Questions about the elections, answered only with what the evidence supports.

Questions

• Who won the second round of Brazil's 2026 presidential election?

  Confirmed LULA won the second round of 2026, with 64,256,232 votes, 51.79% of valid votes, according to the TSE consolidated result with 100.00% of polling sections totalized.

• Did the extension of voting hours change the 2026 result?

  Unsupported We do not know yet whether it could have. No collected evidence supports that the extension changed the winner, and the calculation against the final margin depends on the official count of votes recorded after 5 p.m.,…

• Were votes recorded after 5 p.m. in the 2026 second round?

  Unknown We do not know yet. The official TSE publication with the count is not in the collection: the site answered HTTP 403 to the automated client, and that block is not bypassed. Without it, no number of votes after 5 p.m.…

• Did all votes recorded after 5 p.m. in 2026 go to the winner?

  Unknown We do not know yet, and the collected files cannot tell: no collected source splits the votes recorded after 5 p.m. by candidate, and the consolidated result does not split votes by time.

• Did abstention fall between the first and second rounds of 2026?

  Confirmed Yes. Abstention was 21.08% of registered voters in the first round and 19.22% in the second, 1.86 percentage point(s) lower. The change is derived; its cause is not established.

• Who won in each state in the 2026 second round?

  Derived In the presidential election nobody wins a state: the winner is whoever has the majority of valid votes nationwide. In the 2026 second round, of the 27 states, FLAVIO BOLSONARO was the most voted in 14 and LULA in 13;…

• How does Brazil's 2026 presidential election compare with 2022?

  Derived In the second round, abstention was 19.22% of registered voters in 2026 and 20.58% in 2022 (1.36 percentage point lower in 2026). The lead of the first over the second candidate was 3.57 points of valid votes in 2026…

How the answers are built

1. Official source downloaded with URL, date and SHA-256 (TSE, STF).
2. Observation: the number or excerpt read from the file itself, never typed in.
3. Evidence: a graph node linked to the source by observed_in.
4. Derived calculation, when there is one: declared formula, linked to its inputs by derived_from.
5. Claim: linked to evidence by supported_by or contradicted_by; context does not become an edge.
6. Claim state: what the chain supports, with what stays open.

Rules

• Only observed or derived evidence supports a fact. Imported, inferred and synthetic items stay as context.
• A confirmed claim with no evidence in favour drops to unknown at load time; the ledger keeps no confirmation without a chain.
• There is no aggregate confidence score: each item shows its factors (source type, traceability, independence, freshness, corroboration, conflicts).
• A source that blocks automated clients is not bypassed; the item stays unknown and the reason is shown.
• No claim here is an opinion, a voting recommendation or an adjective about a candidate or party.

Evidence states

Observed OBSERVED

read from a primary source with URL and SHA-256 checked

Imported IMPORTED

from an external dataset, API or system

Derived DERIVED

a declared calculation over observed items

Inferred INFERRED

output of a model, researcher or algorithm; does not support a fact on its own

Synthetic SYNTHETIC

simulation, forecast or scenario; never enters as a fact

Unknown UNKNOWN

a legitimate absence: no collected source supports the item