Balls holds the key

Politics has seen many changes over the past couple of decades - but if there is one thing that has changed out of all recognition, it is the science of opinion polling.

I have been in this game just about long enough to remember the infamous BBC exit poll in 1987 predicting a hung Parliament. Mrs Thatcher's Tories won a 102-seat majority.

By contrast, this year's exit poll - also predicting a hung Parliament - was very nearly spot-on, not just in terms of the overall outcome but also in terms of the number of seats won by each party.

But if most elections are becoming easier to predict, Labour leadership election are surely the exception that proves the rule.

There are two fairly straightforward reasons for this. Firstly, the single transferable voting system, which usually means that contests are decided on voters' second and sometimes even third preferences.

Secondly, the make-up of Labour's electoral college, comprising MPs, trade unions, and party members, which makes it nigh-on impossible to conduct a meaningful opinion poll.

So the widespread expectation that South Shields MP David Miliband will be crowned as Gordon Brown's successor later this month needs to be taken, at the very least, with a small pinch of salt.

While the Shadow Foreign Secretary certainly has the most support among MPs, and probably among party members, no-one quite knows what the union ballots will come up with, or how important those second preferences will prove to be.

If anyone is in any doubt about this, they only have to look at what happened in the party's deputy leadership election in 2007, when Alan Johnson and Hilary Benn were seen as favourites by the pundits.

They completely underestimated the level of support among the grassroots for Harriet Harman and Jon Cruddas, whose second preference votes ultimately won Ms Harman the job.

That said, leadership elections are not the same as deputy leadership elections where you might feel more able to vote for someone you like the sound of, without necessarily worrying about whether they are capable of winning a general election.

There is a good argument for saying that if the same six candidates as contested the deputy leadership in 2007 had been contesting the leadership, Mr Johnson would have won.

The conventional wisdom in this election has been that Ed Miliband is everyone's second favourite candidate, and that if David is not sufficiently far enough ahead on first preferences, he risks being overhauled by his brother in the latter stages.

The key to it, as with the 2007 deputy leadership election, will be what happens to the second preferences of the third-placed candidate.

Following his strong performance in bashing the coalition, and showing real fighting qualities over the course of the campaign, I think this will in all likelihood be Ed Balls.

I am quite sure this is why talk of a 'pact' under which Mr Balls would become David Miliband's Shadow Chancellor has been doing the rounds over the past couple of weeks.

As it is, I am not sure there ever was such a pact or whether it would even be deliverable.

Mr Balls and the elder Miliband do not appear to share the same views about the importance of tackling the deficit vis-à-vis the need for economic growth, and that may make his appointment as Shadow Chancellor somewhat problematical.

Either way, by my reckoning Ed Miliband will probably need to win at least three fifths of Mr Balls' transfers in order to pip his brother to the post.

My hunch is that he won't, and that it will indeed be David wearing the crown a fortnight tomorrow.