One of the professional hazards of being a political journalist is that you are expected to predict exact election outcomes . It's meant to be your USP, the basis for your dinner invite. Yet, over the last few weeks, as one has been repeatedly asked to predict election 2009, the response has been to simply and honestly say, "I don't know."

Never before has an Indian election left so many people so confused about the final verdict. Perhaps, the ghosts of 2004 still haunt all of us and have made us a little cautious. More accurately, it is a reflection of the reality than an Indian general election is no longer one election, but possibly 543 elections being played out at the same time, with almost every constituency having its own unique set of issues and individuals to deal with.

Take Mumbai for instance. If in North Mumbai, the absentee MP Govinda has been an issue in itself for voters, in the neighbouring north-west constituency, slum rehabilitation is a major concern. While a section of South Mumbai goes into cataclysms over 26/11, another part of the same constituency is worried about unsafe old buildings. How is it possible to safely predict a trend for even a single city when there are such sharp variations even within a constituency?

Delimitation has made predictions even more hazardous. As the maps of constituencies are redrawn, there is no history to fall back on, no previous trends that can be used as a reliable barometer. Delimitation has dramatically altered the demographics of several constituencies. To take South Mumbai as an example again: affluent Pedder Road must now co-exist with not just middle class Kalbadevi, but also with the Maharashtrian labour stronghold of Parel.

Another new challenge is the sheer competitiveness of each election. Very few constituencies in the country can strictly claim to be bipolar any longer. In almost every constituency, there are now powerful rebels contesting as independents, or smaller parties attempting to play the role of 'spoilers'.

Will Raj Thackeray's MNS play this role in pockets of Maharashtra, will the Muslim-dominated AUDF be a factor in Assam, which major party will Chiranjeevi damage more in Andhra, how much vote share will a Vijaykanth get in Tamil Nadu, and, of course, just how much vote will the Bahujan Samaj party take across the country? In the answers to these questions and more, perhaps lies the key to unlocking the great Indian election puzzle.

And yet, with all the statutory warnings and before being accused of a complete cop out on predictions, it is possible to do some basic trend-spotting. Firstly, it is clear that the definition of a 'national' party needs to be redrawn. The Congress, as a result of its withering away in several major states, is in serious contention in a little over 300 of the 543 seats in the country. The BJP, which has more severe geographical limitations, is in the race in even a lesser number of seats. It should then come as no surprise that both the major 'national' parties are looking to win around 160 seats, or less than a third of the total Lok Sabha, as enough to claim victory as the single largest party.

It is equally true that the so-called Third Front, if not a parking lot or a roadside dhaba, as it has been variously called, is clearly a fragile arrangement designed to further individual ambitions of its gaggle of leaders. A front that is unlikely to cross the 150 mark can hardly offer the basis for a cohesive non-BJP, non-Congress alternative.

Which is why it is increasingly clear that while millions of Indians may have braved heat and cynicism to vote over five grueling rounds, the knockout round - the sixth and critical one - really only begins from May 16th. When netas talk of post poll options, they are effectively acknowledging their failure, either as single parties or even as pre-poll alliances, to have won the confidence of a majority of voters. Post-poll then is a euphemism for reducing an election verdict to an auction house, based less on any ideological commitment but almost entirely on a chilling pragmatism.

In 2009, Indian politics is not short of its pragmatic kingmakers. A Prakash Karat may be driven by ideological antipathy towards both the BJP and the Congress, but almost everyone around him has no similar rigid worldview. A Jayalalithaa's chemistry may be with the BJP, but her arithmetic dictates that it's the Congress that can deliver what she most wants: the fall of the Karunanidhi government. A Mayawati may have taken on both the Congress and the BJP, but she would happily deal with anyone who will make her India's first Dalit prime minister.

A Sharad Pawar may have allied with the Congress in Maharashtra, but is looking to cut a deal with anyone who will coronate him in Delhi. Nitish Kumar may be projected as the new secular hope, but his presence at an NDA rally with Narendra Modi should be seen as a decision based on his desire to remain chief minister of Bihar, atleast for now. Naveen Patnaik may have suddenly discovered the darker side of the sangh parivar just before the elections, but a hung verdict in the Orissa assembly could lead to a potential rediscovery of old allies. Not to forget the even smaller parties - like the PMK, Paswan's Lok Janshakti, -- who will simply ally with anyone who will promise them their permanent seat in the union cabinet.

In a sense, the 2009 verdict must be seen as another act in a chapter of Indian politics that began two decades ago when the Congress's monopoly position was seriously threatened for the first time. Since then, it has been apparent that the vacuum left behind has not been completely filled, either by the BJP or the regional forces. Political India today is a mirror cracked, with each political party reflecting broken glass shards. Holding together the shattered pieces will always be a tough and painful process, which is why we may be in for a long and cruel summer.

P.S. For those still interested in numbers, here is a tip: in the last election, the office sweepstakes on how many seats the NDA would get was won by the office canteen boy. This time, it could well be my local panwallah who has the last laugh.