Regular readers of the magazine, and of the NS blogs, will be aware that my colleague James Macintyre and I have been banging on for a while about the "soft" Tory lead and the muddled media's desire to crown Cameron now, rather than wait for the verdict of a general election next year. We've long predicted that the Tory poll lead will shrink, that Brown's performances will improve as the election nears and that the prospects will increase for a hung parliament -- or, at least, a Tory victory that falls short of a working majority.

Such views have made James and me rather unpopular in the partisan and closed-minded Tory blogosphere. So, imagine my pleasure to see the front page of the Guardian this morning:

Labour back in race as poll gap narrows

"Back in race"? I thought the press pack had decided long ago that the race was over. Cameron had won. This was all just a formality. Wasn't it?

Let's not forget that today's "back in race" headline comes from the same Guardian which said, only a few months ago, that Gordon Brown should be "cut loose" because "it is too late. The chance for him has passed." Hilarious. How the PM must be smiling in No 10 today -- and that, too, without having to be told to do so by an image consultant.

Then there's Mike Smithson of PoliticalBetting.com. He warned those of us who kept going on about Cameron's poll lead being far smaller than Tony Blair's in 1996 to heed only the ICM results (which showed Cameron's Tories 17 points ahead of Labour in November 2009, and Blair's Labour 19 points ahead of the Tories in November 1996):

I say repeatedly that the only valid comparisons are with ICM.

Tory bloggers like Iain Dale were -- unsurprisingly! -- quick to link to his analysis.

But, hats off to Smithson. With the Guardian's ICM poll today showing the Tory lead down to 9 points -- the first poll by ICM since December 2008 to give the Tories less than a double-digit lead -- here is his admirably honest admission, having already predicted that Labour's progress in the polls had "stalled":

I must admit that I got it wrong.

Nice of you to say so, Mike!

So, point-scoring and ha-ha-ing aside, what does this actually mean, in political and electoral terms? The Guardian suggests that a 9-point lead would still result in a (narrow) victory for Cameron's Conservatives -- especially with the party's poll rating not yet having fallen below the crucial 40 per cent threshold in this latest ICM poll. The paper notes:

One projection suggests a Cameron government would be returned with an overall majority of four.

But my colleague George Eaton has run the numbers through UK Polling Report's Swing Calculator, and says it is

. . . a result that on a uniform swing would leave David Cameron six seats short of a Commons majority.

A hung parliament? Hooray! As the Guardian acknowledges, such polls are "likely to intensify calls for Brown to go to the country on 25 March next year, rather than the 6 May polling day that most at Westminster have been expecting". But there is a danger here: Brown cannot be seen (again!) to be chasing the polls and adjusting his calculations simply on the basis of the latest results from ICM, YouGov, Populus et al. The Prime Minister cannot afford to look like a "ditherer" or a "bottler", or to repeat history in chasing the ups and downs of the polls and giving mixed messages to the media, the party and the public.

Here's Professor John Curtice on Comment is Free:

Of course the mood in politics often reflects the perceived direction of travel rather than current location. And Labour MPs are quite right to claim that their party's vote has firmed up in recent weeks and months. No wonder they are feeling a little better. But all that has happened is that a calamitous average polling figure of 21 per cent at the height of the expenses scandal is now simply a slightly less calamitous 28 per cent.

Labour MPs might care to bear in mind too that this means their party is no more popular now than John Major's Conservatives were at this stage in the 1992-97 parliament. That is hardly a happy precedent.

So there is little evidence in the polls so far to suggest that Brown should opt for March rather than May. But equally the time for dithering is past. If by the end of February the Conservatives' lead has consistently and significantly shrunk further -- to, say, just 5 points or less -- then March will look attractive.

I think, come what may, Brown has to stick with 6 May for a variety of reasons, chief among them the local elections and the (negative) impact on the Labour Party's activist base -- or what's left of it -- were he to go sooner. LabourList's Alex Smith is spot on:

I think a ballot on any day other than 6 May would be the wrong move for two main reasons. First -- and most importantly -- with council elections already mandated for 6 May, the cost of holding two elections within six weeks of each other would be excessive -- and would be deemed to be excessive by the public.

The second, related, reason is that an election on any other date would dramatically reduce voter turnout for the local council elections -- and after the decimation of Labour's council position last June, that would be very unhelpful to Labour groups who are battling to maximise support.

So unless the 6 May council elections were also brought forward, which would be difficult to do at this late stage, it would be seen as a crude and unnecessary move which wouldn't play well with either grass-roots activists or the wider electorate.

A week, as one of Brown's predecessors remarked, is a long time in politics. The more weeks between now and the general election, the more time to try to turn things around -- and for Cameron and Osborne to slip up. I doubt Labour can win outright, but I know -- I have a feeling in my gut! -- that the Tories can lose. Stick with 6 May, Gordon.

DISCLAIMER: When I refer loosely to the Tories losing, I mean a hung parliament or a Tory victory short of a working majority, that is, fewer than 20 seats or so.

UPDATE: I have amended the paragraph above to reflect the valid criticism posed by the commenter "Stark Dawning", below. Mike Smithson did, however, seem to be suggesting that ICM is the benchmark for polls and comparisons and that ICM showed a huge Tory lead. It is therefore valid to point out to Smithson and co that their beloved ICM has joined YouGov, Populus, etc in showing a narrowing of the gap between the two main parties. And it is also therefore valid, and not "pointless", to now compare Blair's 19 per cent ICM lead in November/December 1996 with David Cameron's 9 per cent ICM lead in the latest Guardian-published survey.