Friday 28 November 2014

#GE2015 Forecast update: 28 November 2014

Forecast b 141128
Another mixed bag of polls this week: cherry-pick your favourite and you can claim anything from a 5-point Labour lead to a 1-point Conservative one. Lump them together in our polling average, though, and the result is a Labour lead of 1.
The Tories have lost the point they gained last week, taking them back down to 32%, while Labour remain on 33% for the fifth week in a row. UKIP are the beneficiaries of the Tory fall, up a point to 16% – perhaps as a result of their second by-election win last Thursday.
In our forecast, that means a small shift from the Tories to Labour. The Conservatives are still the slight favourites to be the largest party in May, but their chances are down from 58% last week to 53% now. Labour’s are up from 42% to 47%.
Our central forecast tightens again: it now has the Tories winning the popular vote by 3.5 points, enough to give them 5 more seats than Labour (298 to 293).
With the forecast tightening and time running out for one party to break ahead, the chances of a Hung Parliament are now up to 57%.

Date of forecast: 28 November 2014
Days till the election: 160
Inputted current average poll shares
Con: 32%
Lab: 33%
LD: 8%
Others (inc. UKIP): 27%
– UKIP: 16%
Forecast Election Day Shares (with 95% Prediction Intervals)
Con: 34.4% (±6.7, i.e. 28% – 41%)
Lab: 30.9% (±5.0, i.e. 26% – 36%)
LD: 10.9% (±6.7, i.e. 4% – 18%)
Implied point estimate shares for:
– Others (inc. UKIP): 23.8%
– UKIP: 14.1%
Forecast Election Day Seats (with approximate 95% Prediction Intervals)
Con: 298 (226 – 380)
Lab: 293 (216 – 361)
LD: 28 (23 – 34)
(Prediction intervals assume LD & others shares at central forecast, Con & Lab shares vary as per prediction intervals)
Central forecast: Con largest party, but short of a majority by 28
Approximate probabilities of key outcomes
Con largest: 53%
… with a majority: 23%
Lab largest: 47%
… with a majority: 20%
Hung Parliament: 57%
… with Con largest: 29%
… with Lab largest: 28%
(probabilities may not sum due to rounding)

No comments:

Post a Comment