Forecast - Brazil 2026 general elections: State legislatures

SummaryChamber of DeputiesFederal SenateGovernorsState legislaturesPresidential
FrançaisEnglishPortuguês
União Progressista leads across the 27 legislatures
Le Millénaire · 20 000 simulations · updated October 2, 2026 · first round on October 4, 2026
Polls through 2026-10-01 · election in 2 days
The balance of power on October 4
192.3
União Progressista · expected seats
90% : 120–281 · largest 55%
158.1
PT federation · expected seats
90% : 94–236 · largest 24%
145.7
PL · expected seats
90% : 86–220 · largest 16%

Read the result, then trace it back to the races and sources. Each state has its calculation card: hover over the map, search a name or pick a state.

The 27 federal units, in three views

One race. Its numbers. Its calculation. Hover, tap or search for a state.

Balance of power

1059 seats. Lula’s coalition : 256.3 (186–336) · Flávio Bolsonaro’s PL : 145.7 (86–220).

ListExpected seats90%2022Largest party
União Progressista192.3120–28118555%
PT federation158.194–23615524%
PL145.786–22012616%
MDB114.268–169943%
PSD91.651–143791%
Republicanos88.649–139761%
Podemos49.826–8148
PSDB-Cidadania47.925–7874
PSB45.924–7455
Renovação Solidária37.118–6254
PSOL-Rede27.012–4729
PDT25.412–4344
Avante15.57–2814
Novo9.83–205
Mobiliza5.61–116
Agir3.91–85

States where first place is within 3 points

StateLeadingSecondSeats
RN Rio Grande do NortePT federation 26.8%PL 26.6%24
MT Mato GrossoMDB 18.0%Podemos 17.7%24
GO GoiásMDB 16.4%União Progressista 15.7%41
SP São PauloPL 18.0%PT federation 17.3%94
ES Espírito SantoUnião Progressista 14.5%Republicanos 13.5%30
DF Federal DistrictUnião Progressista 15.1%PL 13.9%24
MS Mato Grosso do SulUnião Progressista 22.4%PL 21.1%24
RO RondôniaUnião Progressista 16.5%Renovação Solidária 15.2%24
TO TocantinsUnião Progressista 18.8%Republicanos 16.8%24
RS Rio Grande do SulPT federation 18.5%União Progressista 16.4%55

States where the leading list would change

Compared with the largest party by seats in 2022. 11 of 27.

State2022Leading in 2026Seats
RN Rio Grande do NortePSDB-Cidadania 10PT federation 26.8%6
MT Mato GrossoUnião Progressista 5MDB 18.0%4
GO GoiásUnião Progressista 9MDB 16.4%7
ES Espírito SantoPL 5União Progressista 14.5%5
DF Federal DistrictPL 4União Progressista 15.1%4
MS Mato Grosso do SulPSDB-Cidadania 7União Progressista 22.4%6
TO TocantinsRepublicanos 7União Progressista 18.8%4
AP AmapáRenovação Solidária 4União Progressista 19.0%5
CE CearáPDT 13PT federation 20.5%9
MA MaranhãoPSB 11MDB 22.7%9
PB ParaíbaRepublicanos 8União Progressista 35.1%13

