The first round of voting occurs on October 7th, this Sunday.

Traditional metrics such as polls and betting markets have been all over the place over the last few months.  A variety of candidates have been pushed into front-runner status by these traditional metrics — Lula, Haddad, and Alckmin occupied consensus leadership at one point.

SMI (Social Media Influence), however, has consistently shown Bolsonaro to be the front-runner since 2017.

Going into this Sunday’s vote, Bolsonaro’s SMI points to an astounding 42% of the vote, excluding null and blank votes.  In comparison most polls have him around 30%.

There is a possibility that his SMI has become temporarily elevated due to the stabbing attack and subsequent hospitalization.  In this case, he will still likely outperform polls but not to the same degree (judging from his SMI just prior to the attack).  According to SMI, the real question is if Bolsonaro just outperforms his poll figures or if he beats them by about 12 percentage points.  This is a completely out-of-consensus forecast and should be taken as such.

SMI has also uncovered a late surge by Gomes.  Haddad is seen by most as obviously finishing in the top two in the first round which would qualify him to move to the second.  SMI, however, shows Haddad slipping going into voting with Gomes surging.  It will likely be a case of too little too late for Gomes, but SMI shows that the race for second place will be closer than traditional metrics currently show.

Brazil’s election has been extremely volatile.  It is a highly contentious environment, even on social media.  Additionally, this election has moved fast, with candidates rising and falling quickly.