On election day, Lula da Silva’s lead fluctuates between 7.1 and 14 points
The latest polls forecast a lead of 7.1 to 14 percentage points for Luiz Inácio Lula da Silva (PT, left) over Jair Bolsonaro (PL, right), taking into account only valid votes.
Polls conducted by Ipec, Ipespe, and Datafolha show a gap of 14 percentage points.
Paraná Pesquisas indicates that the former president’s advantage is 7.1 percentage points.
According to the telephone poll, Lula da Silva is 47.1%, and Bolsonaro is 40%.

To win in the first round, a candidate must receive at least 50% plus 1 of the valid votes.
In these cases, the calculation of valid votes considers the number communicated by the companies, which is rounded. The valid votes in these polls are therefore estimates.
The calculation did not consider the percentages of whites, non-voters, and undecided.
PowerData’s poll, conducted Sept. 25-27, shows the former president leading the contention with 48% of the valid votes in the presidential succession. Bolsonaro has 38%.
PoderData’s poll has the highest frequency in the current election cycle.
DIFFERENCES IN THE POLLS
Presidential elections are a great challenge for polling organizations.
Many results indicate divergent signals. It has become difficult to discern the actual trend.
It is important to say that all polls are accurate, each within its chosen methodology.
Each system can have advantages and disadvantages depending on the constellation they intend to identify.
In 2018, for example, there were a lot of “shame votes” for Jair Bolsonaro.
In some in-person polls, capturing this type of preference was difficult.
On the other hand, phone polls gave more certainty to a portion of voters who chose the then PSL presidential candidate (today Bolsonaro is in the PL).
It is still unclear what impact each method has on data collection.
However, it is already known that face-to-face polls tend to result in a smaller lead for Lula da Silva.
And telephone polls (especially automated and neutral polls where a tape asks the questions, as with PowerData) tend to show a closer race.
In the United States, in-person polls have not been used to gauge voting intent at the national level for decades.
The extremely polarized environment makes data collection difficult when the interviewer and respondent are face-to-face.
In short, it is important to note that the issue is not that one poll or another is wrong. It is a matter of different methods.
At the end of this campaign, it will be known which system is best suited to show trends in current Brazilian politics.
Read More from The Rio Times