Analysis

    NYC Primary 2026: Kalshi's 28% Sweep Odds, Three Mamdani Wins, and One Stunning Upset

    Kalshi priced Mayor Mamdani's full NYC congressional sweep at 28% odds. All three candidates won — including a 2,200-vote upset over a five-term incumbent.

    By Prediction Markets US Analysis DeskWednesday, June 24, 20268 min read
    NYC Primary 2026: Kalshi's 28% Sweep Odds, Three Mamdani Wins, and One Stunning Upset

    When New York City Mayor Zohran Mamdani endorsed a trio of progressive challengers for Congress, prediction market traders tried to price the political future. They called two races decisively. The third one surprised everyone—including the market.

    On Tuesday night, all three Mamdani-backed candidates swept their Democratic congressional primaries, confirming a leftward surge in one of the country's most watched political battlegrounds. Kalshi—the federally regulated prediction market—had priced the combined sweep at just 28% before polls closed. But before dismissing that as a miss, a closer read reveals a more nuanced story: two races ran exactly as traders expected, and one was a genuine upset that even well-resourced polling couldn't fully anticipate.

    The Three Races, the Market Prices, and What Actually Happened

    NY-10: The Rout Markets Saw Coming

    In New York's 10th Congressional District, Brad Lander—former city comptroller and the most prominent of Mamdani's endorsees—ran against incumbent Representative Dan Goldman. Before polls closed, Kalshi traders had Lander at 98% to win the Democratic nomination. Polymarket's odds were effectively the same.

    The results matched the market: Lander won with 65.8% of the vote to Goldman's 34%, a decisive 32-point margin. Decision Desk HQ called the race within minutes of poll closing.

    Goldman's loss was months in the making. Emerson College polling had shown Lander ahead by 34 points, and the district's demographics—affluent, progressive, and one of the most Mamdani-friendly in the November 2025 mayoral race—left little room for a reversal. Goldman, a lead counsel in the 2019 Trump impeachment hearings and an heir to the Levi Strauss fortune, had the backing of Governor Kathy Hochul and House Minority Leader Hakeem Jeffries. None of it moved the needle. For prediction markets, NY-10 was effectively a lock, and the final tally confirmed it.

    NY-7: Another Decisive Victory Markets Priced Correctly

    In the 7th District—covering parts of Brooklyn and Queens—state Assemblywoman Claire Valdez ran against Brooklyn Borough President Antonio Reynoso, the handpicked successor of retiring Representative Nydia Velázquez. Despite Reynoso's endorsements from Velázquez and the Working Families Party, Kalshi had Valdez at 82% to win. Polymarket's price was 77%.

    Valdez won with 56% of the vote to Reynoso's 36%, a margin that matched the market's confidence. The district—often called the "Commie Corridor" for its left-leaning politics—proved fertile ground for a DSA-backed candidate running with Mamdani's explicit blessing.

    Both major prediction markets rated this race as a clear Valdez favorite heading in. Both were right.

    NY-13: The Genuine Upset That Priced at 33%

    The biggest story of election night—and the one that makes this a genuine prediction market case study—came in New York's 13th Congressional District. Darializa Avila Chevalier, a community organizer and sociology Ph.D. candidate who helped lead the 2024 pro-Palestine protests at Columbia University, took on Representative Adriano Espaillat, the five-term incumbent and chair of the Congressional Hispanic Caucus.

    The market read going in: Espaillat at 67%, Chevalier at 33% (Kalshi individual odds). The sweep combo market—tracking whether all three Mamdani candidates would win—was priced at just 28%, reflecting the drag of Chevalier's underdog status.

    Chevalier won by approximately 2,200 votes, with 49.4% to Espaillat's 45.9% with over 96% of scanners reported. Espaillat conceded around 10:30 p.m. ET Tuesday, telling supporters, "It wasn't our night."

