How the polls are averaged
Every publicly available poll is included in the average, weighting the poll by its sample size, when it was done and the track record of the pollster to estimate each party’s support among decided and leaning voters. The arrows show how the average has shifted since the last update. A full explanation of the methodology and a list of polls included can be found at the bottom of the Poll Tracker.
How the seat projection works
The seat projections are calculated by transposing each party’s results from the last election onto Canada’s new electoral map, and adjusting those results in each of Canada’s 343 ridings by how each party’s support has shifted in the polls in a given region. Other local factors are also taken into account. The ranges show how many seats each party could potentially win, based on how close individual ridings are projected to be and past discrepancies between polls/projections and actual results. A full explanation of the methodology can be found at the bottom of the Poll Tracker.
How the probabilities are calculated
The probabilities are based on simulations of 5,000 elections, taking into account each party’s seat projection range and how that range overlaps with other parties. This reflects how likely it is that errors in the polls and/or the projection model could change the expected outcome of the election.
Percentages may not add up to 100 because of rounding.
Party support, data table
| Region | |||||||
|---|---|---|---|---|---|---|---|
Definition of terms
Dates refers to the field date range of the survey.
Poll shows the pollster that conducted the survey and the media outlet (if applicable) that either commissioned the poll or first reported on it. When available, a link to the pollster’s report is provided.
Weight refers to the weight (out of 100) the poll carries in the aggregation. Weight is determined by the date of the survey, the sample size and the track record of the polling firm.
Method refers to the mode of contact of the polls:
Margin of Error (MOE) lists the margin of error (in percentage points) of a corresponding probabilistic sample equal to the size of the poll’s sample size.
| Poll dates | Poll | Sample | Weight | Method | MOE |
|---|
How the polls are weighted
The vote projection model averages publicly available opinion polls of decided and leaning voters that meet CBC standards. Only polls commissioned by media outlets or independently self-commissioned by polling firms are included in the aggregate. Polls commissioned by political parties, interest groups or advocacy organizations are excluded.
Polls are weighted by their age and sample size, as well as by the track record and past performance of the polling firm.
Outside of a campaign, the weight of a poll is reduced by about five per cent each day (equating to 35 per cent every 10 days). This weight reduction is incrementally increased once a campaign officially begins until, by the last week of a campaign, the weight of a poll is reduced by 35 per cent each day.
The date of the poll is determined by the last day the poll was in the field, unless the poll was conducted over a period of one week during an election campaign or two weeks outside of a campaign, in which case the date of the poll is considered to be the starting field date plus seven days (or 14 days outside of a campaign).
The sample size weighting is determined by the margin of error that applies to the poll, assuming a completely random sampling of the population.
The weightings for sample size, date and track record are combined to give each poll in the projection model an overall weight. No individual poll is ever awarded more than 66.7 per cent of the total weight of the projection, unless there have been no other recent polls. Generally, a limit of 50 per cent is placed on the weight of any one poll.
Calculating the pollster track record
Polling firms are weighted by their track record of accuracy over the last 10 years. For each election, the accuracy rating is determined using the last poll the firm released during the election campaign, the average error for all parties that earned three per cent or more of the popular vote and the amount of time that has passed since the election.
In order to take into account changes of methodology or improvements made over time, the performance of a polling firm in a recent election is weighted more heavily than its performance in an older election.
The difficulty of each election is also taken into account: elections where the average error was smaller are weighted more heavily than elections in which the error was greater. This is meant to take into consideration elections in which there were particular factors contributing to pollster error that were outside of the pollster’s control. Conversely, outlier pollsters are penalized more for elections where the consensus was close to the mark.
A reduction in the track record weighting is applied for newly published polls that are outside of the confidence intervals of the vote projection, in order to reduce the extent to which potentially outlying polls make the projection swing unrealistically. This reduction is cumulative, so that a polling firm that is consistently outside of the confidence intervals has its weight increasingly reduced over time. The penalty is removed when a firm’s polling is back within the confidence intervals, but returns if the polling results slip outside the confidence intervals again.
The track record rating for a poll that is released without relevant demographic or regional breakdowns is reduced by 50 per cent.
How seat projections work
The seat projection model uses a proportional swing method based on the difference between the results of the last election and current polls.
For example, if a party managed 20 per cent in Quebec in the 2021 federal election and is now polling at 40 per cent in Quebec, the party's 2021 election results in each riding in Quebec are doubled.
This swing is applied to every party in each riding.
How individual candidates are taken into account
Incumbent candidates historically have been less sensitive to broader trends than other candidates. When a party is losing support in a region, an incumbent candidate's support drops less than support for other candidates. Conversely, when a party is gaining support in a region, an incumbent candidate receives less of a boost than other candidates.
When a party's incumbent is not running for re-election, its projected vote share in the riding is reduced. This penalty is also applied when an MP defeated in the previous election does not run again in the subsequent election.
Leaders have historically been able to outperform other candidates when a party is losing support in a region. Leaders running for the first time in a riding have also historically experienced a significant boost in support.
Similarly to when an incumbent candidate does not run again, there is a steep penalty when a leader vacates a riding, either because they lost it in the previous election or because they are not running for re-election.
Star candidates increase their party’s projected vote share by a significant amount. Star candidates are usually former legislators or cabinet ministers, party leaders or well-known figures from the private sector.
When a party is running candidates where it did not have a name on the ballot in the previous election, the riding projection is equal to the regional vote projection for the party.
Independent candidates
If an independent politician is running for re-election, his or her vote is dropped marginally from the previous election, according to the drop in support recorded by independents in other elections. The same penalty is applied to popular independent candidates who were never elected but are running again.
Politicians who left or were forced out of their party caucuses and are running for re-election as independents retain a proportion of their vote share from the previous election, based on the circumstances of their departure from caucus. Those who depart for positive reasons retain much more of their support than those who leave in disgrace. When the circumstances are hard to define, the average performance is used. A no-incumbent penalty is applied to the party the candidate left.
When an ex-legislator attempts a comeback as an independent, using past cases as a guide the independent ex-legislator retains a portion of the vote they received the last time they stood as a candidate.
Byelections and riding-level polls
Byelections are taken into account when the incumbent party lost the seat or if the results were significantly different from that previous vote (when a major party’s support has doubled or been cut in half). The swing from byelections is calculated by how current polling levels differ from where the parties stood in the polls at the time of the byelection.
When quality riding polls are available and when they differ significantly from the projection’s estimations, riding polls are also added to the projection for an individual riding. The projection for each party in the riding is adjusted by half of the difference between the projection at the time the riding poll was conducted and the results of the poll.
How likely seat ranges are calculated
Historical polling error is used to project a range of support for each party in each individual riding. If these ranges overlap for two or more parties, then the seat is considered winnable by those parties.
The projected seats show the best estimate of how many seats each party would win if an election were held today, while the seat projection ranges represent all the seats a party could win at either the 68 per cent (likely) or 95 per cent (min./max.) confidence intervals, taking into account 68 and 95 per cent of all past polling errors. Seat projection ranges in the regional breakdown and tracking chart show the range of seats each party could win at the 68 per cent confidence interval.
The projected seat number for each party will not always be in the centre of the projection ranges. If, for example, a party is leading in a lot of close races, the projected seat total will be closer to the top end of the projection range suggesting that the party has more room to fall than to grow.