[ethan.l.cecc]

Can't understand why we have three or four (or more) "polling decks" in a polling station? Should be putting just one "polling desk" in a single "polling station". Be it a condo meeting room, school gym, church hall, or vacant store. Even a garage in someone's home (unless the garage is packed with their non-automobile stuff).

Better granularity for data and splits the ridings into more manageable lists of electors. In some hyper rural ridings, I'm guessing that some of the polls are alone since there would be so few people to vote there, there's no reason to split it up. There are only so many places that can host elections (can't be actively used during voting days, have to be reservable AND available on voting days, and have enough space for line-ups), which leads to having A LOT of people vote at one location. If there were only one poll per location, then the lists they would have to go through to confirm your info and that you voted would be insanely long. It's also expensive to operate tons of separate polling stations. My polling station (in Toronto Centre) had 5 desks, and that split up the lines enough that some people were able to vote right away (though my line was very slow due to someone having to fill out a special or provisionary ballot).

I agree, though, that we should generally aim for fewer desks per station, both to make it easier to know which line you need to be in and to get polling stations closer to their assigned electorate, especially in larger (in area) ridings

 

[chrisw]

8 polls in one location was quite common in Markham. There might've been up to 400 people per list.

 

[Midtown Urbanist]

Two big surprises in Ontario were Green MP Mike Morrice losing in Kitchener Centre, and long-time NDP MP Brian Masse finishing third in Windsor West.

This was a level of strategic voting we've seen in modern Canadians elections.

Both went blue which is wild.

Maybe it is part of that shift towards conservatism in post-industrial areas (e.g. Oshawa) - and a general trend elsewhere (sic Rust Belt)

AoD

I think it's about housing prices. There was a huge blue shift across the GTA. In Davenport, the CPC went from 10% in 2021 to 23% in 2025. The NDP collapse allowed the Liberals to stave it off in many cases, but not all.

The obvious answer is that there was severe vote splitting in both ridings. The strategic voting seems to only benefit the Liberals, but Liberal voters didn't get the message for Green and NDP incumbents.

However, in Kitchener Centre and London-Fanshawe (another dropped historically NDP seat) I believe there was a lot of local resentment towards the impact of new arrivals attending Conestoga and Fanshawe Colleges in the area. Enough to energize base and some swing voters to come out in high numbers for CPC anyway, while the vote splitting did the rest.

 

[adma]

adma said:
... when everybody votes in the advance ballot, their vote disappears into a big maw, and that negates the worth of e-day polling-station data.

I'm not sure why that would somehow be a big problem

It's not a big problem for the voter. However, it *is* a big problem for those examining granular polling-station data

Of course, if you're the kind of person who doesn't care less about polling station data as long as you get to vote, then it's all immaterial

 

[adma]

8 polls in one location was quite common in Markham. There might've been up to 400 people per list.

However, in the past provincial election, that one location would have counted as one poll. At least the 8 polls break it all down into manageable portions (and makes for more satisfyingly "granular" polling maps)

 

[adma]

The point of an election is to democratically elect people to represent us. Voter data WRT to numbers, timing, etc. is all available from Elections Canada. Perhaps not as fast as some would like.

The concept of 'an election date' is baked into the Constitution. Anything outside of or before that is intended as accommodation in the Elections Act. The availability of polling places and of polling staff in some communities might be an issue. Local schools and community centres are getting harder to find and might not be available In our poll, the advanced voting location was 20km away in a different community.

I know voter data is available. I'm talking about *polling station* data. When one votes in advance, that doesn't count t/w one's polling station--though it may count t/w a larger advance polling subdivision that encompasses said polling station. To take a case: in the 2021 election, Parkdale-High Park had provision for 158 "regular" polling subdivisions, 35 "400-polls" (typically apartments/condos), and 16 advance polls + special balloting. Voting in advance goes into those 16 advance polls, which are like "megapolls" which don't offer the statistical granularity of the e-day polls.

Fortunately, Elections Canada still goes about things in an old-fashioned paper-ballot/hand-count way w/an ample number of polling subdivisions--but a lot of jurisdictions (including Ontario) have adopted electronic balloting and a "one polling location = one poll" formula, which dumbs down the statistical data *a lot*. Or if you want an example of before and after electronic balloting was adopted, take a look at Alberta's polling maps for the most recent election, vs those for preceding elections

election-atlas.ca - Alberta

If you think the new version--where a riding like, say, Calgary-Beddington is now covered by 9 e-day polling subdivisions where there were something like 70 before--is more "statistically satisfying", you're out of your freaking head.

 

[DirectionNorth]

adma said:
... when everybody votes in the advance ballot, their vote disappears into a big maw, and that negates the worth of e-day polling-station data.

It's not a big problem for the voter. However, it *is* a big problem for those examining granular polling-station data

Of course, if you're the kind of person who doesn't care less about polling station data as long as you get to vote, then it's all immaterial

I know voter data is available. I'm talking about *polling station* data. When one votes in advance, that doesn't count t/w one's polling station--though it may count t/w a larger advance polling subdivision that encompasses said polling station. To take a case: in the 2021 election, Parkdale-High Park had provision for 158 "regular" polling subdivisions, 35 "400-polls" (typically apartments/condos), and 16 advance polls + special balloting. Voting in advance goes into those 16 advance polls, which are like "megapolls" which don't offer the statistical granularity of the e-day polls.

