BTG/Nexus: Lula leads in the 1st round and technically ties with Flávio in the 2nd

According to a survey released this Monday (25), the president appears with 47% of voting intentions, against 43% for the senator

PR and Senado Agency
In the other scenarios tested by the research, without the presence of Flávio, President Lula maintained the leadership of voting intentions in all second round simulations presented by the survey.

The third survey of Nexus Research and Intelligencein partnership with the BTG Pactualreleased this Monday (25), points out President Lula (PT) numerically ahead of Senator Flávio Bolsonaro (PL) in a possible second round of this year’s presidential election.

According to the survey, Lula appears with 47% of voting intentions, against 43% from Flávio Bolsonaro — a numerical difference of four percentage points between the candidates. According to the survey’s margin of error, which is plus or minus two percentage points, the two are technically tied.

Others 9% said they voted for blank, null or none of the candidatese 1% did not know or chose not to respond.

In the other scenarios tested by the research, without the presence of Flávio, the President Lula maintained the lead in voting intentions in all simulations second round presented by the survey.

BTG/Nexus: Lula appears ahead in the 1st round; dispute with Flávio is tied for 2nd

BTG/Nexus.

Lula leads in 1st round

In the first round stimulated scenario, President Lula leads with 40% of voting intentionsfollowed by Flávio Bolsonaro, with 35%.

Then they appear Ronaldo Caiado (PSD), with 5%, Romeu Zema (Novo), with 4%, Renan Santos (Missão), with 3%, Joaquim Barbosa, with 2%e Augusto Cury (Avante) and Cabo Daciolo (Mobiliza), both with 1%.

Another 7% said they voted blank, null or for none of the candidates, while 2% did not know or preferred not to answer.

BTG/Nexus: Lula appears ahead in the 1st round; dispute with Flávio is tied for 2nd

BTG/Nexus.

Methodology

The quantitative research was carried out by Nexus by telephone, using the CATI (Computer Assisted Telephone Interview) methodology, between May 22 and 24, 2026. In total, 2,045 voters were interviewed. The margin of error is two percentage points, plus or minus, with a 95% confidence level.

The survey is registered in the Superior Electoral Court (TSE) under number BR-04193/2026.

source