With little more than a month to go before Toronto’s mayoral by-election, Olivia Chow continues to dominate the polls as the top candidates get set for more debates this week.

According to a new Forum Research poll, Chow is favoured by 34 per cent of decided and leaning voters, while Mark Saunders and Josh Matlow are tied for a distant second, with 12 per cent support each.

Despite her commanding lead, support for Chow dropped two percentage points compared to last week, when Forum conducted a similar poll. Saunders, who was last week’s second place finisher, dropped back six percentage points in this week’s poll.

Mitzie Hunter remains in fourth place, with 10 per cent support among decided and leaning voters, the same as last week, while Ana Bailao and Brad Bradford round out the top six with nine and seven per cent support, respectively.

About one in five of those surveyed were undecided, according to the poll.

“Chow’s lead remains untouchable,” said Dr. Lorne Bozinoff, president of Forum Research.

“Other candidates have yet to connect strongly with voters.”

Forum’s findings are consistent with other recent polling data, which also has Chow as the clear front-runner, however support for the other top candidates varies.

According to Forum’s most recent poll, the most important issues in the mayoral by-election are housing affordability and the cost of living and inflation, consistent with previous weeks.

This new polling data comes before a number of mayoral debates set for this week, including a Wednesday noon-hour debate on housing, which will be broadcast live on CP24, as well as the Toronto Region Board of Trade debate on Thursday evening.

The Toronto mayoral by-election is set for June 26.


The poll was conducted by Forum Research with the results based on an interactive voice response telephone survey of 1,000 randomly selected Toronto residents over the age of 18, 70 per cent by cellphone and 30 per cent on landlines.

The poll was conducted on May 19. Results based on the total sample are considered accurate +/- 3 per cent, 19 times out of 20. Subsample and regional results will be less accurate.