Fortunately, Elections Canada still goes about things in an old-fashioned paper-ballot/hand-count way w/an ample number of polling subdivisions--but a lot of jurisdictions (including Ontario) have adopted electronic balloting and a "one polling location = one poll" formula, which dumbs down the statistical data *a lot*. Or if you want an example of before and after electronic balloting was adopted, take a look at Alberta's polling maps for the most recent election, vs those for preceding elections

election-atlas.ca - Alberta

If you think the new version--where a riding like, say, Calgary-Beddington is now covered by 9 e-day polling subdivisions where there were something like 70 before--is more "statistically satisfying", you're out of your freaking head.

Who cares? The point of an election isn't to gather <<statistically satisfying>> data, it's to enable voters to democratically elect new leaders. If advance polls enable more people to vote, the data is irrelevant.

 

[ethan.l.cecc]

I think @adma 's point is more that advance polls probably should operate the same way as e-day polls, so that analysts can get a better picture of how people voted in advance polls. Yes, an election isn't intended to gather "satisfying" data, but it is an IMMENSE amount of data that should be collected in a way that can be analyzed easily. This data also helps inform future voters: the data from these polls create more accurate polls from pollsters, allows for electoral candidates to target demographics more specifically, and even for Elections Canada to know if there is a need to adjust the size of each poll desk or the number of poll desks at each station. The data isn't irrelevant, it just shouldn't be the reason for whether advanced polls exist or not.


PS. And not to mention, there are a lot of people who DO care about the data.

 

[adma]

Who cares? The point of an election isn't to gather <<statistically satisfying>> data, it's to enable voters to democratically elect new leaders. If advance polls enable more people to vote, the data is irrelevant.

Which is sort of a "as long as you get to a destination, the journey is irrelevant" POV. And a POV which is ultimately...barren.

Besides, my beef is not about advance polls in general, it's about advance polls as a crutch--and ultimately, when the lineups one hopes to avoid on e-day happen in advance voting, it kind of cancels out the anticipated advantage of voting in advance...

 

[UtakataNoAnnex]

However, in Kitchener Centre and London-Fanshawe (another dropped historically NDP seat) I believe there was a lot of local resentment towards the impact of new arrivals attending Conestoga and Fanshawe Colleges in the area. Enough to energize base and some swing voters to come out in high numbers for CPC anyway, while the vote splitting did the rest.

That says more about the Conservatives when folks go to them for bigoted reasons...

 

[adma]

I think @adma 's point is more that advance polls probably should operate the same way as e-day polls, so that analysts can get a better picture of how people voted in advance polls. Yes, an election isn't intended to gather "satisfying" data, but it is an IMMENSE amount of data that should be collected in a way that can be analyzed easily. This data also helps inform future voters: the data from these polls create more accurate polls from pollsters, allows for electoral candidates to target demographics more specifically, and even for Elections Canada to know if there is a need to adjust the size of each poll desk or the number of poll desks at each station. The data isn't irrelevant, it just shouldn't be the reason for whether advanced polls exist or not.


PS. And not to mention, there are a lot of people who DO care about the data.

And remember that it isn't just about candidates targeting demographics more specifically, but neighbourhoods, streets, even buildings more specifically. Not to mention that canvassing, in the door-to-door sense, comes easier when a polling station is of a manageable size--one can focus upon a specific poll, whereas when it comes to provincial-size "megapoll", it's impractical, particularly when one doesn't have past confirmed voter lists to operate by.

And ultimately, when it comes to the stats geek with a knack for urban geography, it can make for a fascinating "electoral psychogeography", those weird blips where one party might get outsized results or strange spot-specific swings happen one election into the next...

 

[vic]

Why can't they just put the advanced poll ballots into ballot boxes marked with the regular poll numbers? Just too many boxes to deal with?

 

[PinkLucy]

That would require unsealing the advanced poll ballot box with the potential for votes to be viewed ahead of election day. Candidate scrutineers could then be present and would have advanced knowledge of voting trends. The current process means no risk of results leading out.

 

[lenaitch]

I will put faith in Elections Canada/Elections Ontario to gather whatever voter pattern data they need (and are allowed to) to hold fair, legal and efficient elections. Whether or not any publicly-available data that they assemble is of any value for mining by political parties or even urban geographers should be irrelevant to them.

Every election is somewhat unique in terms of available facilities, staffing, weather/season, etc. Election officials can only control so much within the parameters of the legislation.

That would require unsealing the advanced poll ballot box with the potential for votes to be viewed ahead of election day. Candidate scrutineers could then be present and would have advanced knowledge of voting trends. The current process means no risk of results leading out.

I recall an article that said a special provision was made in Carleton riding to count the advance boxes in advance in light of the unique situation. Handling a ballot that is several feet long would have its own challenges.

 

[SubHuman]

... even buildings more specifically. Not to mention that canvassing, in the door-to-door sense ...

Living in an apartment building, I know the only ones now who will knock on my door without calling first or using the buzzer from the lobby are salespeople for Rogers or Bell, or election campaign workers, so I know not to answer, particular when there's an upcoming election.