The recent series of state elections was famously described as a mini-general election by the entire club of tele-psephologists. But none dared to hold an opinion or an exit poll. Again, opinion polls are expensive, so perhaps the dwindling revenues of media organisations (TV and print) in this disastrous quarter had something to do with it. But one reason no media organisation was willing to put any money in opinion polls this time was their stunningly consistent record lately of getting it wrong. In the past five years, if anybody has called an election result right, it’s been a mere exception, and that too with many qualifications. Most others have gone wrong.

So what exactly has changed? Has the voter suddenly decided not to share his mind with the pollster? Has something gone wrong with the polling technology? Or, has something changed with our politics so fundamentally that old formulae no longer work, formulae of caste combinations, personal loyalties and enmities, grievances and identity around which the psephologists constructed their swing zones? Is it all to do with just the impact of delimitation of constituencies? Or could it be that our psephologists and other political pundits have missed out on a much more significant delimitation — one in the voter’s mind that is rendering old rules, stereotypes and formulae irrelevant and writing a new script? What is the voter trying to say when she defies anti-incumbency in three states (Madhya Pradesh, Chhattisgarh and Delhi) but not in one (Rajasthan), and then junks all conventional logic in Kashmir to give you this fascinating result?

... contd.