Brazil Presidential Election 2026

Polymarket's Brazil presidential book turned over $922,823 in the last 24 hours, second only to the FedEx St. Jude golf market across every exchange this platform covers. First round is October 4; a runoff, if needed, is October 25.

The prices are settled and cross-exchange agreement is near-total. Lula 66.5% on Polymarket against 64.5% on Kalshi. Flávio Bolsonaro 28.7% against 28.5%. Renan Santos 4.7% against 4.9%. Three candidates, two exchanges, and the widest disagreement anywhere in the book is two points.

The money is somewhere else entirely. The deepest leg of the day is Romeu Zema at 0.2%, taking $294,555—more than three times what Lula's 66.5% leg draws. Second-deepest is Ronaldo Caiado at 0.6% on $192,486. Add Camilo Santana at 0.1% on $50,699 and you have $537,740—58% of the entire book—on three candidates the same book collectively prices under one percent.

And the depth is on one exchange. Polymarket does $922,823; Kalshi does $11,696. That is a 79-to-1 ratio on identical prices—the most lopsided cross-exchange volume split in this catalog. Kalshi's book is a mirror, not a market.

Election calendar.

Four venues, one question. Polymarket carries the flow, Kalshi carries near-identical prices on 1.3% of the volume, Manifold runs $344 a day, and PredictIt's book has gone quiet at zero.

Kalshi's open interest tells a different story than its volume: 408,077 contracts on Renan Santos at 4.9%, 268,610 on Flávio Bolsonaro, 248,771 on Lula. Positions were built there; they are simply not being traded now. The most-owned leg on Kalshi is the third-place candidate.

Brazil's two-round system makes second place the consequential question, and Polymarket prices it at $36,336 a day—triple the volume of Kalshi's entire outright book.

Flávio Bolsonaro is 85.5% to finish second on $19,781. The surprise is what sits behind him: Lula himself at 6.9% on $12,658, which is the market's implied probability that Bolsonaro tops the first round outright and Lula trails into the runoff. That leg is the second-deepest in the book.

Then the pattern from the outright market repeats one tier down. Camilo Santana at 0.5% takes $15,109, third-deepest in a book with an 85.5% favorite; Michelle Bolsonaro at 0.2% takes $9,380; Fernando Haddad at 0.2% takes $9,046. Whoever is buying the tail in the winner book is buying the same names for second place.

Kalshi's mirrors of both books trade $102 and .

Brazil requires more than 50% to avoid a runoff. Kalshi prices anyone winning outright on October 4 at 24.5% on $1,924—roughly one-in-four that the election is over in a single round.

Polymarket's margin ladder is the finer read, and it is unambiguous about who leads: Lula by 5–10% at 40.5%, by 10–15% at 26.9%, under 5% at 19.5%, 15%+ at 5.0%. That sums to 91.9% that Lula tops the first round—against 66.5% that he ultimately wins. The twenty-five-point gap between those two numbers is the runoff risk, priced explicitly.

Flávio's side of the ladder totals 14.3%, which does not reconcile cleanly with 91.9% for Lula. On a book doing $5,657 a day, treat the margin distribution as indicative rather than arbitrage-free.

A note on the Kalshi metadata: several Kalshi Brazil contracts report settlement in October 2027, a year after the election they settle. The tickers are right and the prices are live; the dates on the DTOs are not. The same off-by-a-year defect appeared on Kalshi's Clacton series last week.

• Lula vs. Tarcísio—Polymarket leads on volume. Polymarket's outright Categorical is the price the global book is converging on. When Kalshi or Manifold drift more than 5 points on either name, one book is catching up.
• Outright Single as runoff probability. Kalshi's KXBRAZILPRES1R-26OCT04 binary asks whether the first round ends without a runoff. As of recent reading the implied probability of an October 25 runoff sits well above 50%; any drift toward the outright-Single Yes side is a sign the field has narrowed unexpectedly.
• 2nd-place spread. The 2nd-place Categorical is where the Bolsonaro-vs.-Tarcísio internal fight prices directly. Polymarket lists the family members as separate lines; Kalshi often groups them. Read the two charts together to disentangle which face of the right is mobilising.
• PredictIt's smaller field. PredictIt only books three candidates (Lula, Flávio Bolsonaro, Zema). It is the most retail-American venue and the cleanest read on what U.S.-side traders without ground knowledge of Brazil are pricing—useful as a contrast to the sharper Polymarket book.
• The countdown targets the October 4 first-round vote. Expect the largest re-pricing in the final two weeks as candidate field is locked and last debates take place.