How Lula Won Brazil’s Runoff With 51.79%

BRAZIL ELECTIONS 2026 · RESULTS ANALYSIS

Updated 110 times · Latest: 4 November 2026

Lula won Brazil’s presidential runoff on Sunday 25 October 2026 with 51.79% of valid votes, the Superior Electoral Court (TSE) reported. He beat Senator Flávio Bolsonaro, who took 48.21%, by 4,430,191 votes.

This analysis explains how Lula won, region by region, and what changes in Brasília and in the markets. It is written to stay true for months, so later events are in our news coverage.

Explore the interactive results map above: tap any state to see its vote, turnout and the change since 2022. The numbers behind it are in our results data room, with a CSV download.

What Happened on 25 October

Brazil stayed up late on Sunday 25 October and woke up to a country split almost down the middle. Lula won 51.79% of valid votes, the Superior Electoral Court (TSE), the body that runs Brazil’s elections, reported after it counted all 499,248 polling sections.

Valid votes are the ones that count towards the percentage, so blank and spoiled ballots are left out. Lula received 64,256,232 votes and Flávio Bolsonaro, the 45-year-old son of former president Jair Bolsonaro, received 59,826,041.

Of 158,745,502 registered voters, 128,236,074 cast a ballot, a turnout of 80.78%. Another 1,430,496 voters left their ballot blank and 2,723,305 spoiled it, which together are 3.2% of ballots cast.

The day had one unusual twist. On the afternoon of the vote, Supreme Court Justice Cristiano Zanin ordered polling to stay open until 7 p.m. Brasília time, not 5 p.m., but only at stations still open.

The TSE’s president, Kassio Nunes Marques, told the regional electoral courts at 5:49 p.m. that only stations still open could keep voting. Flávio Bolsonaro later made the extension the centre of his complaint about the result, and the TSE’s president has said it did not change the result.

How Lula Won, Region by Region

Brazil votes in two very different countries, and the map shows it. In the Northeast, Lula won all nine states and took 24.45 million votes to 10.80 million, or 69.4% of the region’s valid votes.

That one region gave Lula a lead of 13.65 million votes, while everywhere else inside Brazil Flávio Bolsonaro led by 9.25 million. The gap between those two numbers, plus Lula’s net lead of 28,657 among votes cast abroad by our subtraction, is the national margin of 4,430,191.

Bahia alone made the difference. Lula beat Flávio there by 6,485,258 votes to 2,608,761, a lead of 3.88 million that is close to the whole national margin.

The seven states of the North (Acre, Amapá, Amazonas, Pará, Rondônia, Roraima and Tocantins) split almost exactly in half. Lula led there by just 1,079 votes out of 10.05 million, with 50.01% of the region’s valid votes.

The Southeast is the prize that decides Brazil, because it holds 41% of all valid votes. Flávio Bolsonaro won São Paulo with 53.85% and Rio de Janeiro with 53.10%, but Lula won Minas Gerais, the big swing state, by 50.43% to 49.57%, or 109,313 votes.

Minas Gerais is Brazil’s classic swing state, which is why campaigns spend so much time there. In the whole Southeast Lula took 47.24% of the valid votes, up from 45.73% in 2022 by our calculation from the TSE’s state files.

The South and the Centre-West, the farm belt around the capital, stayed conservative. Flávio won Santa Catarina with 69.42%, Paraná with 62.58%, Rio Grande do Sul with 57.84% and every state of the Centre-West, including the Federal District around Brasília.

The count produced one more lesson about Brazilian voters. They split their tickets, which means many people chose a conservative governor and Lula on the same day.

In São Paulo, Governor Tarcísio de Freitas of the conservative Republicanos party won re-election on 4 October with 62.65%. Lula still took 46.15% there in the runoff three weeks later.

In Minas Gerais, Cleitinho Azevedo, also of Republicanos, won the governor’s race with 55.40%. Lula carried the state anyway, by 109,313 votes.

Why the Runoff Turned: Three Numbers

Flávio Bolsonaro led the first round on 4 October with 47.03% against 45.16% for Lula, so Lula started the runoff 2.2 million votes behind. Three weeks later he finished 4.4 million votes ahead.

The first number is 9.3 million. That is how many voters picked someone else on 4 October, among them the writer Augusto Cury (Avante) with 2.89%, Renan Santos (Missão) with 2.24% and Ronaldo Caiado (Social Democratic Party) with 2.18%.

Between the rounds Lula added 10,376,694 votes and Flávio Bolsonaro added 3,721,538, so roughly three of every four extra votes went to Lula by our subtraction of the TSE’s two counts. The Southeast supplied 4.80 million of his gains, about 46% of the total, and Flávio Bolsonaro added only 14,110 votes in the whole North.

