The Labour Party will narrowly win more seats than the Conservatives – and the Liberal Democrats will be saved from wipe out by the first-past-the-post system, according to a new electoral forecast by Prof Paul Whiteley at the University of Essex, co-director of the British Election Study from 2001 to 2012.
Whiteley’s forecast, based on mathematical modelling, focuses on what happened to seats in previous election rather than the overall share of the vote and is based on a model developed while the British Election Study was based at Essex. This model successfully predicted the outcomes of the 2005 and 2010 general elections.
The new forecast for the 2015 election released to the Guardian shows Labour on 291 seats, the Conservatives 281, the Lib Dems 48 and others 30. The model produces a result that is more optimistic than many others about the performance of the Lib Dems – a losing fewer than 10 seats and sharply at odds with claims that the party could crash to as few as 25 seats.
This would give a Labour-Lib Dem coalition a total of 339 seats and a majority of 14. A Conservative-Lib Dem coalition would also be possible, but with a tiny majority since the two parties would only muster 329 seats.
The Essex forecasting model works by combining the number of seats won by parties in the previous election with voting intentions data from polls conducted six months prior to the election.
Whiteley said that the model’s track record was good: in 2005, the average prediction error for the Conservatives, Labour and Liberal Democrats was 11 seats, and on that occasion the model predicted that Labour would get 358 seats when they actually won 356.
It was slightly less accurate in 2010 with an average error of 18 seats for the three major parties, predicting that the Conservatives would get 293 seats, for example, when they actually won 307.
Whitely said the wild card would be the performance of the SNP. If they gain as many as 15 seats, Labour’s hopes of being able to form an overall majority just with the Liberal Democrats becomes more difficult. But Whitely said he did not think the SNP would do as well as some voting intention polls show, partly because increasingly Scots see the Westminster election as secondary to the Holyrood poll, and he expected turn-out to be lower than in the high profile Scottish referendum.
Whitely acknowledged that the model forecast a relatively large number of Lib Dem seats, but that appeared to be because there is a stronger incumbent effect for Lib Dem MPs than any other parties, partly due to many of their MPs long holding marginal seats.
guardian.co.uk © Guardian News and Media Limited 2010