All races

StateLeading listSecond listVote gapSeatsRating2022
AcreUnião Progressista 8PDT 317.6 pts24SafeUnião Progressista 5
AlagoasMDB 15União Progressista 631.1 pts27SafeMDB 14
AmazonasUnião Progressista 7Avante 311.0 pts24LikelyUnião Progressista 6
AmapáUnião Progressista 5PDT 43.5 pts24LeansRenovação Solidária 4
BahiaPT federation 17União Progressista 127.0 pts63LikelyPT federation 17
CearáPT federation 9PSB 83.6 pts46LeansPDT 13
Federal DistrictUnião Progressista 4PL 41.2 pts24Toss-upPL 4
Espírito SantoUnião Progressista 5Republicanos 41.1 pts30Toss-upPL 5
GoiásMDB 7União Progressista 70.7 pts41Toss-upUnião Progressista 9
MaranhãoMDB 9PL 67.2 pts42LikelyPSB 11
Minas GeraisPT federation 14PL 104.1 pts77LeansPT federation 17
Mato Grosso do SulUnião Progressista 6PL 51.2 pts24Toss-upPSDB-Cidadania 7
Mato GrossoMDB 4Podemos 40.2 pts24Toss-upUnião Progressista 5
ParáMDB 13União Progressista 616.8 pts41SafeMDB 13
ParaíbaUnião Progressista 13Republicanos 715.4 pts36SafeRepublicanos 8
PernambucoUnião Progressista 12PSB 94.9 pts49LeansUnião Progressista 13
PiauíPT federation 12MDB 106.5 pts30LikelyPT federation 12
ParanáPSD 15PL 911.0 pts54LikelyPSD 16
Rio de JaneiroPL 16União Progressista 108.2 pts70LikelyPL 17
Rio Grande do NortePT federation 6PL 60.2 pts24Toss-upPSDB-Cidadania 10
RondôniaUnião Progressista 4Renovação Solidária 41.3 pts24Toss-upUnião Progressista 6
RoraimaUnião Progressista 8Republicanos 413.9 pts24SafeUnião Progressista 6
Rio Grande do SulPT federation 10União Progressista 92.1 pts55Toss-upPT federation 12
Santa CatarinaPL 10União Progressista 511.6 pts40LikelyPL 11
SergipeUnião Progressista 7PSD 411.4 pts24LikelyUnião Progressista 6
São PauloPL 16PT federation 160.8 pts94Toss-upPL 19
TocantinsUnião Progressista 4Republicanos 42.1 pts24Toss-upRepublicanos 7

How it works

A predictive model is not one more poll. A poll interviews a sample and describes opinion at a given moment. A model interviews no one. It gathers what can be measured, published surveys, past election results and the electoral rules, and derives from them a distribution of possible outcomes, with their probability. It does not say a party will win so many seats. It says within which range it will finish, and what each candidate’s chances of election are.

Why it is sturdier than a single poll

Two pollsters published on the same day can differ by several points. The model aggregates every survey published in each state and registered with the Superior Electoral Court, taking into account their age, their size and the pollsters’ reliability.

Past polling errors

Polls get it wrong, sometimes badly. The model builds in the size of the errors seen in past elections, without assuming in advance that one side will be underestimated. That is what gives each candidate their chances, and each list its range.

From votes to seats

Brazil elects its deputies by proportional representation, state by state, under a complex allocation rule. The model starts from the last election’s results and the make-up of the 2026 lists, then applies the official rule. Tested on the 2022 votes, it recovers 1542 of the 1572 seats actually allocated.

Linked races

The October 4 elections are not independent. The vote for lists close to the two leading presidential candidates moves with the presidential vote, and the model takes this into account.

Twenty thousand simulations

A single forecast would give a false sense of certainty. The model therefore replays each race twenty thousand times, varying what is genuinely uncertain. The probability shown is the share of simulations in which the event occurs.

No probability at 0 or 100%

Uncertainty is this model’s product, never its flaw. The simulations are reproducible, and the forecast on the eve of the vote is frozen, then compared with the official results.

Sources

Election results and candidacies: Superior Electoral Court. 2026 polls: surveys published by pollsters and registered with the Superior Electoral Court.

Export assemblees_2026_par_etat.csv ↗ · Export the forecasts as JSON ↗

Credits

All rights belong to Le Millénaire SAS, 47 rue de Vivienne, 75002 Paris.

Model designed by
Project director: William Thay
Deputy project director: Pierre Clairé
Political analysis and data: Matthieu Hocque
Data analysis, mathematics and validation: Florian Gérard-Mercier