Brazil’s presidential runoff is a two-candidate election, but its decisive voters may be people who did not vote for either finalist in the first round. Under Brazil’s electoral system, if no presidential candidate wins more than half of valid votes in the first round, the two leading candidates advance to a second election. First-round votes do not carry over: voters make a fresh choice, and the candidate with more valid votes in the runoff wins.
That makes the voters behind eliminated candidates unusually important. In the 2026 contest, polling before the first round showed Lula and Flávio Bolsonaro with highly committed core supporters, while candidates such as Augusto Cury, Renan Santos, Ronaldo Caiado and Romeu Zema were competing for smaller constituencies. Datafolha found that voters outside the traditional Lula-versus-Bolsonaro camps were considerably more open to changing their choice, making them a potentially important source of runoff votes.
How Brazil’s presidential runoff works
Brazil uses a majority system for the presidency.
To win in the first round, a candidate needs more than half of the valid votes. Blank and null ballots are excluded from the valid-vote calculation. If nobody reaches that threshold, the two candidates with the most valid votes proceed to the runoff.
The second round is therefore not a redistribution of first-round votes by formula. Every voter gets another opportunity to cast a ballot.
That distinction is important because a first-round candidate who wins 45% does not enter the runoff with 45% already secured. Those voters have to turn out again and vote for that candidate—or potentially abstain, cast a blank or null ballot, or change their choice.
For the 2026 presidential election, the scheduled runoff is October 25 if one is required.
| Voter group or factor | Why it matters in a runoff | What to watch |
|---|---|---|
| Supporters of eliminated candidates | They represent votes that can move between the finalists or toward abstention/blank-null voting | Endorsements and second-choice polling |
| Non-aligned voters | They tend to be less attached to either major political camp | Whether campaigns can persuade rather than simply mobilize |
| Core supporters | Their turnout determines how much of each candidate’s first-round support survives | Enthusiasm, organization and participation |
| Lower-income voters | A major constituency for Lula in recent polling | Turnout and economic concerns |
| Higher-income voters | A stronger constituency for Flávio Bolsonaro | Economic and security issues |
| Northeast | Historically and currently a major Lula stronghold | Whether the margin changes |
| South and parts of Southeast | Stronger territory for Bolsonaro-aligned politics | Whether Flávio expands his regional advantage |
| White/null and abstaining voters | They can affect the effective size of the winning coalition | Whether disengagement rises or falls |
The third-party voters are the immediate prize
The most obvious pool of potential swing voters is made up of people whose first-round candidate is eliminated.
That does not mean all of them will switch to one of the finalists.
Some may have a second preference. Others may dislike both remaining candidates. Some may decide that the stakes are high enough to choose strategically, while others may stay home or vote blank or null.
Datafolha’s final pre-first-round survey illustrates the complexity. Among voters not aligned with either petismo—the political current associated with Lula—or bolsonarismo, Lula and Flávio were almost evenly matched in first-round intentions. Among voters who supported smaller candidates, however, the second-round choices varied significantly.
For example, the survey found that among voters intending to support Augusto Cury in the first round, 44% said they would choose Flávio in a Lula-Flávio runoff, 30% Lula and 24% blank/null. Among Ronaldo Caiado’s supporters, the corresponding figures were 46%, 30% and 22%. Renan Santos’s voters leaned more strongly toward Flávio, at 54% versus 28% for Lula, with 16% choosing blank or null.
Those figures are polling preferences, not actual vote transfers. They also come from before the first-round result, meaning the political environment can change once voters know exactly which candidates have been eliminated.
Why endorsements matter—but cannot simply transfer votes
After the first round, defeated candidates and their parties face a strategic choice.
They can endorse one finalist, remain neutral, urge supporters to vote against a particular candidate, or effectively leave the decision to individual voters.
An endorsement can provide organizational support, campaign infrastructure and political legitimacy. But it does not give a candidate ownership of another politician’s voters.
A voter who supported a centrist or smaller-party candidate may have chosen that person precisely because they rejected both major camps. An endorsement from the eliminated candidate therefore may influence the voter without determining the vote.
