Call it lazy modelling, call it ideological comfort, or call it a refusal to confront inconvenient data—but I see the persistence of pollsters in projecting an “edge” for the Trinamool Congress as something far more deliberate. This is not misreading anymore; this is evasion. Because when you strip away the noise, the numbers tell a story that many are simply unwilling to acknowledge—and it does not end in a fourth straight victory for Mamata Banerjee.
Let me begin with what I believe is the single most underplayed statistic in this entire debate: nearly 90 lakh names removed from the electoral rolls after the Special Intensive Revision. In any serious analysis, this alone should have forced a complete rethink. Instead, it is being brushed aside as though it were a minor administrative footnote. The same experts who obsess over marginal vote swings suddenly develop selective amnesia when confronted with a structural shift of this magnitude. The obvious question—one that continues to dodge honest answers—is this: how does a party retain an “edge” after losing access to such a vast pool of voters, particularly minority Muslims, who are known to vote in far tighter consolidation than the fragmented patterns seen among the majority?
To understand why this matters, we must go back to the last assembly election. The Bharatiya Janata Party winning 77 seats was not an accident. It happened despite an electoral ecosystem that was widely believed to be inflated and, at times, compromised. In other words, the BJP rose in spite of structural disadvantages. Now, with a significant clean-up of the rolls, I find it difficult to accept the argument that nothing materially changes. On the contrary, everything changes.
What I find even more puzzling is the way headline statistics are being used to build a misleading narrative. We are told that 63.3 percent of deletions are Hindus and 34.2 percent are Muslims—as if that alone settles the debate. It doesn’t. Elections are not decided by who gets deleted on paper, but by who actually votes and how consistently.
When I break the numbers down, a very different picture emerges. The largest segment—58 lakh voters—falls under the category of dead, absent, or permanently shifted. These are not active voters. They do not influence outcomes in any legitimate sense. Their presence on rolls has always raised questions, and their removal, in my view, only cleans up the system. If anything, it neutralizes avenues that could have been misused.
The next segment—5.5 lakh voters who failed to verify their existence—raises even more doubts. With 96 percent attributed to Hindus, it is easy to jump to conclusions. I don’t. Because the real question is: why did they not show up? In an exercise of this importance, absence speaks volumes. These are, at best, uncertain entries and, at worst, non-existent ones. Either way, they were not reliable voters.
The real electoral impact lies in the final segment—27 lakh voters removed after judicial scrutiny. Here, the composition tilts significantly, with 65 percent Muslims and 30 percent Hindus. This is where the debate should have focused. And even here, when I factor in voting patterns, the conclusions are far from what is being suggested. Only a fraction of the Hindu segment aligns with the BJP, while a significant portion does not. By my estimate, the BJP’s actual loss is in the range of 6 to 8 lakh votes out of the 90 lakh deletions.
That is the number I consider politically relevant. Not the headline figure. Not the percentages quoted without context. But the actual, on-ground electoral impact.
Which brings me to the larger point. For decades, identity-driven politics has relied on the consolidation of predictable vote banks. It worked because the system allowed it to work—unchecked rolls, weak verification, and a narrative ecosystem that discouraged scrutiny. What we are witnessing now is a disruption of that model. The moment you begin to clean the rolls, the very foundation of that strategy starts to weaken.
This is precisely why I believe many pollsters are getting it wrong. They are still analysing a past that no longer exists. Their models are built on assumptions that have just been fundamentally altered. They continue to overestimate bloc voting and underestimate shifts that are far more subtle but equally decisive.
The real contest in West Bengal today, as I see it, is not evenly balanced. It is a test of whether an incumbent can withstand the erosion of its structural advantages. Because once those advantages are diluted, the fight moves to organisation, credibility, and genuine voter connect.
This is where the Bharatiya Janata Party has been steadily building ground, and where the Trinamool Congress is now under pressure. A cleaner electoral roll does not guarantee victory for anyone—but it certainly removes the cushions that once made victories predictable.
To me, the writing on the wall is clear. A political formation that sees a significant portion of its support ecosystem disappear—whether real or artificial—does not walk into an election with an advantage. It walks in with a problem.
Pollsters may continue to hedge. They may continue to qualify their projections. But I prefer to go by numbers when they are this stark.
And those numbers, in my reading, point in only one direction.
Advantage BJP.
A reckoning TMC can no longer postpone.
