Long Range UK general election forecast

http://electionsetc.com/2014/09/05/forecast-update-5-september-2014/


No change for the main parties in the UK Polling Report Average this week – Labour are still on 36%, the Tories on 33% and the Lib Dems on 8%. UKIP have moved up a point to 15%, perhaps helped by Douglas Carswell’s defection and the resulting media coverage.
The Tories’ continued failure to dent Labour’s three-point lead (it’s been 36-33 for the past five weeks now) means the forecast moves slightly towards Ed Miliband’s party again. We still make the Tories favourites to be the largest party, but its getting narrower with every week that passes without them making up the ground our model expects.
We now give the Tories a 53% chance of being the largest party to Labour’s 47% – a touch closer than last week’s 54-46.
It’s a continuation of the long-term trend we’ve seen since the model launched in October: the polls haven’t swung towards the Tories as history suggested they would, and their chances of winning the election have dwindled as a result. Labour have benefited not because they’ve improved in the polls (indeed, they’re down from 39% in the average 11 months ago to 33% now), but simply because there’s less time for the Conservatives to overhaul them.
Our central forecast also sees Labour improve, but again only very slightly. They’re up to 295 seats, just six shy of the Tories’ 301. The Lib Dems are on 26 seats, which would only give them enough to form a very narrow majority coalition with the Tories, and not quite enough to form one with Labour.

Date of forecast: 5 September 2014
Days till the election: 244
Inputted current average poll shares (from UK Polling Report)
Con: 33%
Lab: 36%
LD: 8%
Others (inc. UKIP): 23%
– UKIP: 15%
Forecast Election Day Shares (with 95% Prediction Intervals)
Con: 35.8% (±7.2, i.e. 29% – 43%)
Lab: 32.3% (±5.4, i.e. 27% – 38%)
LD: 11.6% (±7.9, i.e. 4% – 20%)
Implied point estimate shares for:
– Others (inc. UKIP): 20.3%
– UKIP: 13.2%
Forecast Election Day Seats (with approximate 95% Prediction Intervals)
Con: 301 (223 – 390)
Lab: 295 (210 – 367)
LD: 26 (21 – 33)
(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 25
Approximate probabilities of key outcomes
Con largest: 53%
… with a majority: 27%
Lab largest: 47%
… with a majority: 22%
Hung Parliament: 50%
… with Con largest: 26%
… with Lab largest: 25%
(probabilities may not sum due to rounding)