This week’s forecast is by far Labour’s best since our model launched 12 months ago. We now give them a 59% chance of emerging from the General Election with the most MPs – up from 49% last week, and surpassing their previous high of 52%.
The chances of a majority Labour government are also the highest yet, at 30% – up from 23% last week.
But it’s not because of some great poll bounce for Labour, perhaps brought on by their conference in Manchester last week. In fact, they haven’t moved at all in our polling average* – still on 36%, right where they’ve been for most of the last six months.
Instead, Labour’s improvement in our forecast comes on the back of a swing from the Conservatives to UKIP. The Tories are down 2 points to 31%, equalling their worst share in the polling average for the last year. Meanwhile UKIP, following their conference and the defection of Mark Reckless, are up 3 to 16%.
With their poll share down and time to catch up ticking by, the Tories get their worst forecast yet. Their chances of winning the most seats are down to 41% (from 51% last week), and their chances of a majority are down to 17% (from 25%).
We should treat this shift in the forecast with caution, however. Barely any of the polling in this week’s average was conducted after the Conservative conference in Birmingham. It’s worth waiting until the conference season is over to see where the polls – and the forecast – end up.
With the two largest parties now forecast to secure just two-thirds of the vote, the likelihood of a Hung Parliament is up to 53% (from 51% last week), the highest since our forecast began.
Our central forecast is also Labour’s best and the Tories’ worst so far – but it’s still very close and well within Hung Parliament territory. In that scenario, the Tories win the most votes, though by the narrowest margin in our forecast so far (34.1% to 32.5%). Labour get the most seats – 306 to the Tories’ 288 – but are left 20 short of a majority, while the Lib Dems win 28.
*Methodological note: We are now generating our own polling average. Full details of the method to follow.
Date of forecast: 3 October 2014
Days till the election: 216
Inputted current average poll shares
Others (inc. UKIP): 25%
– UKIP: 16%
Forecast Election Day Shares (with 95% Prediction Intervals)
Con: 34.1% (±7.2, i.e. 27% – 41%)
Lab: 32.5% (±5.3, i.e. 27% – 38%)
LD: 11.4% (±7.5, i.e. 4% – 19%)
Implied point estimate shares for:
– Others (inc. UKIP): 22.0%
– UKIP: 14.1%
Forecast Election Day Seats (with approximate 95% Prediction Intervals)
Con: 288 (223 – 374)
Lab: 306 (214 – 374)
LD: 28 (21 – 33)
(Prediction intervals assume LD & others shares at central forecast, Con & Lab shares vary as per prediction intervals)
Central forecast: Lab largest party, but short of a majority by 20
Approximate probabilities of key outcomes
Con largest: 41%
… with a majority: 17%
Lab largest: 59%
… with a majority: 30%
Hung Parliament: 53%
… with Con largest: 24%
… with Lab largest: 29%
(probabilities may not sum due to rounding)