2026 general election · 7 November
Where the parties stand
A sample-weighted trend across every published party-vote poll, with the uncertainty attached. Under MMP a party needs 5% of the party vote — or one electorate seat — to win list seats.
Data as at22 July 2026· Wikipedia poll aggregate
Clears 5%
The whole 95% interval is above the line.
Too close to call
The interval straddles 5%. The polling cannot say which side of the line these parties are on.
Under 5% on the party vote
The whole interval is below the line — but see the electorate seat rule below.
One electorate seat changes the answer
A party under 5% still takes list seats in proportion to its party vote if it wins an electorate. The Electoral Act directs the Electoral Commission to disregard a party’s party vote only if it both falls under 5% and has no electorate candidate elected — so failing one of those two tests is survivable. The figures above are party vote only.
Party vote trend
110 polls from 6 polling organisations, fieldwork between 3 November 2023 and 5 July 2026. The shaded band is a 95% interval, widened to reflect how far New Zealand polls have actually missed in the past — how that is calculated, and who the 6 pollsters are.
Shaded bands are 95% intervals. Fieldwork between 3 Nov 2023 and 5 Jul 2026, from 6 polling organisations. Individual polls are plotted as marks; the line is a sample-weighted trend, not a poll.
Recent polls
All 110 as CSV →The latest polls, newest first. The leading party in each is bold; a dash is a party the pollster did not report. ★ marks a poll a party commissioned about itself.
| Fieldwork | Pollster | Sample | NAT | LAB | GRN | ACT | NZF | TPM | TOP | OTH |
|---|---|---|---|---|---|---|---|---|---|---|
| 5 Jul | Reid Research | 1,000 | 28.7 | 34.0 | 10.3 | 7.8 | 11.5 | 2.3 | 4.7 | 0.8 |
| 3 Jul | Curia | 1,000 | 30.5 | 31.5 | 10.4 | 6.9 | 10.8 | 3.4 | 3.3 | 3.2 |
| 15 Jun | Verian | 1,001 | 29.0 | 32.0 | 13.0 | 6.0 | 11.0 | 1.8 | 4.6 | 1.5 |
| 8 Jun | Freshwater Strategy | 1,038 | 29.0 | 35.0 | 10.0 | 8.0 | 12.0 | 2.0 | – | – |
| 7 Jun | Roy Morgan | 891 | 31.0 | 25.5 | 13.5 | 9.5 | 10.5 | 3.0 | 6.5 | 0.5 |
| 6 Jun | Curia | 1,000 | 30.1 | 32.2 | 11.5 | 7.8 | 11.4 | 3.1 | 3.2 | 0.8 |
| 5 Jun | Talbot Mills | 1,021 | 29.0 | 34.0 | 13.0 | 6.0 | 12.0 | 2.6 | 3.3 | – |
| 10 May | Roy Morgan | 870 | 30.5 | 26.5 | 12.5 | 10.0 | 11.0 | 2.5 | 6.0 | 1.0 |
| 5 May | Talbot Mills | – | 29.0 | 36.0 | 9.0 | 7.0 | 14.0 | 2.0 | – | – |
| 5 May | Curia | 1,000 | 30.0 | 31.9 | 9.7 | 6.5 | 11.7 | 4.1 | 2.8 | 3.3 |
| 16 Apr | Talbot Mills | 1,082 | 29.0 | 36.0 | 7.0 | 8.0 | 15.0 | 1.9 | 2.3 | – |
| 13 Apr | Verian | 1,010 | 30.0 | 37.0 | 11.0 | 7.0 | 10.0 | 1.5 | 3.3 | 1.2 |
| 12 Apr | Roy Morgan | 887 | 25.5 | 34.0 | 11.0 | 10.5 | 11.5 | 3.0 | 4.0 | 0.5 |
| 1 Apr | Curia | 1,000 | 29.8 | 33.4 | 7.8 | 9.0 | 13.6 | 2.6 | 2.6 | 1.3 |
| 16 Mar | Reid Research | 1,000 | 30.8 | 35.6 | 10.1 | 7.0 | 10.6 | 3.2 | 2.0 | 0.7 |
| 8 Mar | Roy Morgan | 872 | 26.5 | 34.0 | 11.0 | 10.0 | 11.0 | 3.0 | 4.0 | 0.5 |
About these figures
- The aggregation states that its dates are the survey dates of the poll, or where survey dates are not stated, the date the poll was released. The two are not distinguished upstream, so a poll may sit up to about two weeks later on the time axis than its fieldwork warrants.
- 9 of 110 polls report no sample size. They are listed here but excluded from the trend, because a poll with no sample size cannot be weighted and inventing a weight would fabricate its confidence band.
- Not every pollster reports every party. The gaps are not random: some houses never publish a figure for the smaller parties. An average over a small party therefore rests on a different set of pollsters than the same average over a large one, so each figure states how many polls and how many organisations it rests on.
- Polls a party commissioned about itself are marked. They are published because they were published, not because they are equivalent to an independently commissioned poll.
- Margin of error is not reported in the source table and varies by organisation and date. Margins shown here are computed from sample size.
Polls shown but not trended
- Curia · 4 December 2023 · no sample size reported
- Talbot Mills · 5 February 2024 · no sample size reported
- Talbot Mills · 5 March 2024 · sample size reported as "1,000+"
- Talbot Mills · 30 April 2024 · no sample size reported
- Talbot Mills · 5 May 2024 · no sample size reported
- Talbot Mills · 5 September 2024 · no sample size reported
- Talbot Mills · 5 November 2024 · no sample size reported
- Talbot Mills · 25 November 2024 · no sample size reported
- Talbot Mills · 5 May 2026 · no sample size reported
Source: the named polling organisations, aggregated by Wikipedia (CC BY-SA 4.0), revision 1364172385, retrieved 2026-07-22.