Electoral Calculus made vote share and seat predictions for the final elections from Britain to the EU Parliament in May 2019. We used our regression analysis of our own commissioned polls, on behalf of Remain United.
This election had some difficult features to predict, particularly the massive flows of voters away from the two larger parties towards the 'Brexit extremes' of the Liberal Democrats and the Greens, and the Brexit party. There was also the likelihood of significant amounts of tactical voting to add more complexity.
Our model used regression analysis with additional modelling on top of that to allow for tactical voting.
In this difficult context, we were pleased that there were many positive features of our prediction, but we didn't get everything right.
We correctly predicted that the Lib Dems would be the leading pro-Remain party in every English region, and would be ahead of the Green party. This prediction was challenged and disputed by Green party supporters, but was accurate. We also correctly predicted that the Brexit party would come top, and that Change UK would do badly.
Figure: Predicted versus actual vote share at the EU Elections 2019
For the large parties, we correctly predicted that their vote shares would decrease significantly, but we under-estimated the extent of Labour's collapse. Nevertheless, the regression analysis was generally better than conventional polling analysis at predicting the regional results.
Full details of our EU predictions, and find out about our regression polling and election prediction services.
CASE STUDY
Electoral Calculus made the most accurate pre-poll prediction of the result of the December 2019 General Election. Our final prediction correctly predicted a Conservative victory with a substantial majority.
CASE STUDY
Electoral Calculus undertook a project to create a political segmentation of the UK voting population by political attitudes. This has become more relevant in recent years since the traditional single left-right axis no longer captures the reality of the public's political attitudes.