This is why second-choice polling can be more informative than endorsements alone.
The question is not simply, “Who did the defeated candidate endorse?” It is “Where do that candidate’s voters actually say they will go?”
The voters who are already committed
The runoff will not be decided exclusively by swing voters.
Both Lula and Flávio Bolsonaro entered the campaign with large and highly motivated political bases. Datafolha’s final pre-first-round survey found that 88% of Lula supporters and 87% of Flávio supporters said they were certain of their choice.
That creates a different strategic problem from winning over undecided voters.
The campaigns must ensure that their supporters actually participate.
A candidate can theoretically gain ground in polling while losing the election if supporters fail to turn out in sufficient numbers. Conversely, a highly motivated base can compensate for weaker persuasion among undecided voters.
This makes turnout particularly important in a close runoff.
Turnout could change the arithmetic
Brazil has compulsory voting rules for many citizens, but abstention remains significant. Non-participation has been a recurring feature of recent presidential elections, and reporting before the 2026 vote suggested that roughly one in five voters could abstain.
The crucial issue is not simply whether turnout rises or falls nationally.
It is whose turnout changes.
If supporters of one candidate are more enthusiastic and organized than those of the other, a modest difference in participation can affect the final margin. The same applies geographically: a small turnout shift in a large electoral region can matter more than a much larger percentage change in a small state.
This is one reason campaign strategies in the final weeks can emphasize mobilization rather than persuasion.
Brazil’s regional blocs still matter
Brazil is too large and politically diverse for the national percentage alone to explain the election.
The Northeast has long been a central base of support for Lula and the Workers’ Party. The South has generally been much more favorable to Bolsonaro-aligned candidates, while the Southeast—Brazil’s largest population center—can be highly consequential because it contains enormous numbers of voters and includes politically diverse states.
The latest Datafolha survey before the first round showed the regional divide clearly: Lula held a substantial advantage in the Northeast, while Flávio led in the South. The two were effectively tied in the Southeast in that survey.
This creates an important runoff question: does either candidate broaden the geographic coalition, or does the second round simply intensify the first-round map?
A candidate does not necessarily need to win every region. What matters is how efficiently votes are accumulated across Brazil’s states.
The Northeast: Lula’s defensive wall
For Lula, the Northeast is more than another region. It has been a crucial component of his electoral coalition.
A large margin there can provide a substantial national vote advantage even if the candidate performs less strongly elsewhere. The latest Datafolha survey put Lula at 56% among first-round voters in the Northeast compared with 26% for Flávio.
The runoff question is whether Lula can preserve that advantage while preventing Flávio from making sufficient gains elsewhere.
For Flávio, narrowing the deficit in the Northeast would help, but the more realistic route to a national majority may be maximizing support in regions where Bolsonaro-aligned politics is already stronger and competing aggressively for voters in the Southeast.
The South: Flávio’s strongest territory
The South presents almost the reverse calculation.
Datafolha’s final pre-first-round survey showed Flávio at 50% among voters in the South, compared with 30% for Lula.
A strong performance there gives Flávio an important base, but it does not by itself guarantee a national victory.
The strategic challenge is to convert that regional strength into additional votes in the much larger and more politically mixed Southeast and elsewhere.
The Southeast could be the pivotal battleground
The Southeast contains São Paulo, Minas Gerais, Rio de Janeiro and Espírito Santo and represents a huge share of Brazil’s electorate.
Its political diversity makes it particularly important in a close presidential contest.
The latest Datafolha survey had Lula and Flávio effectively tied in the region, with 40% and 39%, respectively, in the first-round scenario.
That means small movements can have national consequences.
A candidate who improves by several points in a large Southeast state can potentially gain far more votes than a much larger percentage movement in a smaller electorate. This is why national polling should always be read alongside regional results.
Income, religion and education create another map
Regional geography is only one way of understanding the runoff electorate.
The same Datafolha survey found Lula substantially stronger among lower-income and less-educated voters, while Flávio performed better among higher-income groups. Lula also led among Catholics, while Flávio had a substantial advantage among evangelicals.
These divisions overlap with geography but do not perfectly correspond to it.
