Bihar 2025

Figures as of 30 September 2025, when the final roll was published.

Elections

    Find a constituency or district

    Each constituency is shaded by the names left off or deleted for every 100 electors on its 2025 roll, which lets large and small seats be compared. Tap a constituency on the map or type its name, and its figures appear below with a link you can send.

    Constituency boundaries from DataMeet's assembly constituency map (CC BY 2.5 India).

    1. Names left off or deleted, against the 2025 margin

    Names left off the draft or deleted after it in each seat, set against the margin by which the seat was won in 2025. The margin is the winner's votes less the runner-up's. Where more names were left off or deleted than that, those names were enough to have decided the seat. The lists cannot show whether they did.

    After the final roll

    The draft roll of 1 August 2025 was open to claims, to add a name, and objections, to delete one. The final roll of 30 September 2025 is the result. Here it is set against the roll before the revision. A seat's roll before is its draft roll plus the names on its phase 1 list of names left off the draft, because the roll of June 2025 is published only as a state total. The net fall is the roll before less the final roll.

    Each reason for leaving a name off the draft, on its own, against the margin

    The phase 1 list gives the officer's reason for every name left off the draft. That reason is what the officer wrote down, and nobody checked it. Each reason is taken on its own here. The chart counts the seats where the names left off for that reason alone are more than the 2025 margin, by the alliance that won.

    Could the names left off or deleted have changed who won?

    How the electors left off or deleted might have voted, estimated three ways, and one figure that needs no estimate at all. Each estimate starts from the seat's names still off the final roll, less those entered as dead, and assumes they would have voted at the seat's 2025 turnout. That count is on the low side, because the names added after the draft include new electors. So every estimate is also run on all the names left off or deleted, with nothing added back.

    2. Seats, grouped by who held them

    Names left off or deleted per 100 electors, with each seat labelled three ways. These are the 2020 Assembly winner, the party that led the seat's segment of the 2024 Lok Sabha vote, and the 2025 Assembly winner.

    Did seats that lost more names swing to the NDA?

    If the names left off or deleted had handed seats to the NDA, then among the seats INDIA held in 2020, those that lost the most names should have gone to the NDA most often in 2025. Here those seats are split into thirds by names left off or deleted per 100 electors.

    Did the vote move where more names were left off or deleted?

    Between the 2020 and 2025 Assembly elections the NDA parties gained votes across Bihar. If the names left off or deleted had taken away more of INDIA's voters than the NDA's, INDIA's share should have fallen most where the most names came off the roll. Here every seat is split into thirds by names left off or deleted per 100 electors, and both elections are counted with the parties grouped as they stood in 2025.

    3. Booths within the same seat

    Comparing whole seats mixes politics with geography. Inside one constituency every booth has the same registration officer and much the same villages and migration, so comparing booths there takes most of the geography out. Here the booths the NDA led in 2024 are compared with the booths INDIA led, seat by seat.

    NDA-led minus INDIA-led booths, seat by seat

    The average names left off the draft per NDA-led booth, less the average per INDIA-led booth, in each constituency with at least 20 of each. A bar above zero means NDA-led booths lost more.

    4. Parties and reserved seats

    By the party that won in 2025

    Names left off or deleted per 100 electors, with seats grouped by the party that won them in 2025. The number of seats is in brackets.

    Reserved and general seats

    38 seats are reserved for Scheduled Castes and 2 for Scheduled Tribes.

    5. Seats that changed hands

    The alliance that won each seat in 2020 against the one that won it in 2025, with the names left off or deleted per 100 electors in each cell.

    The 2024 booth leaders come from Form 20, the booth-level count, and the top two parties in every Lok Sabha seat match the declared result. The NDA here is the BJP, JD(U), LJP(RV), HAM(S) and RLM, with the VIP in 2020. INDIA is the RJD, INC, the three Left parties, the VIP from 2024, and the IIP. Everything else is Other. Method.

    How sure are the findings?

    A pattern seen across all seats can weaken, or even reverse, once each seat is compared only with seats like it. So each finding here was measured again within each district, within each seat, and with Seemanchal left out.

    The swing, measured six ways

    Each row is one way of asking whether seats that lost more names moved further towards the NDA from 2020 to 2025. The square is the estimate, in points of NDA lead for each extra name left off or deleted per 100 electors. The line is its 95% interval, the range the true value most likely lies in. A line that crosses zero cannot rule out no effect. The red row is the comparison that allows for each seat's 2020 lead, which the others leave out.

    Questions about this page

    Download the figures for Bihar 2025

    Excel workbook, every table What each sheet holds

    CSV: constituencies.csv districts.csv

    Cite this page

    Raman, Varna Sri. "Elections, Bihar 2025." Off the Roll: SIR analysis for India, ImpactMojo. https://offtheroll.in/bihar/results/. Build 305ca314004f, 15 November 2026. Accessed 15 November 2026.

    Share a listFind your constituency