CEO, Pallas Data. PhD, 2013, London School of Economics. Opinions are my own, RT != Endorsements
Toronto, Ontario
Joined February 2012
- Tweets4.1K
- Following1.1K
- Followers1.9K
- Likes2.5K
Very good explainer from someone who clearly understands basic statistics. I'd add that, based on my polling at the time this poll was released, Chow was leading by 7.
This was never a competitive mayoral race.
First of all, when calculating accuracy for polls you have to check if it falls within the margin of error.
For example: Forum’s final poll on Oct 25th (53-43) was in the +-4 MoE and the final result of 54-41.
So what’s the margin of error and why does it matter, then? In 95% of cases if you ran the poll again, it would produce a result within 4 points each way. The MoE shrinks with increased sample size. Simply put, you can say with 95% confidence that the true result is contained in that space.
If the actual result fell outside of the MoE at 95% confidence level it would be ~2 standard deviations away from the mean.
Build Canada’s October 16th poll (Bradford +4.7) had an MoE of +-3.1. The polling miss of 17.1 points (for the margin of victory) meant that the actual result was nearly 5.62 standard deviations away from the poll’s mean.
If this was just because of pure random sampling luck, this has a 1 in 51 million chance of occurring.
The poll did not reflect the voting public that showed up on October 26th. To attribute this miss solely to voter preferences swinging double digits in a week is an excuse for poor methodology.
Joseph Angolano retweeted
First of all, when calculating accuracy for polls you have to check if it falls within the margin of error.
For example: Forum’s final poll on Oct 25th (53-43) was in the +-4 MoE and the final result of 54-41.
So what’s the margin of error and why does it matter, then? In 95% of cases if you ran the poll again, it would produce a result within 4 points each way. The MoE shrinks with increased sample size. Simply put, you can say with 95% confidence that the true result is contained in that space.
If the actual result fell outside of the MoE at 95% confidence level it would be ~2 standard deviations away from the mean.
Build Canada’s October 16th poll (Bradford +4.7) had an MoE of +-3.1. The polling miss of 17.1 points (for the margin of victory) meant that the actual result was nearly 5.62 standard deviations away from the poll’s mean.
If this was just because of pure random sampling luck, this has a 1 in 51 million chance of occurring.
The poll did not reflect the voting public that showed up on October 26th. To attribute this miss solely to voter preferences swinging double digits in a week is an excuse for poor methodology.
When you look more broadly than one cherry-picked poll for comparison, the polling story of this election is actually quite interesting.
The pattern across all 11 of the final week poll results shows that the older the fieldwork, the bigger the miss. This points to a significant late stage shift in sentiment back towards Chow, not bad polling.
We polled the Toronto mayoral race in the field from Oct 11-16 and found Bradford +4.7. The final result had Chow +13 on the night. That means out of the 11 final-week polls, ours was the farthest off. If I had a do-over, we should definitely have also run a Wave 2 poll Oct 24-25.
I'm confident in our methodology, which is the same as many of the other pollsters on this list, and has been reviewed and validated by CRIC (the industry standards body for public opinion research).
I'm excited to continue to scale Build Canada's polling operations. We have big plans to do so. And I'm even more excited to be able to field polls at a price that democratizes access to public opinion research.
We're just getting started.
#cdnpoli #topoli
Polymarket just double-counts polls.
My friend pointed out a 2% spike for Chow after I released my poll on Sunday night.
Bang on. At no point did I have Bradford leading or close to leading. This was never really a contest. Olivia Chow was comfortably ahead the whole time.
Too many media want the easy - already written up - story....
Like, the BS Polls suggesting that Brad Bradford was tied and even ahead of Olivia Chow.
Too many media ran with it - any poll and press release - already done for them.
Based on the polling I had, Bradford was 10-points ahead with males over 54 years of age and only 3-points behind with females over 54 years of age.
For Chow to have won by 13-points, clearly means Chow had a much higher quality list of "identified supporters", with names, addresses & phone numbers - and they were able to get them out to vote to overcome the advantage Bradford had with 55+ voters.
Since our founding, we've dropped polls the night before E-Day six times.
We took the top spot in three of them (BC 2024, BC 2026, Toronto Mayoral 2026). And we missed the top spot in Quebec by 0.5%.
@pallasdata: We are very, very good at what we do.
Our final Toronto poll: @oliviachow 53, @BradMBradford 43.
The result: Chow 54, Bradford 41.
Number one. First with the most accurate result in the Toronto mayoral race. #TOpoli
Thank you Dean!
A big congratulations to @JosephAngolano and Pallas Data — quickly becoming one of Canada's best and most reliable polling firms.
They had the closest results in BC and Toronto, and were almost perfect in Quebec.
Our final @pallasdata #topoli poll is out.
Most likely scenario: Olivia Chow will be re-elected tomorrow, but GOTV will be a big factor, as it always is in municipal races with lower voter turnout.
🚨🚨🚨 And that's a wrap! Our final Toronto poll for this election
@oliviachow - 53%
@BradMBradford - 43%
#TOpoli
In this column, Eric ranks @pallasdata as the best performing firm in last night's election in BC.
What happened in B.C.'s election, plus how the polls and the projection did: thewrit.ca/p/conservative-la…
Joseph Angolano retweeted
What happened in B.C.'s election, plus how the polls and the projection did: thewrit.ca/p/conservative-la…
Joseph Angolano retweeted
So it seems we are ready to declare a winner among the pollsters
And it's Pallas. Only firm with all 3 parties within the MoE. Pollara (using their numbers adjusted for the list of candidates) has a lower average error though but I feel getting all 3 parties within MoE is more important
Our final @pallasdata #bcpoli poll is out.
Tomorrow's most likely outcome: @Conservative_BC will form the next government.
🚨🚨🚨 And that's a wrap! Our final BC poll for this election.
@bcndp - 39%
@Conservative_BC - 45%
@BCGreens - 12%
#bcpoli
The Mississauga mayoral race is a coin flip with four days to go. Either Parrish, Crombie, or Tedjo can win this race.
It all comes down to turnout. Which voters are the most motivated, and which campaign has the best GOTV operation?
Game on.
🚨🚨🚨 New Mississauga poll
We've found a three-way battle for the top spot, with only four days to go. 🔽🔽🔽
pallas-data.ca/2026/10/22/pa…
