Ontario Polling Tracker
Public voting-intention polls for Ontario's next provincial election. Hover for a snapshot, click anywhere on the chart for the nearest poll's full details.
Showing 6 of 45 polls. Compiled from publicly released polls; each poll attributed to its pollster. Shaded bands show each poll's margin of error.
Compare pollsters
Pick pollsters in the Filters panel (top right) — or click any row below — to focus the chart and highlight their rows. All figures cover the selected date range (May 26, 26 – Aug 24, 26).
Averages & lean vs. the field, by pollster
| Pollster | Polls | Last poll | PC | OLP | NDP | GPO |
|---|---|---|---|---|---|---|
| All pollsters | 6 | — | 36.2 field | 33.0 field | 23.2 field | 3.5 field |
| Liaison Strategies | 3 | Aug 24, 26 | 35.7 — | 37.0 — | 22.0 — | 2.7 — |
| Abacus Data | 1 | Jun 17, 26 | 41.0 — | 32.0 — | 19.0 — | 5.0 — |
| EKOS Research | 1 | Jul 16, 26 | 32.0 — | 26.0 — | 28.0 — | 5.0 — |
| Pallas Data | 1 | Aug 9, 26 | 37.0 — | 29.0 — | 26.0 — | 3.0 — |
Top figure: unweighted mean of the pollster's published toplines in the selected range. Below it: how far the pollster runs above (+) or below (−) other pollsters' polls fielded within two weeks of theirs — so nobody is called high or low just for polling during a surge. Bold ▲/▼ marks a consistent lean of at least a point across three or more polls; dashes mean too few contemporaneous polls to compare.
Bias map
Each dot is a pollster's mean lean for one party, in points against other pollsters' polls fielded within two weeks. Dots right of the line run high, left run low; a gold ring marks a consistent lean of at least a point across three or more polls. The party chips above the map hide or show parties; clicking a row filters the main chart to that pollster. Filtering by method — in the Filters panel, top right — turns this into a within-mode bias map.
Compare methodologies
The same comparison, grouped by how polls are conducted. The method filter in the Filters panel (top right) scopes the chart and every table; clicking rows here does the same. These tables always cover every method in the selected date range, so the comparison keeps its full context while you filter.
Averages & lean vs. the field, by method
| Method | Polls | Last poll | PC | OLP | NDP | GPO |
|---|---|---|---|---|---|---|
| All methods | 6 | — | 36.2 field | 33.0 field | 23.2 field | 3.5 field |
| IVR | 4 | Aug 24, 26 | 36.0 — | 35.0 — | 23.0 — | 2.8 — |
| Online | 1 | Jun 17, 26 | 41.0 — | 32.0 — | 19.0 — | 5.0 — |
| Telephone | 1 | Jul 16, 26 | 32.0 — | 26.0 — | 28.0 — | 5.0 — |
Top figure: unweighted mean of the toplines from every poll using the method in range. Below it: how far those polls run above (+) or below (−) polls using other methods fielded within two weeks — the mode-effect analogue of a pollster's house effect. Bold ▲/▼ marks a consistent lean of at least a point across three or more polls.