    The result was stunning. Espaillat had won his most recent primary unopposed and took the general election with over 83% of the vote in 2024. He had the backing of Jeffries, Hochul, and Attorney General Letitia James. Chevalier had never held elected office.

    What the 28% Sweep Odds Actually Meant

    The quick narrative—"markets got the NYC primary wrong"—misreads what prediction market probabilities represent. Kalshi's 28% sweep odds were not a failure of pricing. They were an accurate quantification of real uncertainty.

    A 28% probability is not "no chance." It's roughly the odds of rolling a 1 or 2 on a standard six-sided die. Low-probability outcomes materialize regularly—that's what probability means. Prediction markets perform well not when they call every result, but when their prices reflect the true distribution of possible outcomes.

    The useful breakdown from Kalshi's combo contracts heading into Election Day:

    OutcomeKalshi OddsWhat Happened
    Lander + Valdez win, Chevalier loses54%Did not happen
    All three Mamdani candidates win (sweep)28%✅ Happened
    Only Lander wins20%Did not happen

    Markets assigned the highest probability to a 2-for-3 split. What they got was 3-for-3. But they didn't ignore the sweep scenario—they priced it as a real, if lower-probability, path. That's a fundamentally different story than missing it entirely.

    For comparison: traditional polling might have listed Espaillat as the "favorite" without ever attaching a number to the uncertainty. Kalshi's 33% Chevalier price explicitly communicated that an upset was plausible—one-in-three odds is not an edge case.

    Why NY-13 Was Always a Market Challenge

    The 13th District race illustrates a known limitation of prediction markets in low-information electoral environments. In nationally prominent races—presidential contests, high-profile Senate battles with abundant polling—markets benefit from rich data. They aggregate surveys, fundraising signals, and real-money trader conviction to produce well-calibrated prices.

    Down-ballot primaries operate with far less information. There were only two meaningful public polls of the NY-13 race, and they diverged. Emerson College showed a competitive Valdez race in the adjacent 7th District; an earlier Data for Progress survey showed Espaillat ahead in NY-13. Many voters—43% in Emerson's survey—were undecided well into June.

    In that information vacuum, markets relied on structural indicators: incumbency advantage (historically strong), Espaillat's five-term tenure, Chevalier's first-run status, and the endorsement weight of establishment Democrats. The market's 67/33 Espaillat lean was a rational price given available data. Chevalier's victory doesn't invalidate the market's process—it confirms that low-probability outcomes do occur, and that the market correctly identified a 1-in-3 shot rather than pricing it at zero.

    Prediction Markets as Political Intelligence

    The integration of Kalshi data into mainstream pre-election coverage marks a significant shift in how political media covers down-ballot races. CNBC ran a dedicated piece on Kalshi's NYC primary odds the day before the election, noting that "traders on prediction market platform Kalshi think the mayor will go two for three." The article detailed the individual race prices and explained the combo contract structure to a general audience.

    This is the same trajectory prediction markets followed during the 2024 presidential election, when Kalshi and Polymarket prices became standard reference points for political journalists and analysts—cited alongside traditional polls rather than treated as fringe instruments. The 2026 midterm cycle appears to be accelerating that normalization into lower-profile primary races.

    What the NYC results add to that narrative: prediction markets are now visible enough that when the 28% scenario materializes, it becomes part of the next-morning political story. That's not a liability—it's evidence that the market is being taken seriously.

    What It Means for NYC's Congressional Delegation

    The political stakes extend well beyond market accuracy metrics. For the first time in years, New York City's congressional delegation shifts significantly to the left:

    Dan Goldman's defeat removes one of the Democratic Party's most prominent Trump-era figures—a lead counsel in the 2019 impeachment hearings who had become a fixture on MSNBC. Goldman spent millions of his own money and still lost by 32 points.

    Espaillat's loss ends the tenure of the first Dominican American member of Congress and the sitting chair of the Congressional Hispanic Caucus, whose district delivered him 83% of the vote just two years ago. His successor is a DSA-aligned first-time candidate who led campus protests and has called for abolishing ICE and ending military aid to Israel.

