Following my blog earlier this week on historic election spend, I’ve been playing with a predictive model using just the amount of donations to political parties in the 2015 election as the basis for prediction. I wanted to get it ‘out there’ before results came in. It really is just a bit of fun, and I had to include a ‘Lib-Dem Suicide Factor’ where the popular perception is that they didn’t fulfill their 2010 election promises, so there will be a lot lower number of people voting for them than would otherwise.
The first four columns are actual data from the Electoral Commission and Wikipedia. Firstly there is the predicted number of votes based on money spent in the campaign, then if we were a proportional representation (of some type) democracy then a prediction of the number of seats. Finally and even less certainly is a prediction of number of seats on the basis of ‘first past the post’. This is done on a very broad brush way, not considering individual constituencies margin for changing and individual personalities.
|Political Party||Number of Candidates||Donations (£k)||£ / Candidate Fielded||Predicted Number of Votes||Predicted Number of Seats (Proportional Representation)||Predicted Number of Seats (First Past the Post)|