It looks like a dead heat in Punjab. As curtains come down on one of the most keenly and bitterly contested elections in Punjab, it is difficult to say who will form the next government. You have to wait till the mid-day of February 27 to find out the real winner in this cliffhanger.
You don't expect to read these lines in an exit poll based report. Exit poll, after all, is as close one can get to the real result as is possible through the technique of opinion polling.
But the assembly elections in Punjab show us the limits to what any poll can tell us. Only a brave pollster can make a definite forecast for this election. Only a believer would bet on the basis of these forecasts.
For the less credulous, here are some estimates and projections based on the HT-CNN-IBN Exit Poll for Punjab carried out by the Centre for the Study of Developing Societies.
Our exit poll, based on interviews with 5217 respondents outside 115 polling booths in 40 assembly constituencies spread across the state, shows that both the contenders for power are placed evenly in terms of votes.
We estimate that the Akali-BJP alliance and their rival the Congress are tied at 41 per cent each of the popular votes. What we can say with fair degree of confidence is that this is not a wave election, that we are looking at a keen contest. We cannot be very confident about who will touch the tape first in a photo finish.
The remaining 18 per cent of votes are likely to go to other smaller parties and independents. The BSP appears to have shed a small slice of its votes and may finish around 5 per cent.
The CPI which did not have any alliance with the Congress this time and joined the CPI(M) and the Lok Bhalai Party to contest separately, does not appear to have damaged the Congress very much. Akali Dal (Mann) also appears to have shed some votes.
The real question is: how would this translate in terms of seats? The tricky part comes here. If both the contenders get the same share of votes, it does not mean that they will get the same number of seats.
Recent electoral history
The recent electoral history of Punjab suggests that when both the sides get the same proportion of votes, Congress gets more seats. This is exactly what happened in 2002 assembly elections.
The Congress-CPI alliance and the Akali-BJP alliance were tied in terms of votes. But in terms of seats, the Congress led alliance won 64 seats, leaving the Akali-BJP combine far behind at 44 seats. Akalis won a few seats with huge margins but lost many seats with small margin. This may not, of course, happen every time.
The real spin in the story is the regional break-up. The exit poll findings suggest that the traditional political geography of Punjab is in for a toss. The Akalis are set to make major gains Majha and have an upper hand in Doaba region. Both these regions have traditionally favoured the Congress.
One important reason for this is that, unlike last time, the BJP has proved to be a useful ally for the Akalis in these two regions with substantial Hindu population.
However, the Congress is set to gain an upper hand in the agrarian belt of Malwa, the traditional bastion of the Akalis. Since the Malwa region has 65 seats in the 117 member assembly, this can offset other losses for the Congress.
Taking everything into account we expect the Congress to win between 50 and 60 seats, and finish neck and neck with Akali-BJP combine that may also win 50 to 60 seats. In political terms this projection leaves all the three possibilities open.
One possible scenario is that the Congress will secure the largest number of seats and with some luck may even win the 59 seats required for majority.
The second possible scenario is that the Congress and Akalis both may finish with equal number of seats and the balance will be held by the half a dozen MLAs from outside the two major blocs. The third scenario, of Akalis finishing at the top and forming the government too cannot be ruled out.
An election like this can humbled any pollster in the world. The only honest response is to raise one's hands and say that this election is too close to call, to bow one's head and acknowledge that psephology is not so developed a science as to call such a close election with precision. And to invite you all to watch the television when the EVMs are opened on the morning of February 27.