than $100, and together they put up under 1% of the money. The 321 wallets that traded $100,000 or more put up more than half. Across the whole site, a small group of round-the-clock traders makes most of the trades, as we found when we measured how much of Polymarket is bots.

09 · TrustHow far to trust the price

So how far should you trust a price set by a few hundred whales? Polymarket has a record to check this year. In 273 primary races for the Senate, the House and governor, the favourite on the morning of the vote won 87% of the time. Its picks didn't always win, though. Of the favourites priced at 80 cents or more, 13 lost.

Outside checks line up too. Newsweek put the Senate price at 52% for Democrats on 8 September, and our data gives 51% for that day and the one before.

A price is only as good as the money behind it. One whale can swing a thin market, and lomneivan was on one side of most of the trading in South Carolina. Prices can also move on news some traders get first, as with the bets the CFTC is looking into and the accounts that beat the odds on KPMG's clients.

Polymarket even lists bets on its own Senate price. One asks whether the GOP's odds will dip to 25% by the end of October, and it trades at 19¢. Anyone holding a lot of that bet gains if the main price falls. We have not checked whether anyone has tried to push it.

10 · The answerSo, can Democrats flip the Senate?

Traders think so. They give Democrats a 65% chance, a full reversal since January. Over the past five weeks the crowd kept buying the Democratic side, and one whale with more than $1 million on the GOP took the other side of it. The seats market and the Balance of Power market back the same number, and the race markets agree on the Democrats' chance.

The bet settles on 3 November, and one side of it gets paid. If Republicans win every race lomneivan backs, the account collects about $1.78 million. Traders even give the vote itself a 98% chance of going ahead on schedule. You can track any of these markets through our Polymarket API.

11 · The recordAddresses behind this story

FAQ

What are Polymarket's odds for the 2026 midterms?

In the first week of October 2026, Polymarket's traders gave Democrats about a 65% chance of winning the Senate and a 92% chance of winning the House. In January they had the GOP as the favourite to keep the Senate.

Can Democrats flip the Senate in 2026?

Polymarket's traders give them about a 65% chance. Democrats need a net gain of four seats, and traders back someone other than the Republican in five seats the GOP holds now: North Carolina, Alaska, Texas, Ohio and Maine.

How many Senate seats are up for election in 2026?

There are 35 seats on the ballot, with special votes in Ohio and Florida. Republicans hold the Senate 53 to 47 going in.

What are the Texas Senate odds on Polymarket?

In the first week of October, Polymarket gave Democrat James Talarico a 64% chance against Republican Ken Paxton. Last November the Democrat was a long shot at about one in five. Texas has drawn more money than any other Senate race on the site.

How does Polymarket work?

People buy shares in an outcome. A share pays $1 if the outcome happens and nothing if it doesn't, so its price is the crowd's guess at the chance. The site records every trade on Polygon, a public blockchain.

How much money has been bet on the midterms on Polymarket?

About $103.4 million traded on Polymarket's midterm markets on Polygon in the 13 months to early October. Control of Congress took 18% of it.

Is Polymarket accurate for elections?

In this year's primary races, the favourite on the morning of the vote won 87% of 273 races for the Senate, the House and governor. Its odds are only as good as the money behind them, and a few accounts dominate some of the single-race markets.

Where can I get Polymarket data?

Bitquery records every Polymarket trade on Polygon. You can query it through the Bitquery Polymarket API or buy the full history as files from the Bitquery Data Store.