6 December 2012

Predicting an election


Correct as of 1852 GMT, 07-Dec-12


Deep in the underground headquarters of Benjamin's Blog, the political department sprint to get to work. They have used Excel to produce a rough prediction of what the election will entail. The psychology department ceased on their topic of "what pregnancy does to a person's mind" and tried to wonder how the results have come about. The maths and statistics departments checked over and over. All were still shocked, and when I, the editor, entered the room, I was greeted by a great sense of awe. "Are you sure?" I asked. "Yes," came the reply. It's such a confusing story that I'm running it.

If you follow my Twitter account, @TheOtherRB, you will find that I make predictions about what will happen in by-elections, and usually, they turn out to be true. Those who know me best will know that I do "exit polls" in anything in school remotely like an election, and, again, they're usually correct. It's a shame that my impressions of Peter Snow at the swingometer leave a lot to be desired (for those of you who don't know, go on YouTube and watch Election '92 and look for Rory Bremner's Peter Snow impressions... if you can stand all fourteen hours). So when I was told of this prediction, I too was sceptical.

Here, then, is our prediction. Bear in mind that a lot can happen in two years, and that there is a surprising result.

Liberal Democrats: 28% (+4%)
Labour: 28% (-2%)Conservatives: 26% (-11%)
UKIP: 17% (+14%)
Others: 1%. (-6%)

Our political map which you saw in the last post cannot make UKIP predictions, although, I daresay it soon will. Nonetheless, our map has churned this up:

Labour: 297 (+39)
Conservatives: 208 (-98)
Liberal Democrats: 115 (+58)
Others + UKIP: 30 (+1)

Here's our map. The prediction is on the left, the 2010 result is on the right.


The one other gain is Wyre Forest from the Conservatives. Look at how the Liberal Democrats flood in. There's really going to have to be a coalition if that is the case.

Other notable seat changes:

Hampstead and Kilburn, the tightest three-way marginal in the country, would fall to the Liberal Democrats.
Liberal Democrats clean sweep Cornwall.
The seat which Zac Goldsmith has indicated he would give up for Boris Johnson would go yellow.


12.5% swing from CON to UKIP.
8% swing from LAB to UKIP.7.5% swing from CON to LD.
5% swing from LD to UKIP.4.5% swing from CON to LAB.
3% swing from LAB to LD.

So I can beg that the question on your lips is "how?". This takes every opinion poll up until now, adding in an Excel "trendline" (or ten, due to two graphs for better accuracy). It capitalises on a Lib Dem surge, which is occurring at the moment. OK, it's small, but it's still happening. This could be what drives the Lib Dems forward.

Labour are faltering (again!) in the winter, leading to their long-term decrease (according to our graphs). The Conservatives would gently fall away. UKIP would gently gain, but too gently to really be a factor. The others, would fall away.

Election 2015 is years away, but it promises to be exciting.

No comments:

Post a Comment