Datafolha’s final survey before Brazil’s first-round vote places President Luiz Inácio Lula da Silva ahead of Flávio Bolsonaro, but neither side presents the result as a decisive advantage. The poll shows Lula on 45% of valid votes and Bolsonaro on 42%—a three-point gap that is described as a technical tie.¹

The left-leaning account stresses both Lula’s numerical lead and the speed with which Bolsonaro has closed ground. Lula remained at 45% from the previous Datafolha release, while Bolsonaro rose from 40% to 42%, according to the report. It also notes that valid-vote calculations exclude blank, null and undecided respondents, and that a first-round victory requires more than 50%.

The right-leaning coverage reaches the same central conclusion but places greater weight on the uncertainty around the figures. It reports a two-percentage-point margin of error at a 95% confidence level, meaning the first-round result falls within that range. In its simulated runoff, Lula holds an even slimmer 47% to 46% lead over Bolsonaro, with 6% blank or null and 2% undecided.²

Both accounts therefore distinguish a poll snapshot from an election forecast. The left-leaning report points to abstention, late shifts among voters and turnout differences as factors that could alter the outcome. The right-leaning article similarly cautions that survey methods and question design can affect results, while arguing that polls are informative rather than predictive.

A post by commentator Rodrigo Constantino highlights the contrary case for treating an eve-of-election lead as meaningful, saying that in six previous contests “the Datafolha leader on the eve was the most voted in the first round.”Rodrigo Constantino (Rconstantino)³ That historical observation does not resolve the current contest, but it underscores the tension in the final numbers: Lula leads, while Bolsonaro remains close enough for the margin of error to dominate the last-day debate.