BJP won 206 seats in the 2026 West Bengal elections, ending Mamata Banerjee's 15-year rule. But in nearly 50 constituencies, voter deletions under the Election Commission's SIR exceeded the winning margin. Here is what the data shows.
The 2026 West Bengal Assembly elections have culminated in a historic and seismic shift in the state's political landscape. For the first time since Independence, the Bharatiya Janata Party (BJP) has secured a decisive mandate in the eastern stronghold, winning 206 seats and dismantling the 15-year incumbency of Mamata Banerjee’s Trinamool Congress (TMC).
While the BJP attributes this "Saffron Surge" to a narrative of good governance and an anti-incumbency wave, a shadow of controversy hangs over the victory: the Special Intensive Revision (SIR) of electoral rolls, which saw the removal of millions of voters just months before polling.
The West Bengal Assembly election results 2026 represent a complete reversal of fortunes compared to 2021. The BJP, which held 77 seats in the previous assembly, surged to 206, well past the majority mark of 147.
Conversely, the TMC was reduced to just 81 seats, a staggering drop from its 215-seat tally in 2021.
The Fall of Strongholds: The TMC lost 78 of its 124 "safe" seats (which it had won consistently since 2011).
BJP Expansion: The BJP won 100% of the 142 seats it had previously led in, during various Lok Sabha and Assembly cycles, and added 65 new constituencies to its tally.
The Electorate Shift—2021 vs 2026: The most contentious aspect of this election was the shrinking voter base. Following the Election Commission’s Special Intensive Revision (SIR), the total number of registered voters dropped by approximately 10.94%.
The election recorded a historic voter turnout of 92.47%, among the highest in the state's post-Independence history. The BJP needed 148 seats for a majority. It won 58 more than that.
Between 2021 and 2026, the TMC lost vote share in 268 out of 293 counted constituencies. In 69 of those, it shed more than 10 percentage points. The BJP gained vote share in 270 of 293 constituencies, adding over 10 percentage points in 95 of them.
The SIR resulted in a 10.94% reduction in the total electorate. While the absolute number of deletions was highest among Hindus (57.5 lakh), the percentage-based impact on the Muslim voter base was disproportionately high.
| Demographic | 2021 Share | 2026 Post-SIR Share | Share of 91L Deletions |
|---|---|---|---|
| Hindus | 70.5% | 71.4% | 63.4% |
| Muslims | 27.0% | 26.1% | 34.3% |
| SC/ST | 28.5% | 27.4% | 19% |
In the "Adjudication" phase—where 27 lakh names were removed for logical discrepancies—the exclusion rate for Muslims reached 65% in border districts like Murshidabad and Malda.
In the 20 constituencies that saw the highest number of names removed (ranging from 23,000 to over 74,000 deletions), the TMC actually emerged as the dominant force:
| TOP 20 seats with most deletions during SIR adjudication | |||||
|---|---|---|---|---|---|
| Votes deleted | 2021 winner | 2026 winner | 2026 margin | Difference* | |
| SAMSERGANJ | 74775 | AITC | AITC | 7587 | -67188 |
| LALGOLA | 55420 | AITC | AITC | 18960 | -36460 |
| BHAGABANGOLA | 47493 | AITC | AITC | 56407 | 8914 |
| RAGHUNATHGANJ | 46100 | AITC | AITC | 40555 | -5545 |
| METIABURUZ | 39579 | AITC | AITC | 87879 | 48300 |
| FARAKKA | 38222 | AITC | INC | 8193 | -30029 |
| SUTI | 37965 | AITC | AITC | 12357 | -25608 |
| MOTHABARI | 37255 | AITC | AITC | 10496 | -26759 |
| JANGIPUR | 36581 | AITC | BJP | 10542 | -26039 |
| RATUA | 35573 | AITC | BJP | 32562 | -3011 |
| KARANDIGHI | 31562 | AITC | BJP | 19869 | -11693 |
| GOALPOKHAR | 31384 | AITC | AITC | 83790 | 52406 |
| MALATIPUR | 29489 | AITC | AITC | 59747 | 30258 |
| CHOPRA | 27898 | AITC | AITC | 69124 | 41226 |
| SUJAPUR | 26829 | AITC | AITC | 60287 | 33458 |
| KETUGRAM | 26780 | AITC | BJP | 27610 | 830 |
| RAJARHAT NEW TOWN | 24132 | AITC | AITC | 2074 | -22058 |
| BASIRHAT UTTAR | 23900 | AITC | AITC | 57270 | 33370 |
| MANICKCHAK | 23726 | AITC | BJP | 13938 | -9788 |
| MONTESWAR | 23423 | AITC | BJP | 14798 | -8625 |
Murshidabad: 4.55 lakh
North 24 Parganas: 3.25 lakh
Malda: 2.39 lakh
Constituency outlier: Jorasanko in North Kolkata lost 36.85% of its entire electorate.
