Reading Indian Elections is Increasingly Becoming Futile

In the days following electoral results in West Bengal and Assam, analysts have been busy explaining what happened. Anti-incumbency after 15 years of Trinamool Congress rule in Bengal, urban-rural divides, Hindu consolidation across caste lines, the Matua vote, the school recruitment scam, women’s safety concerns after the R.G. Kar case, employment anxieties, welfare fatigue, corruption, leadership centralisation – all these have been pressed into service. In Assam, development delivery, Himanta Biswa Sarma’s personality politics, and opposition weakness have been the preferred frames.

The analyses are diligent, often sophisticated – but largely beside the point. When the conditions under which elections are conducted are themselves severely compromised, no amount of psephological ingenuity can reliably tell us what voters actually wanted.

Let’s begin with the voter rolls. Ahead of the 2026 West Bengal elections, 91 lakh names were deleted from the rolls of the 2024 general elections. Of these, about 58 lakh were deleted under standard categories – absent, shifted, dead or duplicate.

The truly unprecedented element was a separate category called “logical discrepancy”, a classification with no basis in the constitution or electoral law, invented specifically for this exercise and applied nowhere else in India. Through it alone, 27 lakh additional voters were deleted after rushed adjudication, with the Supreme Court freezing the rolls before appeals could be heard, effectively stripping them of their franchise.

In Nandigram, Muslims – 25% of the population – accounted for 95% of those deleted under this category. Across six other constituencies combined – Manikchak, Mothabari, Samserganj, Baharampur, Bhabanipur and Ballygunge – where Muslims together form about 52% of voters, they made up nearly 93% of logical discrepancy deletions. The Election Commission of India (ECI) has not produced a single verified figure of illegal immigrants actually detected. Analysts examining turnout figures and booth-level patterns in Bengal are working from a voter base that was engineered before even a single vote was cast.

Separately, nearly seven lakh last-minute voter additions prompted grave allegations of impropriety.

In Assam, the distortion took a different form. The 2023 delimitation exercise had already systematically reduced Muslim electoral influence, cutting the number of Muslim-majority constituencies from around 32 to approximately 22 – through fragmentation of Muslim-dominated areas across Hindu-majority seats, reservation of erstwhile Muslim-majority seats for Scheduled Castes or Tribes, and the merger of seats that had traditionally returned Muslim legislators.

An analysis by Scroll found that the BJP-led alliance directly benefited in 19 seats; Muslim representation in the assembly fell from 31 legislators in 2021 to 22 in 2026. Standard seat-change analysis will record these gains without registering that many were demographically pre-determined.

Then there is money. According to ADR data, the BJP’s total income in 2023-24 stood at Rs 4,340 crore against the Indian National Congress’s Rs 1,225 crore – more than three-and-a-half times as much. But income figures alone understate the accumulated asymmetry. By March 2024, the BJP’s bank balance stood at Rs 7,113 crore against the Congress’s Rs 857 crore – over eight times as much. The TMC, though regionally dominant, had a total income of approximately Rs 646 crore in 2023-24 – a fraction of its opponent’s resources.

With no legal caps on party spending and no serious regulatory mechanism to enforce parity, comparisons of campaign effectiveness across parties are exercises in false equivalence. A party that can outspend its rivals at this scale is not competing on the same terrain.

The financial lop-sidedness is compounded by targeted agency action. Weeks before major polling phases in West Bengal, the Enforcement Directorate arrested Vinesh Chandel, co-founder and director of I-PAC, the TMC’s campaign consultancy firm, in a money laundering case. The day after polling concluded, a Delhi court granted him bail with no objection from the ED. 

In the interim, I-PAC significantly curtailed operations across the state. The alignment of the timing is its own argument, and no electoral analysis can account for campaigns dismantled by state power mid-election.

The campaign was further defined by rhetoric of identity threat that rendered issue-based analysis inadequate. In Assam, Sarma, who has been re-elected as chief minister now, cast Bengali-speaking Muslims as “Miya infiltrators” engaged in fabricated “jihad” – demographic, land, love, and vote – and posted, then retracted, a video merging footage of himself with rifles with AI-generated images of Muslims as targets.

In West Bengal, Prime Minister Narendra Modi and Union home minister Amit Shah amplified the infiltration narrative rally after rally, with Shah framing the BJP’s victory explicitly as a matter of “national security”. Suvendu Adhikari, now the chief minister, described the TMC as a “Muslim government”, and declared in the assembly that the BJP government led by him would “throw out the Muslim MLAs”. Standing before the Bangladesh Deputy High Commission, he also said Muslims “have to be taught a lesson like Israel did by bombing Gaza”.

None of these statements attracted meaningful action from the ECI, whose Model Code provisions prohibiting communal appeals exist, it now appears, chiefly on paper.

There are also unresolved structural questions about the integrity of the count itself. In the past, most notably in Maharashtra, significant discrepancies between real-time and post-polling turnout figures remained unexplained. The ECI, supported in the Supreme Court by the BJP-led government, has opposed demands for 100% VVPAT verification. Access to Form 17C and CCTV footage from polling booths has been denied to independent auditors. The final numbers that any analysis relies upon are therefore not independently verifiable.

One possibility deserves acknowledgement. The results may well have been the same even without voter deletions, engineered delimitation, agency raids and unchecked communal mobilisation. Especially in Bengal, there was real anti-incumbency, genuine governance failure, and documented corruption. But that is beside the point.

Electoral integrity cannot be judged solely by whether the winning party might have won anyway. It is judged by whether the process allowed democratic expression to occur freely and fairly. When that cannot be established, no analytical framework can retrieve what the process has already prevented.

Psephology was always an imprecise science, but it operated within a shared understanding that the data it analysed broadly reflected democratic will. That understanding has been steadily hollowed out by mass disenfranchisement, gerrymandering, financial capture, institutional coercion and majoritarian rhetoric left unchecked.

Analyses conducted without seriously engaging with these conditions fall short of accuracy. But even more significantly, they are complicit in normalising what is, by any honest measure, a severe and accelerating deterioration in Indian democracy – most starkly demonstrated, in 2026, in West Bengal and Assam.

Vishal R. Choradiya is an assistant professor with the Department of Professional Studies, Christ University, Bengaluru

The Wire is now on WhatsApp. Follow our channel for sharp analysis and opinions on the latest developments.