The Britain Elects Nowcast


Updated: Morning of 08 June, 2017

Our nowcast is a picture as to how the nation would vote according to the polls.

Our model, combining historical data as well as national and regional polling, estimates the probability a party has at holding onto each seat. This forecast is theoretical and does not consider targeting. Regional variation and parliamentary by-election results are, however, accounted for.

The Nowcast


Based on current polling, we expect the composition of the House of Commons to be thus:



Seats likely to change hands


Based on current polling, of the 144 seats we expect are at risk of changing hands, 43 will more likely than not (probability >50%) do so. Mapped below are the constituencies where the probability of it changing hands exceeds 50%, coloured according to the expected winning party.

Table: Probability (by %) of seats changing hands


The table below displays the seats we think are at risk of changing hands, however small.

For those using mobile devices, the table is horizontal scrollable.



Methodology


The methodology combines data from our poll of polls polling model, representative surveys of the regions, constituent nations (Scotland and Wales), constituencies, parliamentary by-elections and historical results. The first layer of our model, simply, operates on a uniform national swing according to our poll of polls. The second layer involves modifications to UNS so as to account for regional variations. The third and final layer involves the randomisation of expected share changes – within the 4pt margin of error – to project whether X, Y or Z party wins a certain constituency in a given number of tests.

Example: According to polling, Party A is expected in Region X to improve upon their support by 4.8pts on the 2015 election result. Constituency Y, which is within that region, should therefore see support for that party improve by between 0.8pts and 8.8pts. The model performs a repeated number of tests in that constituency, randomising an improvement in support between those two shares. These tests determine the probability as to how likely the seat will change hands.

This model does not account for the potential concentration of resources or issues within certain constituencies. A Lib Dem focus on Vauxhall, for example, or a Plaid Cymru one on the Rhondda, will not show up in our forecast.

I am under no illusions that this model is not perfect, but nor is pure UNS perfect, or any other form of forecast for that matter, but I hope that our forecast/nowcast, regardless, serves as an extra set of numbers in snapshotting the state of public opinion for the coming general election.

Our methodology, the TL;DR version:
A uniform national swing, but, not… so, a Non Uniform National Swing? The NUNS model?

With thanks to Ipsos Mori, YouGov, TNS, Panelbase, GfK, ICM, Survation, ComRes, Opinium, ORB, BMG and Lord Ashcroft for the polling used in our model, and the British Election Study for publicly available data.