Seats Distribution Election Forecaster

developed by Prof. Werner Antweiler

British Columbia Provincial Election 2013

Guess the probability with which a voter in the past election will vote for a particular party in the forthcoming election. The sum of each row in the table below must be equal 1.000 precisely. Once you have completed your guess of the voter migration matrix, press the FORECAST button. The forecaster program will apply your voter migration matrix to the results from the last election to find out which party will win in each riding in the forthcoming election. Leaving the matrix unchanged will display the results from the last election. You can also display results riding by riding in descending order of the vote share to identify "safe" and "marginal" seats for individual parties. For some elections regional breakdowns are also available.

Show results for individual constituencies:
Predict election outcome for
Press to apply your voter migration matrix.
Press to set the voter migration matrix to the identity matrix.

Methodology:

Elections in Canadian jurisdictions employ a "first-past-the-post" voting system. Thus the prediction of the election result in terms of the seats distribution in the federal parliament or provincial legislature is based on a prediction of the election result in each constituency. The use of a voter migration matrix reflects the notion that voters change their opinion about candidates and parties in a similar way across the entire province. That is, a party that gains or loses vote in one constituency tends to gain or lose votes in other constituencies. In the simplest form, this "swing" can be applied across all constituencies.

Voter migration analysis has to be carried out with consideration of the limitations of this method. First, the method does not reflect the change in population. Some children have reached voting age, some (mostly older) people have died, and other people have moved into or out of a jurisdiction. Second, transitions between voters and non-voters are difficult to predict because public opinion polls only poll the current electorate and typically do not report whether someone has voted in the last election or not, or if they intend to abstain. Third, the voter migration matrix is applied identically across all constituencies or across regions. This ignores constituency-specific factors (e.g., incumbency).