again: i will be surprised if the conservatives poll above 35%. am i ignoring the data? no - i'm just reading it correctly.

the polling data does not suggest that the conservatives will poll at 38-39%. what it suggests, rather, is that there's a ninety some percent chance that they poll in the range of [34, 43]. polling is never an exact number, but always a range. but, we know from experience that the telephone polling has inherent biases that favour the conservatives. so, a prudent reading of the data suggests they should end up near the bottom of it, and that there's a higher chance of them polling polling around 34-36 than there is of them polling around 42-43.

i'll acknowledge a level of intuition in pegging them at the very bottom of this range. this election has been defined by exceedingly nasty media coverage of the ruling party, which is clearly on the opposite side of the broader electorate. further, doug ford is clearly not fit for the job. in electing doug ford, ontario would be suffering a collective experience of mass idiocy that it is not known for. i would consequently assign most of that error to the liberals out of a shy effect, rather than to the ndp.

might i be wrong, though? sure.

so, let's say i'm wrong. let's say the conservatives get 39% of the vote.

look as i may, i cannot find an example of the conservatives winning a majority government in ontario with under 40% of the vote. harris polled around 45% in both wins - numbers that nobody is talking about. bill davis had majorities at 44% and minorities at 36% and 39%. frank miller had a minority at 37%. if we trust the elections bodies to distribute the ridings fairly, we can only conclude that ontario is less rural today than it was in the late 70s, and that it would be harder to win a majority in the high 30s today than it was then.

looking at federal data, stephen harper got 44% in 2011 and 39% in 2008 - which is the difference between a majority and a minority.

so, even if i concede that the conservatives may be running in the middle of their range, i don't concede that that makes a majority government likely. i would rather argue rather forcefully that it makes a minority government likely.

you can come up with whatever theory you want, but i think it is clear that the empirical data suggests that the conservatives need to be running in the 40s to make a majority government an idea to even talk about, and should be running closer to 45 to make it likely.

again: my reading of the data is that the ndp is on the brink of a sweep in toronto. i think we're looking at an ndp majority government.

but, if i'm wrong, this legislature is more likely to end up looking something like the minority government of the 70s than it is to end up looking like the majority government of the 90s - even if we have no real idea of how doug ford would actually govern.