The second number is 1.86 points, the rise in turnout from 78.92% to 80.78%. The Northeast turned out at 83.5%, up from 81.6%, while the North was the only region where turnout fell, from 80.6% to 78.8%.

The third number is 4.8 million, the growth in valid votes from 119.3 million to 124.1 million. About 1.8 million fewer voters spoiled or left blank their ballot, and the final margin of 3.57 points came from that larger pool.

Nothing in the official data tells us how the third-placed voters or the returning abstainers split. No pollster has published a verified breakdown, so any claim about who decided the race is an estimate.

What the count does show is where the gains came from. Lula’s biggest rises over his 2022 runoff share came in the Federal District (+4.5 points), Amapá (+4.2), Acre (+3.7), Roraima (+3.6) and Rio de Janeiro (+3.4).

He lost ground in Tocantins (−2.7 points), Maranhão (−1.6) and Rio Grande do Sul (−1.5). Two states changed hands from 2022: Amapá went to Lula and Tocantins to Flávio Bolsonaro.

Trade also shaped the campaign. A 25% US tariff on a wide range of Brazilian goods, with long lists of exemptions, took effect on 22 July 2026 under Section 301, a US trade law.

Lula accused Flávio Bolsonaro of helping to provoke the tariff after a trip to Washington. The TSE does not measure motives, so we cannot say how much the tariffs moved votes.

We can say where Lula gained. He added 2,613,861 votes in São Paulo, 1,162,163 in Minas Gerais and 872,372 in Rio de Janeiro between the rounds, about 45% of his total gain.

How the Result Compares With 2022, 2018 and 2014

In the 2022 runoff Lula beat Jair Bolsonaro by 50.90% to 49.10%, a margin of 1.80 points, on 118.6 million valid votes. This time the margin doubled to 3.57 points on 124.1 million valid votes.

Lula’s 64.26 million votes are about four million more than the roughly 60.3 million he won in 2022. Flávio Bolsonaro’s 59.83 million are about 1.6 million above his father’s 58.2 million.

The 2026 margin is still modest by Brazilian standards. In 2018 Jair Bolsonaro won the runoff by 55.13% to 44.87%, a gap of more than ten points, and in 2014 the then president Dilma Rousseff beat the senator Aécio Neves by 51.64% to 48.36%.

Three of the last four runoffs, in 2014, 2022 and 2026, were decided by less than four points. In each of them the Northeast stood on one side and the South and the farm belt on the other.

The state map changed little. Lula won 13 states in 2022 and 13 in 2026, and Lula’s share moved by less than one point in 11 of the 27 voting areas.

The first-round comparison is more telling. Flávio Bolsonaro won the first round with 47.03% and still lost three weeks later, because 9.3 million voters had backed other candidates and the pool grew in the runoff.

What Changes in Brasília

Lula is sworn in on Tuesday 5 January 2027 for a four-year term that runs to January 2031. The date moved from 1 January because of Constitutional Amendment 111 of 2021, and new governors take office the next day.

The courts must certify the winners in a ceremony called the diplomação by Friday 18 December, the deadline in the TSE’s 2026 calendar. The new Congress takes its seats on Monday 1 February 2027 and picks the leaders of both houses.

Lula starts his new term without a majority. The Chamber of Deputies has 513 seats, and Flávio Bolsonaro’s Liberal Party (PL) is the largest party with 121, or 23.6%.

Lula’s Workers’ Party (PT) has 70 seats, Brazil Union, a centre-right party, 46 and the Social Democratic Party (PSD) 43. No party is close to a majority, so every law needs a deal.

In the Senate, voters filled 54 of the 81 seats on 4 October and the PL won 19 of them. With its senators already in office, the party is projected to hold about 28 seats from 2027, which would make it the largest party in the Senate, by press counts of senators in office.

That arithmetic shapes everything. Ordinary bills need the centre and centre-right, such as the PSD, the Brazilian Democratic Movement (MDB) and Progressistas (PP), and each will want ministries and budget money.

The first test is the 2027 federal budget, which Congress must pass by the end of the year. The second is the cabinet, above all the finance ministry, which Dario Durigan has run since March 2026.

Durigan took over when Fernando Haddad, the previous finance minister, left the post to run for governor of São Paulo. Haddad lost that race with 36.42% against 62.65% for Tarcísio.

The fourth term is also the last in a row. The Constitution bars a president from a third consecutive term, so 2030 will be an open race and Lula, who turns 85 that year, cannot be on the ballot.

That is why traders already price the 2030 winner, a market we show below. Brazil’s next president after Lula will be chosen in an open race.

Did Anyone Challenge the Result?

