Place & Politics in Toronto - Part 10

The 2026 mayoral election: What kinds of neighbourhoods voted for whom?

Zack Taylor, Aniket Kali, and Jeff Allen

November 10, 2026

Chow's vote share

Bradford's vote share

Boundaries

% of households that rent

Median age of residents

% of population that is a visible minority

% of commutes by car

Median household income

% of dwellings that are single-detached

Census tracts (2021 boundaries) are shaded in four equal groups (quartiles), from the lowest values (lightest) to the highest (darkest). Vote shares are shares of valid 2026 election-day votes, apportioned from polls to tracts; advance and mail-in votes are not included. Neighbourhood characteristics are from the 2021 Census of Population.

Chow and Bradford appealed to different kinds of voters. The Star’s Vote Compass project found that people who favoured one or the other candidate seemed to live in “different cities”, with Chow supporters being most concerned about housing and transit, while Bradford’s reported property tax rates and crime and policing as their most important issues. As our social geography correlation tool shows, this variation is reflected in their geographic bases of support. Each candidate tended to receive support from different types of neighbourhoods – reflecting deeper patterns across the city.

Transportation behaviour gives a strong hint about what separates neighbourhoods that support Chow and Bradford. Commuting by bicycle or walking, and neighbourhood walkability are among the strongest correlates of neighbourhood support for Chow. By contrast, commuting by car is Bradford’s strongest.

Chow’s neighbourhoods are also disproportionately home to younger, “creative class” types – people who work in arts and culture occupations, are unmarried or living common-law, and rent. Bradford’s neighbourhoods are where “homevoters” live: people who are married, live in single-detached houses, and have higher household incomes. This may be interpreted as something of a class divide. Bradford’s geographic coalition supports a narrative of pocketbook voting: people voting for low taxes and protection of property and property values. The characteristics of Chow’s coalition are consistent with the prioritization of preserving and expanding collective amenities.

Neighbourhood characteristics most correlated with each candidate's vote

Top ten correlations across census tracts with 2026 election-day vote share

Correlation with Chow's vote share

1 Marital status: Common-law 0.62
2 Commute: Bicycle 0.59
3 Occupation: Arts, recreation, sport 0.59
4 No religion 0.55
5 Commute: Walk 0.52
6 Marital status: Never married 0.45
7 Rented dwellings 0.39
8 Population density 0.39
9 Population age 30–64 0.37
10 Commute: Transit 0.36

Correlation with Bradford's vote share

1 Commute: Automobile 0.61
2 Marital status: Married 0.61
3 Single-detached dwellings 0.58
4 Household income: Median 0.44
5 Owner-occupied dwellings 0.42
6 Occupation: Health 0.41
7 Household income: Average 0.41
8 Average dwelling value 0.39
9 Religion: Jewish 0.38
10 Religion: Catholic 0.36

Each number is a correlation across the city's 585 census tracts, not a share of the vote: 1 means a characteristic rises and falls exactly with the candidate's vote share, and 0 means no relationship. Both lists share one scale, from 0 at the left of each line to 1 at the right. Shown are the ten highest of 85 characteristics (2021 census variables and distance from City Hall). Tracts count equally, whatever their size. Vote share is the share of valid election-day votes; advance and mail-in votes are not included. These are associations between places, not evidence about how any group of people voted.

Growing the pie

In the previous post we noted that Chow won with a “core-plus” strategy – building a coalition of neighbourhoods from the centre outwards toward the periphery. This was not only the result of a greater number of votes from higher turnout, but also a substantial increase in the percentage of votes for her as well.

Where did all of these new votes come from? Half of them came from census tracts that she won in 2023 largely in Old Toronto, East York, and northern Scarborough. But a substantial number also came from parts of the city that she flipped, largely in the rest of Scarborough and along the Weston Road corridor. Moreover, even places that Bailão won in 2023 which stuck with Bradford in 2026, she increased her share of the vote from 21.7% to 39.3%.

Who led each tract in 2023 and 2026

Chow both timesFlipped to ChowOpponent both timesOther

Census tracts by who led, Chow or her main opponent, in each election; black lines are the former municipalities. Other is two tracts Chow led in 2023 but not 2026, and four exact ties. The bars show the votes Chow added from 2023 to 2026 in each kind of tract: 215,705 in all, 1,394 of them in the other tracts. Votes cast rose 32%, so part of each gain is higher turnout. Mapped poll votes only (about 78% of valid votes in 2023 and 76% in 2026); advance and mail-in votes are excluded.

Who exactly are these new supporters? For the most part, they resemble the same kinds of voters that Chow won in 2023: transit and bicycle riders, renters, young people, and lower income residents. Across the city, her increase in vote share was tied to neighbourhoods with a higher proportion of renters, and this is especially prevalent in Scarborough. In one sense, this isn’t very surprising – Chow has pursued numerous pro-renter policies including a renoviction bylaw, increased support to RentSafeTO, and remedial action at 500 Dawes Road. But in another sense, it reflects the growth of half of the city as a political constituency in its own right, overcoming its historical association with low turnout.

At the same time, some of her weakest gains were in neighbourhoods with a high proportion of visible minority residents. Conversely, this is where Bradford began to make inroads, adhering to a broader trend of an openness to right-wing politics from visible minorities, or plain abstention, which School of Cities research has previously highlighted.

Neighbourhoods where each candidate did best

Their top fifth of census tracts by 2026 vote share; dotted lines are medians

Chow's top fifth Bradford's top fifth All other tracts

Each dot is one of the 585 census tracts (2021 census). Each top fifth is the 117 tracts with the candidate's highest share of valid 2026 election-day votes: Chow 66.8% or more (tract median 53.5%) and Bradford 52.3% or more (40.9%). Dotted lines are unweighted medians of tract values. Not drawn, but included in the medians: 24 tracts with incomes above $160,000 or below $40,000 and 9 tracts with median ages outside 30 to 60. Advance and mail-in votes are excluded. These describe neighbourhoods, not how any group voted.

