How accurate the pollsters were
For every pollster: the average difference between its final poll and the result, how many elections that rests on, and the period those polls span.
| Pollster | Average difference | Elections | Period |
|---|---|---|---|
| Liaison Strategies | 0.4 percentage points | 1 too few elections | 2025-04-27 – 2025-04-27 |
| Kolosowski Strategies | 0.7 percentage points | 1 too few elections | 2025-04-19 – 2025-04-19 |
| Pallas Data | 0.9 percentage points | 1 too few elections | 2025-04-26 – 2025-04-26 |
| Cardinal Research | 1.0 percentage points | 1 too few elections | 2025-04-23 – 2025-04-23 |
| Nanos Research | 1.1 percentage points | 7 | 2006-01-22 – 2025-04-27 |
| Léger | 1.1 percentage points | 6 | 2006-01-17 – 2025-04-25 |
| Mainstreet Research | 1.1 percentage points | 4 | 2015-10-15 – 2025-04-27 |
| Research Co. | 1.2 percentage points | 3 | 2019-10-20 – 2025-04-27 |
| Decima Research | 1.3 percentage points | 1 too few elections | 2006-01-15 – 2006-01-15 |
| Angus Reid | 1.3 percentage points | 6 | 2008-10-13 – 2025-04-26 |
| Ipsos | 1.4 percentage points | 7 | 2006-01-22 – 2025-04-26 |
| Harris-Decima | 1.5 percentage points | 2 too few elections | 2008-10-12 – 2011-05-01 |
| Forum Research | 1.6 percentage points | 5 | 2011-05-01 – 2025-04-27 |
| Campaign Research | 1.8 percentage points | 2 too few elections | 2019-10-20 – 2021-09-16 |
| Abacus Data | 1.8 percentage points | 5 | 2011-04-29 – 2025-04-27 |
| Counsel Public Affairs | 1.8 percentage points | 1 too few elections | 2021-09-18 – 2021-09-18 |
| EKOS | 1.8 percentage points | 7 | 2006-01-20 – 2025-04-27 |
| Strategic Counsel | 1.8 percentage points | 2 too few elections | 2006-01-22 – 2008-10-11 |
| YouGov | 1.8 percentage points | 2 too few elections | 2021-09-13 – 2025-04-25 |
| Asking Canadians | 1.8 percentage points | 1 too few elections | 2008-10-13 – 2008-10-13 |
| MQO Research | 1.9 percentage points | 1 too few elections | 2025-04-26 – 2025-04-26 |
| DART/Maru | 1.9 percentage points | 1 too few elections | 2019-10-16 – 2019-10-16 |
| Pollara | 2.0 percentage points | 1 too few elections | 2025-04-24 – 2025-04-24 |
| Deltapoll | 2.3 percentage points | 1 too few elections | 2025-04-09 – 2025-04-09 |
| Innovative Research | 2.3 percentage points | 5 | 2011-04-25 – 2025-04-26 |
| Focaldata | 2.4 percentage points | 1 too few elections | 2025-04-23 – 2025-04-23 |
| Segma | 2.4 percentage points | 1 too few elections | 2008-10-09 – 2008-10-09 |
| Environics | 2.5 percentage points | 1 too few elections | 2011-04-21 – 2011-04-21 |
| JL Partners | 2.8 percentage points | 1 too few elections | 2025-04-16 – 2025-04-16 |
| COMPAS | 2.8 percentage points | 1 too few elections | 2011-04-29 – 2011-04-29 |
| Empirical Intelligence | 2.9 percentage points | 1 too few elections | 2025-04-07 – 2025-04-07 |
| TNS Canada | 6.1 percentage points | 1 too few elections | 2011-04-08 – 2011-04-08 |
A pollster with fewer than 3 scored elections is listed here with its figures and marked as such. Its average is shown for completeness, not as a verdict: an average over one or two elections says more about those elections than about the pollster.
Every election we hold a result for
- 2025 federal election 2025-04-28
- 2021 federal election 2021-09-20
- 2019 federal election 2019-10-21
- 2015 federal election 2015-10-19
- 2011 federal election 2011-05-02
- 2008 federal election 2008-10-14
- 2006 federal election 2006-01-23
- 2004 federal election 2004-06-28
How this is measured: for each past election we take each pollster’s last poll in the 30 days before polling day and compare it, party by party, with the result. The figure is the mean absolute difference across the parties that appear in both. Only parties the pollster actually reported are counted. A late shift in opinion is not the same thing as a polling error.