Youth of today
According to the half-dozen or so polls that have been published since the weekend, the surge in support for the Liberal Democrats following the first televised leaders' debate is holding up. With several polls indicating that nearly one in three voters plan to support the third party, the election currently looks almost certain to result in a hung parliament. The Conservatives are going to find it very hard to win enough marginal seats from Nick Clegg's party to make the gains of 117 a majority would require.

Across the polls, support for the Conservatives seems to have been hit marginally harder than backing for Labour has. Over the weekend before the debate, the Tories were polling between 37 and 39 per cent - excluding the odd rogue poll, that has now dropped to between 31 and 33 per cent. Labour have seen a shift from around 30 per cent to 26 to 28 per cent.

The Lib Dems seem to be holding on to the vote share they achieved in 2005, and at the same time are attracting some of the former Labour supporters who had switched to David Cameron's Conservatives. In addition, the party has been picking up a lot of backing among 18-34-year-olds (see chart, below right). But this support is not necessarily to be relied upon: this group are less likely than any other demographic to be on the electoral register, and are much less likely to vote.

Clegg's factor
The first leaders' debate has been portrayed as a sort of political version of TV programmes such as Britain's Got Talent or The X Factor, with Nick Clegg cast as a Susan Boyle or a Jedward. A complete unknown until he made his dramatic appearance, so the narrative goes, his demeanour immediately chimed with voters looking for change. Instant polls taken on the night had him winning the debate by a distance, and his party's reputation has continued to soar. The implication behind the comparison is that, as acts such as Jedward have discovered, a sudden rise to the top can result in a very swift fall.

But is this necessarily the fate that awaits the Lib Dems? There is evidence that the shift in support started earlier in the campaign. A week before the debate, ICM finalised the interviews for a poll of key Labour-Conservative marginals for the News of the World. Although it did not attract much attention, the result was quite startling. Tory progress in these key areas had stalled and in seats where you would have expected the Lib Dems to be suffering a squeeze from the bigger parties, ICM recorded a vote share that was equivalent to nearly 25 per cent in national terms. Exactly a week later, ICM finalised a standard national poll and found a sharp improvement in the Lib Dems' share, which had reached 27 points (see
graph, above).

Almost all the interviews had taken place before the live debate started in Manchester. But because the poll was only published two days later, the result was swept up into the debate-impact narrative.

Turning out nice again
Changes to voting preferences on the scale we have seen so far during this election campaign make it very difficult to forecast the final vote shares, let alone the number of seats. The standard calculators that seek to translate poll shares into seat projections have never been tested in conditions like these. But it is possible that Labour will win the most seats, despite coming in third place on votes.

This anomaly does seem to have increased public interest in the election. Some polls even suggest that voter turnout will receive a boost - perhaps outstripping the 61 per cent who voted in 2005.

Mike Smithson is the editor of politicalbetting.com