Identity voting?

Toronto is a highly ethnically diverse city, and so we might expect to see ethnic affinity voting for Chow. In 2023, and again in 2026, pundits suggested that as a Chinese woman, Chow would potentially attract support from the city’s large Asian community. In fact we find this to be correct, if weakly. Neighbourhood vote shares for Chow are weakly correlated with neighbourhood Chinese population share in 2014 and 2026, regardless of location. In 2023 there was a stronger correlation in suburban areas, but a weaker one in 2026. Without individual-level evidence from surveys we cannot definitively explain this pattern. Nevertheless, we may conclude that Chinese ethnic affinity voting likely occurs, but it is a marginal factor, and was more present in 2023 than in 2026 or 2014.

Chow's vote share in neighbourhoods with more Chinese residents

More than 20% Chinese All other tracts

2014

Old Toronto 35% 42%

Suburbs 17% 19%

2023

Old Toronto 43% 51%

Suburbs 28% 34%

2026

Old Toronto 65% 72%

Suburbs 48% 49%

Chow's share of election-day votes in the census tracts where more than 20% of residents are Chinese (visible minority, 2021 census) and in all other tracts, in Old Toronto and in the suburbs (everywhere outside the former City of Toronto). Only 17 Old Toronto tracts are over 20% Chinese. The correlations with the Chinese share of residents in 2014, 2023 and 2026 are 0.20, 0.24, 0.23 in Old Toronto and 0.19, 0.43, 0.12 in the suburbs. Advance and mail-in votes are excluded. These describe places, not how any group voted.

In the context of the polarized debate on Israel and Palestine, the media reported criticism from the Jewish community that Chow was insufficiently supportive of them, while others have pushed back on that criticism as relying on anti-Palestinian tropes. Bradford called her out for not showing up to a Walk for Israel and a debate on antisemitism, but also came under criticism for courting Islamophobes.

Do we see any evidence of Jewish neighbourhoods disproportionately supporting Bradford or Muslim neighbourhoods disproportionately supporting Chow? Again, we have weak evidence that this is the case. As the proportion of the neighbourhood population that is Jewish increases, that neighbourhood was more likely to support Bradford over Chow. The corresponding trend for neighbourhoods by Muslim population is much weaker.

Chow's and Bradford's vote share in neighbourhoods with more Jewish residents

Chow Bradford All others

More than 20% Jewish24 tracts 34%64%

All other tracts559 tracts 56%40%

Share of election-day votes in the census tracts where more than 20% of residents are Jewish (2021 census) and in all other tracts. Across all tracts, the correlation with the Jewish share of residents is −0.32 for Chow and 0.38 for Bradford. Advance and mail-in votes are excluded. These describe places, not how any group voted.

Chow's and Bradford's vote share in neighbourhoods with more Muslim residents

Chow Bradford All others

More than 20% Muslim40 tracts 60%33%

All other tracts543 tracts 54%41%

Share of election-day votes in the census tracts where more than 20% of residents are Muslim (2021 census) and in all other tracts. Across all tracts, the correlation with the Muslim share of residents is 0.07 for Chow and −0.16 for Bradford. Advance and mail-in votes are excluded. These describe places, not how any group voted.

Plus ça change…

In the end, the geography of the 2026 election, along with the limited polling data reported in the media, suggests that the contest hinged on divergent neighbourhood interests rather than on other potential divides, including ethno-racial group conflict. The 2026 election is not unique. With the three-way 2014 election being the principal exception to the trend, we find that transportation behaviour and housing type and tenure, which are in turn related to distance from City Hall, are powerful positive or negative correlates of support for winning mayoral candidates across time. Toronto politics is to a large degree a politics of mobility and location – where people live, the nature of their relationship to housing, and how they move through the city.

Absolute correlations between neighbourhood traits and the winner's vote, 1997–2026

Absolute Pearson correlations across census tracts, unweighted, between the winner's share of election-day votes and each measure, using the census nearest each election. They show strength, not direction. Apart from Tory in 2014 (negative for car commuting and distance), four measures are positive for the conservative mayors and negative for Miller and Chow; the share of households that rent runs the other way. Tracts, not voters.

Data and methods

All data are for Toronto’s 585 census tracts, on 2021 boundaries. Official election results and the voting subdivisions (poll catchment areas) were downloaded from the City of Toronto, and election results and census characteristics were each apportioned to these tracts. Everything here describes neighbourhoods, not how individual voters, or any group of people, voted. The code and data used to build this page and its graphics are on GitHub.

Elections: poll-level mayoral results, 1997 to 2026, split into dissemination blocks by population and summed to 2021 census tracts. Only election-day votes are included, as advance and mail-in votes can’t be placed in a neighbourhood. The map and the votes Chow added use the polls that can be mapped, about 78% of valid votes in 2023 and 76% in 2026, so the 2023 baseline is not the official result, in which Chow led Ana Bailão 37.2% to 32.5%.
Census: Statistics Canada’s 2021 Census of Population, including median age. The chart of the winners’ vote over time uses the census nearest each election, 1996 to 2021; data from 1996 to 2016 were apportioned to 2021 census tracts using the Canadian Longitudinal Census Tract Database (Allen and Taylor, 2018).
Correlations: Pearson coefficients across tracts, unweighted, between a candidate’s share of valid election-day votes and a tract characteristic.
Boundaries: census tracts from Statistics Canada and former municipalities from the City of Toronto, as in other posts in this series.