Before a single vote was cast, West Bengal's electoral map had already been redrawn. The Special Intensive Revision removed around nine million voters from the rolls in West Bengal, representing about 12% of the electorate. Over six million were categorised as absentee or deceased, while the status of 2.7 million remained pending before tribunals.
Observers noted that roughly 65% of the undecided group were Muslims, while Dalit Hindus, especially from the Matua community, were also affected in certain districts.
The demographic breakdown of deletions tells a more granular story.
The Election Commission defended the exercise as a necessary revision to remove deceased, duplicate and ineligible entries, including foreign nationals. The TMC said the exercise risked disenfranchising genuine voters, while the BJP defended it as a purge of bogus entries and illegal migrants.
One of the most revealing data points from 2026 is the BJP's performance in seats it had previously won. According to ECI data, the BJP had won 54 constituencies consistently across the 2019 Lok Sabha, 2021 Assembly and 2024 Lok Sabha results.
In 2026, the BJP retained all 54. It then won 100% of the 142 constituencies it had won in any of those three prior elections, and added a further 65 constituencies it had never previously won.
The TMC's collapse was equally consistent in the other direction. Of the 124 constituencies the TMC and its allies had won in all three of the 2011, 2016 and 2021 elections, it lost 78 in 2026 and won just 34 seats outside that core stronghold group, according to ECI data.
West Bengal had 39 constituencies that consistently elected Muslim MLAs in 2011, 2016 and 2021. In 2026, 34 of those 39 again returned Muslim MLAs. shows ECI data. The total number of Muslim MLAs in the new Bengal assembly stands at 40, nearly identical to 2021. The BJP won the remaining five, and in three of those, a second Muslim candidate split the vote against the TMC.
What changed is internal TMC composition: the share of Muslim MLAs within its legislative group has risen to 42.5%, the very dynamic the BJP used to consolidate Hindu votes across the state.
The most difficult question the data poses is whether the SIR deletions directly determined outcomes. In nearly 50 constituencies across the state, the number of voters removed under the SIR exceeded the winning margin on counting day.
Deletion vs. Margin: In 47 seats, the number of excluded voters was higher than the victory margin.
BJP Gains: Of the 119 seats won by the BJP where more than 5,000 names were deleted, 28 seats saw deletions that exceeded the victory margin. Interestingly, 26 of these 28 seats were won by the TMC in 2021.
The Case of Champdani: In this South Bengal seat, BJP’s Dilip Singh unseated the TMC by a razor-thin margin of 13 votes, while the constituency saw 7,600 deletions during the SIR process.
Jangipur and Karandighi: In Jangipur, the BJP won by ~10,500 votes against 36,581 deletions. In Karandighi, the margin was ~19,800 against 31,562 deletions.
Supreme Court Observation: During hearings on the SIR exercise, the apex court noted that if a winning margin is around 2% while 15% of the electorate is unable to vote, the court would "definitely have to apply minds" regarding the fairness of the outcome.
An analysis of the 187 seats that saw over 5,000 names deleted found the BJP won or was leading in 119 of them. In 28 of those BJP victories, the number of deleted voters exceeded the party's winning margin. Of those 28 seats, 26 had been won by the TMC in 2021.
The TMC, however, also benefited from the same dynamic in constituencies it retained. In Samserganj, with a margin of just over 7,500 votes, over 74,000 voters remained under adjudication. In Lalgola, against 55,420 adjudication cases, the winning margin hovered near 19,000.
The range of deletions in those seats ran from 2.1% to 38.6% of the electorate, with a median of 9.5%, yet the vote share shifts did not track those deletions in any consistent direction.