    Valdez's win extends the progressive hold on the 7th District, the seat held for over three decades by Velázquez, whose handpicked successor Reynoso was backed by labor and the Working Families Party.

    Mamdani, who was not on the ballot, demonstrated a degree of endorsement power that exceeded market expectations—particularly in NY-13, where his late-in-the-cycle endorsement of Chevalier was treated as a longshot by both prediction markets and the political establishment.

    The Accuracy Verdict

    Evaluated on its own terms, the NYC primary results present a clean picture of prediction markets working as designed in two races, and encountering the genuine difficulty of low-information upsets in a third:

    RaceKalshi FavoritePre-Election OddsActual WinnerVerdict
    NY-10: Lander vs. GoldmanLander98%Lander (65.8%)✅ Correct
    NY-7: Valdez vs. ReynosoValdez82%Valdez (56%)✅ Correct
    NY-13: Chevalier vs. EspaillatEspaillat67% (Chevalier 33%)Chevalier⚠️ Upset (priced as 1-in-3 risk)
    Sweep combo (all 3 win)—28%Happened⚠️ Lower-probability path materialized

    Two clean calls with high confidence. One genuine upset that the market correctly identified as a real risk—not a certainty, not a zero—and accurately priced at 33%.

    What traders should take from Tuesday night: in low-information primaries with insurgent energy and sparse polling, even the underdog's 33% deserves serious weight. It's not a coin flip, but it's not a lock either.

    FAQ

    Did Kalshi predict the NYC congressional primary correctly? Partially. Kalshi correctly predicted Brad Lander (98% pre-election) and Claire Valdez (82%) would win their races—both did. In NY-13, Kalshi gave Darializa Avila Chevalier a 33% individual chance, with the full sweep priced at 28%. That lower-probability sweep occurred. The market correctly communicated meaningful uncertainty; the outcome fell in the 28% scenario.

    What were the NYC primary results on June 24, 2026? All three Mamdani-backed candidates won: Brad Lander defeated incumbent Dan Goldman in NY-10 (65.8% to 34%), Claire Valdez defeated Antonio Reynoso in NY-7 (56% to 36%), and Darializa Avila Chevalier defeated five-term incumbent Adriano Espaillat in NY-13 (~49.4% to 45.9%), winning by approximately 2,200 votes.

    What was Kalshi's sweep odds for the NYC primaries? Kalshi's combo contract—tracking whether all three Mamdani-endorsed candidates would win—priced the sweep at 28% leading up to Election Day. The most likely scenario, per Kalshi, was that Lander and Valdez would win while Chevalier (the underdog) would lose, priced at 54%. The sweep outcome (28%) materialized.

    How did prediction markets cover the Mamdani endorsement story? Kalshi published pre-election analysis through its news platform tracking the individual race odds. CNBC cited Kalshi's market prices in pre-election coverage. The Chevalier upset in NY-13 became a case study in how markets handle low-information primaries where even 1-in-3 odds can resolve in the underdog's favor.

    Who is Darializa Avila Chevalier? Darializa Avila Chevalier is a community organizer and sociology Ph.D. candidate who helped lead the 2024 pro-Palestine protests at Columbia University. Endorsed by Mayor Zohran Mamdani and the DSA, she defeated five-term incumbent Adriano Espaillat—chair of the Congressional Hispanic Caucus—by approximately 2,200 votes in the June 2026 Democratic primary for New York's 13th Congressional District.

    Where can US users trade political prediction markets? U.S.-based Kalshi offers political event contracts to verified U.S. users. Polymarket's global platform (polymarket.com) covers political events but is not accessible to U.S. users via the QCX LLC designated contract market. U.S. political markets are available through Kalshi and Robinhood's prediction market product.

    Sources & Verification

    Back to Articles