Yes, but with little legal force so far. On election night Flávio Bolsonaro said he could not recognise the result and would ask for the election to be annulled, pointing to the two-hour extension of voting.

By the next day he had dropped the annulment plan. His camp said it would instead sue for abuse of political and economic power, a case that could in theory remove Lula and his running mate, Geraldo Alckmin.

On 28 October he published a letter calling himself the president-elect, then changed it 22 minutes later to read “leader of Brazil”. The TSE has confirmed Lula as the winner, and the PL recognised his victory on 26 October.

Outside Brazil, US President Donald Trump said on 26 October that the runoff “looked like another rigged election”, without giving evidence. Alckmin called the remark unfounded.

The TSE’s president has said the extension did not change the result. Whether the courts will hear the complaint, and on what grounds, is not yet known.

Traders on Polymarket, a prediction exchange, see little chance of it. On 4 November they gave about 3% to the presidential election being annulled by 31 December.

What It Means for Markets, Investors and Readers

The Ibovespa, the main stock index of the São Paulo exchange, fell 13.88% on Monday 26 October, from 205,082.56 to 176,608.23. It closed at 165,528.92 on 3 November, 19.3% below its level before the vote.

The real fell with it, as the central bank’s official dollar rate (PTAX) rose from R$5.0377 per US dollar on 23 October to R$5.5007 on 26 October, a jump of 9.2%. It stood at R$5.4935 on 3 November.

The central bank’s policy committee, Copom, decides on the Selic, the benchmark interest rate, this evening, 4 November, Brasília time; the exact release minute is set by the central bank. The Selic stood at 13.75% on 4 November, and the next decision comes on Wednesday 9 December.

Other dates from the calendar of RT market data, our market-data provider, will show how markets read the new government. The IPCA consumer-price index for October is due on 12 November, the budget balance for October on 30 November and third-quarter GDP on 2 December.

For US investors, the result means continuity in policy but more fiscal risk than before. A Chamber led by the PL favours slow deals with the centre over fast reforms, so the real and Brazil’s bonds will trade on budget headlines.

Brazilian companies such as Petrobras, the state-controlled oil company, and Vale, the mining group, also trade in New York as American depositary receipts, certificates that stand for shares held abroad. Their dollar price moves with both the share and the real, so a weaker real hurts US holders twice.

On trade, the 25% Section 301 tariff on Brazilian goods that are not exempt remained in force when we last checked on 4 November, and we found no announced change since the vote. Any change would need Washington and Brasília to talk, and Lula asked Trump in a call on 21 August to resume tariff talks.

For travellers and expats, an election result does not change entry rules or visas by itself. Check the Brazilian consulate and the US State Department before you book.

In the wider region, Brazil is Latin America’s largest economy, so the result is felt in Mercosur, the trade bloc it founded with Argentina, Paraguay and Uruguay. Lula’s win keeps the country’s foreign and trade policy with the same government for another four years.

What Prediction Markets Say

Prediction markets, where traders bet real money on outcomes, have moved on from the runoff to what comes next. On Kalshi, an exchange regulated in the US by the Commodity Futures Trading Commission (CFTC), the market on which party wins the 2030 presidency last traded at PT 30%, PL 22% and the conservative Republicanos 20%; the latest trade was on 27 October, and the page was read on 4 November at 3:11 p.m. ET.

The Social Democratic Party (PSD) stood at 5%, the Brazilian Socialist Party (PSB) at 3% and Missão at 2%. Only 7,787 contracts had traded, so treat that market as a mood, not a forecast.

On Polymarket, a crypto-based prediction exchange, the market on the central bank’s November decision gave about 78% to no change in the Selic and about 22% to a quarter-point cut, with US$186,859 traded; that is a clear shift towards a cut since this morning, when no change stood at about 96%. A market on whether the presidential election is annulled by 31 December gave about 3%, with US$53,589 traded.

These prices are bets, not polls, and another market can show a different number. Why we show this: prediction markets turn real-money bets into a live probability that moves within minutes of the news, which is why investors, campaigns and newsrooms in the United States now follow them closely. We show them next to polls and official results, never instead of them.

Three Scenarios to May 2027

None of these is a forecast. Each is a path with the conditions that would make it likely, and we will judge them against the facts that come in before May 2027.

Scenario 1: A Workable Deal With the Centre

Lula builds a coalition with the PSD, the MDB and other centre parties by 1 February and passes the 2027 budget with a credible deficit target. The conditions are a cabinet that gives those parties real posts, and a finance minister that investors trust.

In this case the real stabilises near R$5.4 to R$5.5 per dollar, the Selic starts to fall slowly after 13.75%, and the Ibovespa recovers part of its drop. This is the path that needs the fewest things to go wrong, but it depends on choices not yet made.

