"Los Angeles Mayor Race 2026: Polls Say Raman, Markets Just Flipped to Bass Before the November 3 Runoff"

Six days before Los Angeles votes, the polls and the prediction markets are telling opposite stories about the same race, and one of them has just changed its mind. The runoff on November 3 pits incumbent Mayor Karen Bass against City Councilmember Nithya Raman, the first Los Angeles mayoral runoff involving an incumbent in more than two decades. Every independent poll tracked for the race puts Raman ahead. The markets, until about a day ago, agreed with them. Now they do not.

The market move is the sharp part. On Polymarket, the price for a Raman win stood at 54.3% roughly a day ago and is now at 40.5%, a fall of nearly 14 points, on about $157,000 traded in the last 24 hours. Kalshi tells the same story: Raman is now priced at 42% against 57% for Bass, after standing at 56% the day before. Manifold's market, where traders use play money rather than dollars, lines up too, at about 42% for Raman and 58% for Bass. Agreement across three different crowds is harder to dismiss than a single price twitching.

The polls have not moved with them, at least not in the published record. The latest independent survey, by McLaughlin and Associates on October 15-18 among 400 likely voters, put Raman at 47% and Bass at 39%, with 14% undecided. An Impact Research poll on October 2-7 had Raman ahead 43% to 33%, with 23% undecided. A Berkeley IGS poll in mid-September, the largest of the lot at 2,150 likely voters, put Raman at 39% and Bass at 28%, with a third of voters undecided. Different pollsters, different months, the same leader. The gap between the polls and the markets is now roughly ten points in opposite directions, and it is the undecided voters, somewhere between 14% and a third of the electorate depending on the poll, who will decide which side was reading the race correctly.

So what changed in the markets, if the polls did not? No single event cleanly explains a 14-point swing, but the last few days have handed Bass-friendly traders several things to trade on. Bass announced a utility-bill relief plan on October 21, letting strained customers pay what they can afford and pausing shutoffs for the rest of the year; Kalshi's news service noted her price rose nearly six points in 24 hours around that announcement. More recently, reporting noted that former rival Rick Caruso, who lost to Bass in 2022, has endorsed her for this runoff, and that the Raman campaign filed an ethics complaint over advertisements it says falsely portray Raman promoting homeless camps, steering viewers to an interactive phone line using a voice identified as hers. None of that proves causation, and markets can overreact to a news cycle. But a trader watching an incumbent collect endorsements and announce bill relief while her challenger disputes attack ads could reasonably conclude that late momentum is real, even if the polls have not caught it yet.

The underlying contest helps explain why both readings feel plausible. Bass's standing has been battered by her handling of the 2025 Palisades and Eaton fires, which destroyed more than 16,000 structures while she was out of the country, and her campaign has leaned on wildfire recovery and the argument that experience matters with the 2028 Olympics approaching. Raman, endorsed by the Democratic Socialists of America and by County Supervisor Lindsey Horvath, has campaigned on wildfire crisis response, homelessness, and policing, including a proposal to replace LAPD Chief Jim McDonnell. This is not a race about one gaffe or one endorsement. It is a referendum on an incumbent's record after a disaster, fought by a challenger who has led the polls for months.

My own reading sides with the polls, but only narrowly, and the reason is the shape of the lead rather than any single number. Raman does not lead in one poll; she leads in all of them, across pollsters with different methods and sponsors, over two months. A lead that survives that much variation is usually real. Against that, I weigh three things: the undecided share is enormous, late-deciding voters often break toward the better-known name, and the market flip itself is evidence that people with money at stake have seen something the published polls have not yet measured. If the polls are right, the flip will look like a late overreaction to endorsement news. If the markets are right, the polls will have missed a genuine late swing toward the incumbent. The vote on November 3, with counting that may take days as it did in 2022, will settle which instrument was listening more carefully.