Featured, Need to Know | 17th April 2015

Election 2015: New forecast confirms the campaign hasn’t changed polls – how will Tories keep power?

A new academic prediction suggests very little has changed in the polls over the past fortnight, despite much campaigning.

Photo: Getty


Who is going to win the 2015 general election? Two weeks ago our latest election forecast predicted Labour would win slightly more seats than the Tories, in contrast to every other prediction at the time.

Now – after a fortnight of manifesto launches, party political broadcasts, a widely viewed leaders’ debate, and high profile interviews on an almost daily basis – there has been virtually no change in the balance of power between the ‘Big Two’.

Our vote forecast now predicts a narrow Conservative win in the popular vote, but now suggests Labour will win 10 more seats, up from a 5-seat lead two weeks ago (albeit down on the 20-seat lead we gave them last month).

We forecast the Tories to win 33.6 per cent of the vote, down 0.4 points on April 1, and back to the level forecast in early March. Labour are forecast to be 1.1 points behind on 32.5 per cent, down 0.5 points. The modest recovery in Liberal Democrat fortunes continued in the first half of the month – we now project them to win 10.3 per cent of the vote.

Screen Shot 2015-04-17 at 15.16.34

The stability in the polling is also reflected in our latest seat estimates. Our median estimate for the Conservatives’ is down by two seats to 269 (a 1.2 per cent change), Labour tick up by two to 278 (a 0.7 per cent change), and the Liberal Democrats are up one to 28.

This doubles the Labour seat lead from 5 to 10 seats, but, as the figures in brackets indicating the confidence in our estimates show, this projected lead is highly uncertain. A significant lead for either party in seats is quite possible, though a majority for either remains extremely unlikely.


There is evidence that the events of the past fortnight are is starting to filter down to voters. Last week Populus reported for the first time that the general election was the UK’s most noticed news story, and Lord Ashcroft’s polls of marginal seats are now showing often rates of contact above 50 per cent by the main parties.

Screen Shot 2015-04-17 at 15.34.19

The intense political activity of recent weeks may now be reaching voters, but there is little evidence it is changing minds. Although a couple of polls have shown Labour or the Conservatives with a decisive lead – and been widely shared by each party’s supporters – this picture has not been backed by the majority of polling, suggesting they are outliers (or at the very edge of the typical ±3 point sampling error).

Our central projection nicely illustrates the fragmentation of British politics since 2005.

These small shifts in projected seat totals have little impact on the coalition arithmetic. The most optimistic Conservative-led combination – in which the Liberal Democrats, DUP and UKIP all support the party – still falls well short of the 323 votes needed to be sure of winning a confidence vote.

However at the upper end of our projection these four parties would win 340 seats (the lower and upper end of our predictions are in brackets).

Labour—278 (256-300)
Tories—268 (245-293)
SNP—49 (36-56)
Lib Dems—28 (23-34)
Ukip—3 (1-4)
Others—6 (4-8)

Our current central projection remains one of a Labour government supported by multiple parties. On that projection, a Labour-Liberal Democrat coalition with informal support from the SNP would have 355 seats – exactly the number held by Labour alone after the 2005 election.

This nicely illustrates the fragmentation of British politics since that time – the parliamentary power held by one party in 2005 would, on our projections, be shared between three parties in 2015.

Polling Observatory are the academics Robert Ford, Will Jennings, Mark Pickup and Christopher Wlezien. Full Polling Observatory archives are now available at pollob.ca.