Scenario 2: Gridlock and Fiscal Stress

The PL, with 121 deputies and 28 senators, and an angry centre block the budget or demand spending Lula cannot afford. The conditions are a cabinet picked mainly from the PT, a fight over the leadership of the two houses on 1 February, or a bad inflation reading on 12 November.

Here the real weakens beyond R$5.5, the central bank keeps the Selic at 13.75% or higher, and Brazil’s dollar bonds lose value. Lula would still govern, but with fewer laws passed.

Scenario 3: Legal and Street Conflict

Flávio Bolsonaro’s court challenge grows, or his supporters take to the streets. The conditions are a formal filing at the TSE and a court that agrees to hear it, neither of which we have confirmed.

Courts and the electoral authority have so far backed the count, and traders price annulment at about 3%. If this path opened, the main effect would be on markets and on Brazil’s image abroad, not on who holds office.

What We Do Not Know

We do not know whether Flávio Bolsonaro has formally filed any suit with the TSE, because we could not confirm one from the court’s own records. We also do not know how the voters of the third-placed candidates split in the runoff.

No official date has been confirmed for the diplomação, only the 18 December deadline. Lula has not yet announced his cabinet, including his finance minister.

We have not seen any change to the US tariff since the vote. We also do not know whether the tariff talks that Lula asked for before the election will restart.

Before the Vote: A Short Look Back

The final polls pointed the wrong way. Datafolha and Quaest, two Brazilian polling firms, each had Flávio Bolsonaro ahead by 51% to 49% of valid votes on 24 October, and the count gave Lula 51.79%.

Both firms missed in the same direction, by about 5.6 points on the gap between the two men. No cause has been confirmed, and our poll tracker keeps every published survey for anyone who wants to test it.

The lesson for investors is to read polls and prices as ranges, not as forecasts. Our poll tracker and the prediction-market pages keep the numbers as they stood before the vote.

Keep Reading: The Results Pages and the Data

• Final result of the runoff, with the count by state.
• Brazil election data room, with polls and embeddable cards for other sites.
• Results data room and CSV behind the map above.
• First round, 4 October, and how the polls compared.
• Lula starts building his next cabinet, and the market report for 26 October.
• The new Congress, state by state, and all Brazil Elections 2026 coverage.

Frequently Asked Questions

Who won Brazil’s 2026 presidential election?

Lula won the runoff on Sunday 25 October 2026 with 51.79% of valid votes, or 64,256,232 votes. Flávio Bolsonaro took 48.21%, or 59,826,041 votes, according to the TSE.

How wide was Lula’s margin?

Lula won by 4,430,191 votes, or 3.57 percentage points. In the 2022 runoff his margin over Jair Bolsonaro was 1.80 points.

Which states did Lula win?

Lula won 13 states: the nine of the Northeast plus Amazonas, Amapá, Pará and Minas Gerais. Flávio Bolsonaro won 13 states and the Federal District, including São Paulo and Rio de Janeiro.

Did the polls get the result right?

No. Datafolha and Quaest each had Flávio Bolsonaro ahead by 51% to 49% on 24 October, and Lula won 51.79% to 48.21%.

What date was set for Lula’s inauguration?

Lula’s inauguration is set for Tuesday 5 January 2027, after the courts certify the winners by Friday 18 December. Governors take office on 6 January.

Did Flávio Bolsonaro accept the result?

No, not fully. He said on election night that he would seek annulment, dropped that plan the next day and moved to a suit alleging abuse of power, while his Liberal Party recognised Lula’s win on 26 October.

How did Brazil’s markets react to the result?

The Ibovespa fell 13.88% on 26 October and the dollar rose from R$5.0377 to R$5.5007 on the central bank’s official rate. By 3 November the index was 19.3% below its level before the vote.

Can Lula run again in 2030?

No. The Constitution bars a president from a third consecutive term, so the 2030 election is open.

Sources: Superior Electoral Court (TSE) results portal, president files for the first round (4 October) and the runoff (25 October 2026), and governor files, read 4 November 2026; TSE open-data archive for the 2022 runoff by state; TSE election calendar 2026 (Resolution 23.760); Constitution of Brazil and Constitutional Amendment 111 of 2021; Federal Senate and Chamber of Deputies seat data; Banco Central do Brasil (central bank) Selic target and PTAX dollar rate, read 4 November 2026; RT market data (Ibovespa) and economic calendar; Office of the US Trade Representative, Section 301 action on Brazil (Federal Register, July 2026); Datafolha and Quaest final runoff polls of 24 October; Polymarket and Kalshi public market data, 4 November 2026, 3:11 p.m. ET; statements of the candidates, the Supreme Court and the TSE. As of 4 November 2026.

More: Brazil news in English, every day from The Rio Times.