Time for a look at prospective election outcomes in close detail. The numbers haven’t moved much: if they were share prices, they would be described as ‘trading within a narrow range’. The realwinning line, as I argued in my last post, is 323. At the moment no party lookseven vaguely likely to get there alone. Instead, the bookies and the pollsters both have for some time had the Tories wobbling around somewhere in the range of 280 to 290 seats. Labour’s vote inScotland has collapsed, as everyone knows; if it hadn’t, they would have a clear lead. Instead they are wobbling around the 260 to 270 level.

With numbers like this, the other parties decide the next government. There’s a sense in which it doesn’t much matter how many seats the SNP win in Scotland, since Labour and SNP areboth anti-Tories: when it comes to the question of who’s next through the door of 10 Downing Street, 265 Labour and 45 SNP is in effect the same as 270 Labour and 40 SNP. Let’s give the SNP 48seats, because that’s what they were on in Newsnight’s latest projection last night. On the same basis the Lib Dems can have 27 seats. The Ulster Unionist parties have come to a deal tonot contest each other’s seats, so that will probably give them 10 between them. Say five for Sinn Féin, three for the SDLP, four for Plaid Cymru, Caroline Lucas holds Brighton for the Greens, Ukiphold one of their Tory-defected seats and pick up another (as per Newsnight). That makes the conveniently round number of 100 seats between the smaller parties, with Labour and the Toriesfighting over the other 550.

The crucial break point is between 290 and 280 Tory seats. The Tories will, you have to assume, be able to cut a deal with the Unionists. 290 plus 10 Unionists gives 300, at which point the 27 LibDems can put the Tories back in office either by voting with them or promising to abstain: 300 plus 27 is a winning 327. Labour plus all the others would be on 318.

The maths of 280 Tory seats are very different. Cut a deal with the Unionists, chuck in the Lib Dems, and the Tory alliance would still only be on 317. Lob on a couple of Ukip, wildly implausiblethough that multi-party alliance would be. They’re still four seats short of 323, and there doesn’t seem to be any way of getting there. Labour would have 270 seats to start with: give them the SNPand they’re one ahead of the Tory-Lib-Dem-Unionist mash-up, on 318. But they would have a lot more material to work with: Plaid, the SDLP and the Greens would give them eight seats and take them toa cosy, luxurious (by these standards) 326.

So the Tory numbers are 290 for government, 280 for opposition, with the Labour numbers being 270 and 260 respectively. These are useful figures to keep in mind when following the coverage.Newsnight currently has the projection as 279/270, which puts Miliband in Downing Street.

What about the Tory numbers in between 290 and 280 – the ones currently predicted by spread bettors, who have historically tended to be more accurate than pollsters? If you think the outcomes I’vealready discussed are unacceptably messy, look away now. 285 Tories plus 10 Unionists plus 27 Lib Dems is 322. Labour would be on 265: that plus SNP plus everybody else on the left would take themto 321. The two Ukip seats would mean that Farage’s party held the balance of power. Give the Tories one more seat: 286 Tories plus 10 plus 27 would be 323, the narrowest win. Give them 284: thatwould leave Labour with a chance of putting together a winning anti-Tory total of 324, but only with the help of Ukip. The same electoral outcome means the Tories too could make it to 323, but,again, only with the help of Ukip. The negotiations around all these possibilities would be gruesome.

Note that in all the scenarios outlined above, the Lib Dems have control over who occupies 10 Downing Street. Even with the Tories doing ten seats better than the best case and getting 300 seats,Cameron isn’t prime minister unless the Lib Dems say so. (If they chose not to do that, the Labour route to power would be 250 Labour seats plus 48 SNP plus 27 Lib Dems equals 325, almost certainlyvia minority government rather than coalition.) But it would surely be very difficult for the Lib Dems to exercise their kingmaking role in favour of Labour. If the Tories are ahead in both thepopular vote and the number of seats – which is the projection implied in all these numbers – it would be hard for the Lib Dems to jump the other way. Also, they’ve just kept the Tories in powerfor the last five years. In effect, they would be voting against their own record.

All true. The only way the Lib Dems could plausibly deny office to the Tories (on these numbers) is if they were able to establish a narrative about how the Tories’ plans for the next term areunconscionable. They would have to say things like, we know what the Tories are like in private, and what they really think; we know how far they want to go. One term was justifiable, but give themfive more years and they will destroy the welfare state. The national interest dictates that we have no choice except to kick them out of government.

If the Lib Dems were contemplating doing that, they might well start leaking stories about terrible things the Tories had thought about doing in the last Parliament. The stories might look a lotlike the one on the front of today’s Guardian: ‘Revealed: Tory plan to slash £8 bn benefits.’ In that scoop, Danny Alexander revealed ‘secret’ Tory plans to cut £8 billion in welfarespending, mainly from child benefit. Or they might set up a story like the one on the front of today’s Times, which says ‘Lib Dems to revolt over fresh pact with Tories’. First sentence:‘Nick Clegg will face huge resistance from within his own party if he tries to push through a second coalition deal with the Conservatives, according to Liberal Democrats at every level.’

I’m not saying the Lib Dems are definitely plotting to dump the Tories. If they were, though, this is exactly how they would be giving themselves the political space to do it.