That matters because campaigns can target voters around issues rather than simply around state borders. Economic conditions, public security, religion, corruption, poverty and inequality can motivate voters across regional boundaries.
Polling reported before the election indicated that Flávio had advantages on issues including inflation, corruption and crime, while Lula was viewed more favorably on poverty and inequality, foreign affairs and health.
The runoff therefore becomes partly a contest over which issue voters consider most important when choosing between the two finalists.
The voters who dislike both candidates
Perhaps the most difficult group for either campaign is not the undecided voter who is waiting for more information.
It is the voter who has already decided that neither finalist is attractive.
That group can choose one of three broad paths: reluctantly support one candidate, vote blank or null, or not participate.
This matters because the winning candidate needs more votes, not necessarily more enthusiasm in absolute terms.
Datafolha’s first-round polling showed both Lula and Flávio with very high rejection levels: 45% of respondents said they would not vote for either candidate under any circumstances.
In a polarized runoff, campaigns therefore face a difficult balancing act. They need to motivate their own supporters without making themselves so unacceptable to the other side that potential swing voters consolidate against them.
What the first-round result will reveal
The actual first-round result will make the runoff math clearer than any pre-election poll.
Three numbers will be especially important.
First, the gap between Lula and Flávio. A close result means both campaigns start the runoff with plausible paths to victory. A larger gap does not make the result predetermined, but it changes the amount of persuasion and turnout movement required.
Second, the size and distribution of the eliminated candidates’ vote. A third-place candidate with a large national vote share can create a significant pool of potential runoff voters. But their geographical distribution matters just as much as their national total.
Third, turnout. Analysts will need to compare participation with previous elections and examine whether abstention was unusually concentrated in particular regions or demographic groups.
Only after those figures are known can the campaigns calculate where their most realistic gains are.
Why the runoff polls should be read carefully
A close runoff poll does not mean the election is necessarily a coin toss.
Polling has a margin of error, and a national result can depend on turnout assumptions, undecided voters and changes in preferences during the final campaign.
The final Datafolha survey before the first round showed Lula at 48% and Flávio at 45% in a hypothetical runoff, with 6% choosing blank/null and 1% undecided. The three-point lead was within the poll’s two-point margin of error and was therefore classified as a technical tie.
The important lesson is methodological: a runoff poll describes voter intentions at a particular moment; it does not convert those intentions into a guaranteed electoral outcome.
The actual second-round electorate can differ from the first-round electorate because some voters who supported eliminated candidates may behave differently from committed supporters of the two finalists.
What to watch between the two rounds
The most informative developments will be less about campaign rallies than about changes in the underlying coalition.
Watch for:
- Endorsements from eliminated candidates and major political parties, particularly where those endorsements come with organized regional networks.
- Second-choice polling among third-party voters, rather than simply national headline numbers.
- Turnout intentions, especially among voters who are less enthusiastic about both finalists.
- Regional movement in the Southeast, where a small shift can translate into a large number of votes.
- Whether Lula maintains his Northeast advantage.
- Whether Flávio preserves his strength in the South while expanding in the Southeast.
- Changes among lower-income, evangelical, Catholic and younger voters.
- Whether the campaigns moderate or intensify their messages in an attempt to win voters who rejected them in the first round.
These indicators are more useful than treating every campaign statement or daily poll as evidence of a decisive shift.
The voters who could actually decide the presidency
The most important runoff voters are unlikely to belong to a single demographic category.
They are more likely to be a combination of supporters of eliminated candidates, politically non-aligned voters, people who dislike both finalists but are willing to choose strategically, and committed supporters whose participation is uncertain.
The regional map then determines where those voters matter most.
That is why the runoff should not be understood simply as “Lula versus Flávio.” It is a contest to assemble a new majority from the electorate revealed by the first round.
The decisive question after October 4 will therefore be narrower and more practical: which candidate can convert the first round’s fragmented electorate into the broader coalition required to win a majority on October 25?
The answer will depend less on who wins the argument over the first-round result than on who can persuade—or mobilize—the voters who were never firmly in either camp to begin with.