UK 2015 – Forecast – 19th January 2015


This is the latest of our forecasts for the UK General Election in 2015. To see our methodology click here. To read a note on our updated forward looking projection methodology, click here.

Forecast

Our mid point and confidence interval forecast for seats is as follows.

Party Seats – Mid Point Seats 50% CI Seats 95% CI
Conservatives 290 270 – 300 247 – 325
Labour 288 275 – 300 244 – 318
Liberal Democrats 9 6 – 11 4 – 21
UKIP 17 3 – 29 0 – 72
SNP 21 17 – 29 10 – 39
PC 4 2 – 4 2 – 4
Others inc. Green (GB) 3 1 – 7 0 – 15
Irish 18 n/a n/a

Our mid point and confidence interval forecast for votes (mainland GB) is as follows.

Party Vote% – Mid Point Vote% 50% CI Vote% 95% CI
Conservatives 30.5 29.3 – 31.9 27.5 – 34.5
Labour 29.9 28.3 – 31.3 25.2 – 34.7
Liberal Democrats 11.1 10.2 – 12.1 8.6 – 13.3
UKIP 16.5 14.3 – 18.6 10.3 – 22.9
Green 5.9 4.4 – 7.4 1.4 – 10.7

Our probability of different outcomes is as follows.

Event Probability
Conservative Majority 0.7%
Conservative Minority 40.3%
Exact Tie Labour and Conservative 0.9%
Labour Minority 57.7%
Labour Majority 1.1%
UKIP more votes than Liberal Democrats 90.5%
Greens more votes than Liberal Democrats 3.2%
Greens more votes than UKIP 0.8%
Green > UKIP > LD (votes) 0.1%
LD > Green > UKIP (votes) 0.8%
LD > UKIP > Green (votes) 8.8%
UKIP > Green > LD 3.1%
UKIP > LD > Green 87.3%
UKIP get more votes than Labour 0.2%
SNP largest party in Scotland (seats) 32.9%

What are the main points of your forecast?

We identify the following events / features.

  • There is now increasing evidence that the Conservative vote is not moving up as we would normally expect in the run up to an Election. This means our expectation of the Conservative’s final position has been dampened.
  • Although the Conservative lose seats to Labour, Labour lose a significant number in Scotland to the SNP, all but destroying their chance of a workable plurality.
  • The Liberal Democrats will struggle to poll significantly above 11%. This will have a dramatic effect on their attempts to retain seats, even with the enhanced Liberal Democrat incumbency effect.
  • The SNP are set to at least double their seats in Scotland. If they maintain their current poll rating up to the General Election (at the moment we are expecting the SNP support to fall back in the next few months) we expect them to even outperform Labour on polling day in Scotland.
    At the moment we are weighting down the current surge in SNP support. From our next forecast we will begin to increase their support as we become more sure of the sustainability of the increase in support.
  • UKIP support will be significant. The lack of any evidence of a rise in support for the Conservatives means that UKIP could possibly make some dramatic gains. Our analysis shows that as the level of support for the Conservatives drops to around 30%, UKIP become significant challengers in many seats, even if they poll less than 20% national share.
  • Wales – We have improved our modelling for Wales and this is reflected in the figures above.

The key aspect of our forecast is uncertainty. When the levels of support for the two main parties drop to around and below 30% each, the variability of scenarios increases. We forecast 17 UKIP seats, but the reality is that it is very hard to accurately assess the chances of UKIP in up to 50 seats.

Why is your forecast different to that of Stephen Fisher?

We use the same fundamental forecasting principle as Professor Stephen Fisher with the following crucial differences.

  • We use a different sample of elections to model movement of party support prior to a national vote
  • We specifically model UKIP support and do some additional local analysis on where UKIP might do well
  • We have regional models which help to capture specific local peculiarities (for example, we model Scotland independent of the rest of Great Britain, allowing us to capture the current spike in support for the SNP – Stephen Fisher currently does not do this).
  • We have used some components of this methodology before and that allows us to make corrections as we compare our outcome to real results.

What improvements are you hoping to implement next?

At the moment the model isn’t taking into account polling in individual seats. As such data becomes richer and more

Leave a Reply