For years the idea of Ed Miliband in No 10 was judged unthinkable, even laughable. But with the general election just over a fortnight away, a consensus is gradually emerging that his premiership is now the most likely outcome.
Winning general elections is all about managing expectations – so it really matters that many are quietly deciding the Labour leader is more likely to end up in Downing Street after polling day.
It all comes down to numbers. Try playing the BBC's majority-builder game, which offers some likely scenarios and gives you the chance to work out plausible governments. More often than not you're likely to find the Conservatives fall short, even with Lib Dem and DUP support. Labour is often relying on either the SNP and/or the Lib Dems to get into power.
That reflects the balance of probabilities as assessed by the Political Studies Association, a group of unspeakably clever academics who suggest the most likely scenario is bad news for Cameron. "The single most likely outcome is at the bottom of the pie chart," says Dr Stephen Fisher of the University of Oxford. "That is a seriously hung parliament with the Conservatives as clearly the largest party but a majority on the left, including the SNP and Liberal Democrats."
His findings are supported by the academics at electionforecast.co.uk, which give the Conservatives a mean 284 seats, compared to 276 for Labour, 25 for the Liberal Democrats and 41 for the SNP.
Exactly how this will all play out remains to be seen. There are lots of variables, but the overall assessment is one that the New Statesman's spin-off website May2015.com makes rather boldly. "Who will win the 2015 general election? With only 18 days to go, we now think it is likely that Ed Miliband will become prime minister this summer," it predicts. The maths just works better for Labour, it seems, because "there are few scenarios in which David Cameron will survive a vote of confidence".
The pollsters are starting to agree. YouGov's Peter Kellner, writing today, says he thinks the Tories are losing both the air and ground war of the campaign, which has wiped out their previous, slim advantage. He expects a late incumbency bonus will help Cameron's cause, but that won't be enough to escape the logic that everyone seems to be coming to.
"Even if the Conservatives do remain the largest party, a result anything like my forecast would probably lead to Miliband becoming prime minister," he argues. "Even if he could secure the support of the Lib Dems and around ten Ulster Unionists, Cameron would be able to count on 317 MPs, while Labour, the SNP and the smaller left-of-centre parties in England, Wales and Northern Ireland would have around 325." On those figures, it's clear the Tories couldn't govern. But Miliband would need the support of the SNP.
Kellner's assessment highlights just how tight this is going to be. Governing is possible with limited majorities, but it is not straightforward. Votes can and may very well be lost. And there are other uncertainties, too – like how the party's supporters will respond to any particular deal.
This is where new polling from Survation out today comes in handy. Of the 634 voters it spoke to who ranked Labour as their first preference, 49% said they would prefer a minority government, compared to 30% who wanted to see a coalition with the Liberal Democrats. That suggests a division which is likely to leave no-one especially happy.
The same is not true for Conservative supporters, interestingly. While 42% would like to see a Tory minority, 41% would prefer a repeat of the last five years. Experience of sharing power seems to have accustomed right-wingers to the idea. Their real problem, as many are now concluding, is that the odds are stacked against them.
That raises questions of legitimacy. If the Conservatives are the largest party in the Commons but can't command a majority, some fear they're not going to exit power quietly. "They would instantly denounce such a government as illegitimate," Jonathan Freedland of the Guardian writes. "Backed by a Tory press in full cry, they would say Miliband had no mandate. They would call him a squatter in Downing Street, insisting he had usurped power… cartoons would appear of Red Ed in the silver medal position on an Olympic podium trying to wrench the gold from David Cameron's grasp.
The struggle for legitimacy is likely to occur in most scenarios as the real weaponisation of 2015 takes place. Forget the NHS; it's our constitution which could end up being commandeered by the losing side after polling day.
There have been the usual gripes about factual inaccuracies and scorn for candidates' tactical voting claims, but nothing has quite topped the SNP's suggestion that a Lib Dem MP is retiring – when he very much isn't.
But then the stakes are rather high north of the border, aren't they? Maybe this explains why Labour seems to be working quite so desperately hard to demonstrate they know which part of the country they're in.
For some reason it does appear rather as if it's Labour's leaflets which are getting the most stick. They don't have a monopoly on candidates which aren't particularly photogenic, it's true, but still….
The data suggests it has singlehandedly persuaded over 100,000 voters to change their minds and back another party. Should we be paying a little more attention to Vote Match?
As this campaign progresses a mental image keeps popping into my head. It just won't shift itself. It's of those invisible airwaves that science tells us makes the radios work buzzing around ordinary voters, unseen and unknowable, as they scratch their heads and wonder who on earth they're going to support. On one level, this whole struggle for power is like that. To many millions of people, who instinctively find all things political tedious and couldn't possibly care enough to bother sitting through a TV debate, all the arguments of the coming weeks are figuratively – and in the case of those airwaves, literally - going over their heads.
