Brazil presidential runoff 2026: first-round votes, latest polls and the ten-day contest

Brazil will hold its presidential runoff on 25 October after neither candidate secured more than half of valid votes on 4 October. The contest is between Senator Flávio Bolsonaro of the Liberal Party and incumbent President Luiz Inácio Lula da Silva of the Workers’ Party.

The official first-round count placed Bolsonaro first with 56,104,268 votes, or 47.03% of valid ballots. Lula received 53,876,617, or 45.16%. Bolsonaro’s official advantage was 2,227,651 votes and 1.87 percentage points. These two differences are Nine145 calculations from the TSE totals.

The election recorded 125,272,513 ballots in total. After subtracting 2,300,781 blank ballots and 3,674,149 null ballots, the valid-vote denominator was 119,297,583. This distinction is essential because Brazil’s presidential percentages are calculated from valid votes, excluding blank and null ballots.

Official first-round result table

Measure	Flávio Bolsonaro	Lula	Other/total
Valid votes	56,104,268	53,876,617	119,297,583 total valid votes
Valid-vote share	47.03%	45.16%	7.81% for other candidates
Vote difference	+2,227,651	—
Percentage-point difference	+1.87	—
Combined votes of finalists	109,980,885	—
Combined share of valid votes	92.19%	—	47.03% + 45.16%
Valid votes for other candidates	—	—	9,316,698, or 7.81%
Total ballots cast	—	—	125,272,513
Blank ballots	—	—	2,300,781 or 1.84% of ballots recorded
Null ballots	—	—	3,674,149 or 2.93% of ballots recorded

Runoff polling table

Polls using valid votes exclude blank, null and undecided responses. Polls reporting total voting intention retain at least some of those categories. Figures from the two formats must not be compared without conversion.

Pollster	Fieldwork	Sample	Reported measure	Bolsonaro	Lula	Margin/registration
AtlasIntel/Bloomberg	3–8 Oct	5,026	Valid votes	52.8%	47.2%	±1 point; BR-03663/2026
Datafolha	6–7 Oct	2,520	Total intention	49%	45%	±2 points; BR-02949/2026
Datafolha	6–7 Oct	2,520	Valid votes	52%	48%	Derived by pollster
BTG/Nexus	10–12 Oct	Not stated in cited summary	Valid votes	52%	48%	Technical tie reported; BR-07997/2026
PoderData/AYA	Published 13 Oct	Not stated in cited summary	Valid votes	51%	49%	95% confidence; BR-01209/2026
Quaest	10–13 Oct	2,154	Total intention	46%	44%	±2 points; BR-08729/2026
Quaest	10–13 Oct	2,154	Valid-vote equivalent	51%	49%	Rounded conversion reported by pollster/media

How to read the table

Four findings are visible without constructing an artificial poll average.

1. Bolsonaro led the official first round and leads numerically in each listed post-first-round survey.
2. Several leads fall within the poll’s stated margin of error and are consequently reported as technical ties.
3. The polls do not use identical methods: Datafolha interviewed respondents in person, AtlasIntel used digitally recruited online responses, and other institutes applied their own sampling and weighting systems.
4. The apparent spread changes when blank, null and undecided responses are removed. A 46–44 total-vote poll can become approximately 51–49 among valid votes without any respondent changing preference.

The Quaest snapshot

The newest fully documented poll available at the cut-off was Quaest, released on 14 October. It recorded Bolsonaro at 46%, Lula at 44%, blank/null or non-voting intention at 9%, and undecided voters at 1%. With a two-point margin of error, the two candidates were technically tied.

The poll interviewed 2,154 people from 10 to 13 October. A poll is a sample-based estimate of opinion during those dates—not a count of ballots that will be cast on 25 October.

The electorate outside the finalists

The first-round finalists together received 92.19% of valid votes. The remaining candidates received 9,316,698 votes, or 7.81% of the valid total. Separately, more than 33.4 million registered electors did not attend the first round. Brazil’s Senate information service has clarified that an elector who missed the first round may still vote in the second, provided the voter’s electoral status is regular.

The runoff is therefore not decided merely by retaining first-round supporters. The campaigns are competing for voters who backed eliminated candidates, cast blank or null ballots, abstained, or remain undecided.

First round versus latest poll

Candidate	Official first-round valid share	Latest Quaest valid equivalent	Direction, not forecast
Flávio Bolsonaro	47.03%	About 51%	Starts the runoff ahead in the official count and Quaest snapshot
Lula	45.16%	About 49%	Requires gains among non-finalist and returning voters

The two columns measure different things. The first is an official completed vote; the second is a sample estimate for a two-candidate contest. Their difference should not be described as a literal vote transfer.

What the data establish

Bolsonaro entered the runoff with a 2.23-million-vote first-round advantage and has retained a numerical lead across the principal published runoff surveys listed here. The size of that lead varies, and several surveys classify the contest as a technical tie.

The data do not establish the winner. Ten days remained – for – votes had not been cast for the second round, and differences in turnout can matter as much as persuasion in a close contest.