Top News

When I became a journalist in 1965, a senior colleague advised me, 'Never make a prediction close to an event. If you are wrong, many will laugh at you. So, make predictions well in advance. If you are wrong, most people will have forgotten what you said. If you are right, you can remind everybody what you had predicted.'

Let me violate that norm today. Most exit polls have predicted a big win for BJP in West Bengal, ending 15 yrs of rule by TMC. I say no, Mamata Banerjee will come back as CM.

Also Read | Bengal: Hundreds stage protest over 'threats' by TMC workers

Exit polls are notoriously inaccurate in India. In theory, polling people as they exit a polling booth should be the most accurate way of predicting the outcome. But in India, exit polls vary wildly from one another and the final outcome.

In West Bengal, four exit polls gave BJP victory and two suggested a TMC win. A 'Poll of Polls' gave BJP a comfortable win. But history shows exit polls cannot be taken seriously. Why do exit polls fail so badly? Axis My India was supposed to release its exit poll results one day later than others to focus on accuracy. But, ultimately, the agency refused to release any result saying too many voters refuse to say who they voted for, and that was fatal for accuracy.

Surveys showing majority support for a party or cause can be wildly misleading if many of those polled refused to give an answer. Many voters fear that telling the truth will be used to harm them, and so stay silent. Others lie outright, worsening inaccuracy further.

Also Read | Day of Reckoning: West Bengal, Tamil Nadu, Assam, Kerala & Puducherry assembly poll results set to redefine India's political journey

High non-responses and false responses hurt the accuracy and credibility of a poll. Axis has been gracious enough to say so, although it must have lost a lot of money on an exit poll that it ultimately did not publish. Other polling agencies know how inaccurate their exit polls are, but do not care since there is a demand for such polls regardless of their shortcomings. Ordinary folk see exit polls as a form of entertainment and gossip rather than accuracy. The proven inaccuracy of astrological predictions does not reduce the demand for them. So, too, with exit polls.

What, then, can be firmer grounds for prediction? One is demography. In the last census, Muslims constituted 27% of Bengal's population. That proportion is unlikely to have changed significantly despite BJP's accusation of rampant illegal migration from Bangladesh. Muslim fear of BJP has reached a new high after many were disenfranchised by a faulty revision of voter rolls just before the election. Most fear this will be the first step to stripping them of their citizenship. They will throw their weight strategically behind any candidate most likely to beat BJP, typically TMC candidate.

The Left Front that earlier ruled for 30 yrs is now a shadow of its former self, but still gets around 15% of the vote. This means the non-Muslim, non- Left share of the total vote is just 58%. BJP will have to win 51% of this 58% to get a majority of votes, an impossible task even if anti-incumbency erodes Banerjee's vote share.

BJP says it is not necessary to win a majority of votes. If the anti-BJP vote is split between TMC, Left Front and Congress, BJP can win with barely 40% of the popular vote. True. But in these days of highly polarised voting, chances of such a fatal split are low.

The second key factor is language. BJP has always been seen as a pro-Hindi party that wants to reduce the importance of regional languages. That handicap has been buttressed by the lack of any good BJP speaker in Bengali. Bengalis are proud of their language, literature and songs. Banerjee is a fine Bengali orator who builds instant rapport with her audiences. So, too, does Abhishek Banerjee, her nephew and heir apparent.

The BJP, alas, depends on Hindi speakers like Narendra Modi and Amit Shah. This may not be too big a handicap in national elections, when national issues are at stake. But a state election is inherently local and requires speakers fluent in the local language. Modi is a great Hindi orator, but cannot compete with Banerjee in Bengali.

In the election campaign, BJP harnessed the services of Himanta Biswa Sarma, Assam's chief minister, who took time off from his own state election to campaign in North Bengal. But he, too, spoke in Hindi - he's not a Bengali speaker.

These two factors - demography and language - look enough to sink BJP. So, forget the exit polls. Mamata will win.