And yet these people have a vote. They have just as much influence as you or I or the next political nerd. They have been well brought up and so they intend to vote, because they understand that doing so is a civic virtue and they understand if they do not it will lead to another Hitler. They are determined to put a cross in someone's ballot box. But they have steered clear of the news with sufficient vigour over the last five years that, really, they don't have the foggiest who exactly they'd like to support.
If you're reading this you're probably someone who does rather care a lot about the news. You like to keep in touch. You can probably rattle off a list of the coalition's biggest policy failures and maybe even one or two of its successes. You know who Nick Clegg is.
So you might struggle to even accept that there are some hapless individuals who couldn't tell you, even in the simplest terms, what distinguishes one party from another. And that would be your first mistake, for these people are not hapless morons. They are intelligent. They just aren't especially interested. Good for them.
This explains the limited appeal of the 'Why you should vote' series of books now on sale in every good bookshop. Carefully digesting their arguments would be a great idea if you could keep voters awake for hours and hours of reading. But for our group – who are not so much floating as still moored up in harbour – handing them these books is about as effective as giving them to a four-year-old.
Nowadays their problem is even more acute. The old two-party system used to be comfortingly tribal. People would vote the way their family told them to. Allegiances were clear in a country where you were either on one side or the other. Things were straightforward when there was a class war to be fought.
In the 2015 contest things have changed. There are seven leaders in the TV debates. Parties are warning that a vote for one results in the other getting into power. It is all horribly confusing, even for those who spend an unhealthily long time keeping an eye on British politics.
But there is a silver lining. Technology is getting on rather well, too, these days. And, as in 2010, there are numerous ways of starting out on the painful journey of figuring out who to support. One of them is Vote Match, the 'simple quiz' developed by Unlock Democracy which helps people calculate where their unconscious allegiances lie.
The process is simple enough. It begins with a drag-and-drop exercise that gets voters to rank issues like the economy and the environment in order of their importance to you. That sounds clever. It implies weighting of the results, which sounds good. I found some choices straightforward. 'Trending issues' got placed quite high, because as a fully paid-up member of the Westminster bubble I'm incapable of any perspective beyond the here-and-now. But then things got trickier. Do I really care more about education or the NHS? I am as likely to get ill as I am to have a child. Wait, I already have a child! I leave this page feeling like I've done a rather muddled job, which makes me think the validity of the final result will be undermined. But I don't care because I'm desperate to click the alluring 'GET STARTED' button at the bottom.
What follows is the bulk of the quiz - 20 questions to which not just your opinion is tested, but the strength of that view too. It features issues that are definitely not a big part of the mainstream political debate, like whether landlords' annual rent increases be capped. There are excellent opportunities for voter hypocrisy: you can state you want the deficit dealt with in one question, and then insist that public spending is maintained at current levels in the next. And there are loaded questions, like 'the government should have the power to read everyone's digital communications'. It all feels rather comprehensive, despite only taking about five minutes.
And then comes the result. Mine was a shock: the party logo staring back at me wasn't the one I've voted for my whole life. That was one per cent behind the alternative. In my case the Vote Match app had effectively told me I can pick either, opening up all sorts of tactical voting options. In other cases, where it has been more decisive, puzzled voters really have genuinely been helped. The 2010 version had over a million unique users, 4.5% of which said they'd voted as a direct consequence of using the quiz. And, astonishingly, nearly 13% said they'd changed their minds as to who to vote for on the basis of the result.
That's over 100,000 people who weren't swayed by all the noise of the air war, or even because of a politician knocking on their door and appearing likeable. What proved decisive for this sub-group was what a clever online quiz told them.
Yesterday an especially enlightened voter in the Bridgwater and West Somerset called for a computer to run the government . "There wouldn't be no argument," one Mr Webber told the Today programme while waiting for his bus. "We wouldn't have all this fighting and bickering. If a computer could control it, it would be fair."
We're not there just yet – but for those willing to put their democratic rights into the hands of a computer programme, help is at hand. If it helps cut through the buzzing airwaves and helps undecided voters make their minds up, this little app could end up having a big impact on May 7th.
This parliament has had 48 months since first gathering in mid-May 2010 to sort out the arrangements for the next coalition. So why is it that the MP tasked with overseeing our constitutional arrangements is now demanding a change of plan?
Only the Tories stand in the way of a constitutional convention. If a vote were held in the Commons on this issue, there is a strong chance it would be approved. So is a petition of just 15,000 signatures really the best way of underlining its importance?