I’ve proposed a stochastic voting system. This weekend I wrote a program to simulate a stochastic election based on 2004 election data. And the new 2004 election results are in:
| Party | Seats |
|---|---|
| Liberal | 120 |
| Conservative | 96 |
| N.D.P. | 48 |
| Bloc Québécois | 36 |
| Green Party | 7 |
| Independent | 1 |
Results by province and by riding are available. Note: This spreadsheat contains two sheets.
One hundred and fifty-nine elected candidates differ from the actual 2004 election outcome; that seems around the expected result.
This is only one example of the results of a stochastic election. Because of the stochastic nature of the election process, actual results may differ.
In Canada’s election process, it is sometimes advantageous to not vote for one’s preferred candidate. The stochastic election system is the only system in which it always best to vote for your preferred candidate. Therefore if the 2004 election were actually using a stochastic election system, people would be allowed to vote for their true preferences. The outcome could be somewhat different than what this simulation illustrates, since I don’t have data for voter’s true preferences.
In the stochastic election system, every vote counts, whether you vote Liberal or Libertarian.
Although the Green Party has 7 seats, it is Bloc Québécois that hold the balance of power in this government.
Stay tuned for 2006 stochastic election results.