Datafolha SP: Flávio Bolsonaro and Lula Tie

The scenario of the presidential dispute between the voters of São Paulo shows Flávio Bolsonaro (PL) and Lula (PT) technically tied in the first round, according to Datafolha poll released on Friday (2). The survey, commissioned by Globo and Folha de S.Paulo, points to the candidate of the PL with 39% of the voting intentions, against 38% of the petista. The difference of one percentage point is within the margin of error, which is two points for more or less.

In addition to the two leaders, other names appear in the sequence, albeit with much lower percentages. Cury registers 5%, while Caiado and Renan appear with 4% each. Zema has 1% of the mentions. The survey also brings the information that Leonardo Avalanche (PRTB) had not withdrawn from the candidacy on the day of the survey registration.

The figures reflect a portrait of the political moment in the country’s largest electoral college, but should not be read as a definitive projection. The margin of error and the campaign context are factors that can change the framework until the election.

Technical tie dominates the paulista dispute

With 39% for Flávio Bolsonaro and 38% for Lula, the scenario is a technical tie, since the difference of one percentage point is less than the margin of error of two points. This means that, statistically, both candidates may have the same percentage of voting intention among São Paulo voters.

Technical tie is an important concept in electoral polls. It indicates that it is not possible to say with certainty which candidate is ahead, because the sample variation can reverse the order. In the case of São Paulo, both Flávio Bolsonaro and Lula could actually have up to 41% or 37% of the intentions, considering the limits of the margin of error.

This harsh scenario tends to intensify campaign strategies on both sides, with a focus on São Paulo, a state that concentrates the largest number of voters in Brazil and is often decisive in national disputes.

Methodology and scope of research

Datafolha interviewed 1,610 people between September 28 and 30. The margin of error is two percentage points, for more or less, and the level of confidence is 95%. The research registration number is BR-02676/2026.

The 95% confidence level means that if the survey was repeated 100 times with the same methodology, in 95 of them the results would be within the margin of error. It is a common standard in election surveys and gives credibility to the numbers presented.

The sample of 1,610 people is considered robust for a state cutout, allowing for analysis with good accuracy. The three-day collection also helps to capture a specific moment of the campaign, without large temporal variations.

Valid votes and no undecided scenario

Datafolha also released the valid votes. To reach that number, the institute excludes blank, null votes and undecided voters. This calculation is the same used by the Justiça Eleitoral on election day and is usually the closest to the final result.

Although the percentages of valid votes have not been detailed in the disclosure, the practice of Datafolha is to present this cut to give an idea of what the contest would look like if only the votes considered valid were counted. This helps to understand the potential of each candidate in a possible second round.

The exclusion of whites, null and undecided can alter the order of the candidates, but in the São Paulo scenario, with the isolated leadership of Flávio Bolsonaro and Lula, the tendency is that both remain in front, with the other candidates vying for a smaller slice of the electorate.

What to expect from the final line

With the technical tie between the two main candidates, the campaign in São Paulo is expected to intensify in the coming weeks. The dispute over the undecided electorate and the votes of the candidates with the least intention of voting will be crucial to define the result.

In addition, the presence of Leonardo Avalanche (PRTB) in the dispute, even without having formally given up until the registration of the research, can influence the scenario, even if its percentage has not been disclosed. The fragmentation of votes among candidates of lesser expression may be a factor to be noted.

For the São Paulo voter, the time is to follow the proposals and the performance of the candidates in the debates and in electoral propaganda. The Datafolha poll serves as a portrait of the moment, but the vote will only be defined in the ballot box.

The Boca no Trombone Portal will continue to track the unfolding of the presidential race and will bring new updates as more data is released.

Read Also

Elections 2026: aplicativo Pardal allows to denounce irregular electoral propaganda

STF Creates Filter to Bar Orders to Specific Minister

TRE-PR defines this Saturday the 35 ballot boxes that will be audited in the elections

Font

g1.globo.com