New Delhi: West Bengal’s latest electoral-roll revision does not point to one uniform statewide pattern. Instead, the assembly-level data suggest two distinct stories unfolding at once: a heavy concentration of voters marked under adjudication in the border belt, and a sharp contraction of rolls in Kolkata and nearby urban constituencies driven more by deletions than adjudication.An analysis of all 294 assembly constituencies in the state shows that the 2025 pre-draft electorate stood at 7.66 crore, while the 2026 final total fell to 7.04 crore. That is a net decline of about 61.8 lakhs names, or 8.06%. Between the pre-draft and draft stages alone, the rolls shrank by about 7.60%. In the final 2026 list, about 60.1 lakhs voters were marked as under adjudication, equal to 8.53% of the final electorate.But those statewide numbers flatten an important geographic divide.The most striking concentration of under-adjudication appears in the border districts, especially Malda, Murshidabad and Uttar Dinajpur. District-level aggregation shows Malda with an average constituency under-adjudication rate of about 27.78%, Uttar Dinajpur at 22.51%, and Murshidabad at 20.10%. That is far above the state average and far above most non-border districts. Yet, interestingly, the overall shrinkage of the rolls is not confined to this border belt. The average net roll decline from pre-draft to final is remarkably similar in border districts (-8.05%) and non-border districts (-8.37%). This reveals a critical nuance: while the adjudication burden is heavily concentrated at the border, the actual disappearance of voters is a statewide phenomenon.At the constituency level, the clustering is even more dramatic. The highest under-adjudication rates in the state are found in seats such as Sujapur in Malda at 52.5%, Samserganj in Murshidabad at 45.94%, Raghunathganj at 45.90%, Lalgola at 41.40%, and Bhagawangola at 40.19%. Also read: Bengal SIR: In Raninagar, Predominance of Muslim Names in the ‘Under Adjudication’ List Unsettles ResidentsThese are not random seats; they sit exactly at the volatile intersection of border geography, high Muslim concentration, TMC strength, and intense BJP-TMC polarisation over citizenship. In plain terms, in several constituencies in this belt, extraordinarily large shares of the final roll are being carried in an unresolved category rather than fully cleared into the electorate.A simple statistical model reinforces that geographic concentration. Constituencies in border districts show an estimated 5.06 percentage point higher under-adjudication rate than non-border seats, even after controlling for whether the seat was close in 2021 and for the size of the TMC-BJP vote-share gap in the last assembly election. That border effect is the strongest result in the analysis.The second major pattern lies elsewhere. The steepest overall roll decline is concentrated not in the border belt but in Kolkata and nearby urban seats, where the numbers point more strongly to deletion than adjudication. The sharpest net falls from the 2025 pre-draft list to the 2026 final list include Jorasanko at -36.78%, Chowrangee at -35.62%, Howrah Uttar at -28.48%, Kolkata Port at -26.38%, and Ballygunge at -25.27%.These same constituencies also appear among the highest draft-deletion seats. Jorasanko shows a draft deletion rate of 36.85%, Chowrangee 35.44%, Howrah Uttar 26.96%, and Kolkata Port 26.17%. That suggests a different revision mechanism from the one visible in the border belt. In urban seats, the data look less like a surge in adjudication and more like aggressive removal or correction of names before the final list, suggesting a massive urban roll-compression regime targeting duplicate voters, shifting tenant populations, address-cleanup issues, and dead or relocated electors.There is also a political pattern in the data, though it is weaker than the geographic one and should be interpreted with more caution. Constituencies won by the TMC in 2021 had a mean under-adjudication rate of 9.54%, compared with 5.64% in seats won by the BJP. When isolating political strongholds—seats won by a margin of 10 points or more in 2021—the gap widens further. TMC strongholds carry an average under-adjudication rate of 11.61%, compared to just 6.28% in BJP strongholds. In the regression model, each additional percentage point of TMC lead over the BJP in 2021 is associated with roughly 0.265 points higher under-adjudication in the 2026 final roll, and the raw correlation between the TMC’s 2021 lead and the final under-adjudication rate stands at a highly meaningful +0.546. In other words, the more strongly a seat leaned toward the TMC in 2021, the more likely it was to show elevated adjudication in the latest roll revision.That relationship is especially visible in the top under-adjudication constituencies, many of which are TMC-held seats in the Malda-Murshidabad belt. Still, the broader statewide contraction in rolls is not well explained by partisanship alone. When net roll change is modelled directly, neither border status nor partisan margin emerges as a strong standalone predictor. That is important because it suggests the state’s roll revision cannot be reduced to a single political story.Instead, the data support a more layered reading.In the border belt, the central issue is the unusually high number of voters whose names remain in the system but are tagged as under adjudication. In urban Kolkata and Howrah, the dominant phenomenon is a sharp thinning of the rolls through deletions. The two processes overlap in the statewide total, but they do not appear to operate with the same intensity in the same places.That distinction matters politically as well as administratively. If the principal concern in border constituencies is adjudication, then the controversy is likely to revolve around documentary scrutiny, verification, and the eventual disposition of large pools of pending names. If the principal concern in urban seats is deletion, the debate is more likely to focus on the scale and accuracy of removals from the list itself.The result is that West Bengal’s voter-roll revision, viewed seat by seat, looks less like one single exercise and more like two parallel ones. One is concentrated in the border districts, where adjudication is the defining feature. The other is concentrated in urban seats, where deletions drive the numbers. Any political argument that treats the entire state as showing the same pattern risks missing the real structure of what the rolls are showing. Crucially, this data draws strict boundaries around what political parties can honestly claim. The BJP can credibly argue that heightened scrutiny aligns with border districts, but the massive urban deletions in Kolkata hollow out any narrative that this exercise was strictly about border infiltration. Conversely, while the TMC can statistically prove that the adjudication burden falls disproportionately on its geographic and social base, the data alone cannot conclusively prove deliberate partisan deletion, as individual-level data on religion, voter identity, or documentation status remains unobserved.