Brazil - Comments on the first round results
Lula remains the strong favorite to win the election. Political polarization, the strength of the ideological right, and Bolsonaro’s fraud claims will make governing hard.
Lula remains the strong favorite to win Brazil’s election
We wrote last Wednesday that Lula was over 90% likely to win the election. Nothing about the results this past weekend convinced me otherwise.
For all the hype about polling misses, the aggregation of polls across this entire election has been relatively accurate. The polling aggregator we referenced in last Wednesday’s newsletter had Lula 43, Bolsonaro 36 (46-39 with undecideds removed). That’s a seven point lead for Lula. The final aggregation was 44-38 after the late polls were included, a six point margin. The final split for the first round was five points (48-43).
While the national aggregation of polls was essentially within the margin of error, two factors contributed to the perception that the polls were off.
First, some late numbers from pollsters who have historically done well suggested Lula could win in the first round and claimed he would have an overly large margin of victory. Analysts should trust the polling aggregation, not a single poll released in the final days of the race. Lula’s campaign hyped those final polls in their late rallies and the media posted them on the front pages of newspapers and websites, giving them outsized coverage and importance.
This happened before. In 2002, 2006 and 2010, there were late polls suggesting that the election might be over in the first round and then Lula/Dilma fell short. Media headlines then made the failure to finish in the first round seem like a loss. When pollsters tell you in 2026 that Brazil’s election might be decided in the first round, please remember that a two round race is more likely.
Second, there were some serious polling misses at the state levels. Bolsonaro-backed candidates impressively overperformed in São Paulo and Rio Grande do Sul. In fact, Bolsonaro’s margin over the national polling aggregation can be explained almost completely by how well he and his gubernatorial candidates did in those two states.
The perception that polls are off matters because Bolsonaro, the president’s campaign, and his followers are taking it quite personally. He is going to beat this “fake poll” narrative into a key issue of the race and then use it as part of his “proof” of fraud.
If Lula continues to lead the hypothetical second round polling, he wins, even if by a smaller margin than the polls show. The aggregate of the national polls remains quite accurate and they are capturing Lula’s support fairly well, even if there are a few percentage points of “shy Bolsonaro” supporters out there. If the polls start to show the candidates statistically tied or Lula well below 50% with many undecided voters remaining, then the incumbent has a chance. However, that would take a serious shift in the electorate that has yet to occur. The first round results demonstrate that Lula is leading this election right now and is very likely to do so through the second round.
Governance will be a challenge for Lula
While Lula is likely to win the election, the first round results have left the political situation even more difficult for him. Brazil’s right was expected to do well in the Congressional elections, and they did. They overperformed in certain state and municipal elections, leading in key local races. Lula will need centrists to govern, but even with them, he’ll face serious checks on his power from a strong opposition in the Congress and at the local level.
While some politicians on the right may secretly understand that Bolsonaro is particularly unfit to govern, they’re stuck with him, and they will govern according to that reality. Bolsonaro will continue his Trump-like path as ex-president, haunting the political system and making life difficult for both the new administration and his own political allies.
Tomorrow’s weekly politics and polls newsletter for paying subscribers will have more analysis on the results.
Boz-
A few things.
Somebody wrote that the half dozen or so other candidates got 9% (later amended to almost 10%). I have not seen anything saying where those votes will go during the second round.
Somebody also wrote that due to Carnival, thong bikinis and Brazilian wax jobs (I had to look that up) people outside Brazil think it is automatically liberal. The article brought out that Brazil always had a conservative side (not sure how much that is). It must be close to 43%.
Lastly, not a peep about voter fraud or hacked automated voting machines.