What will be the margin of seats won by Labour over National in the 2026 New Zealand general election?

**Current value (as of 2026-09-28):** The latest published poll, the 1News-Verian survey (fieldwork 23–27 September 2026, n=1,004, released 28 September 2026), shows 
National and Labour level at 28 percent
 in party vote, translating into 
National and Labour tied on 34 seats each
. On this poll's modelled seat allocations the margin (Labour minus National) is **0 seats** in both of Verian's scenarios (with or without an overhang from the Māori electorates) — 
National and Labour are tied on 34 seats each, the Greens have 19, ACT 14 and NZ First 11, and the Opportunity Party, on 6% of the party vote, takes seven seats in each version
.

- [2026-09-28] Latest 1News-Verian poll: 
Labour on 28%, down two points since the last poll, and National down one point, also on 28%; the Green Party is on 16%, up four points
 — https://www.thepost.co.nz/politics/361088727/1news-verian-poll-has-both-major-parties-28-greens-sky-rocket
- [2026-09-28] On the 34-34 party-vote-only reading: 
the coalition of National, Act and NZ First would gather 59 seats between them if an election was held when polling took place, while the bloc of Labour, the Green Party, Te Pāti Māori and Opportunity would trump that with 65 seats
 (this bloc figure includes Green/TPM/Opportunity seats, not just Labour vs National) — https://www.nzherald.co.nz/nz/politics/election-2026-green-party-surges-to-record-high-in-1news-verian-poll/LU6JE4FXSZAETCMYWNKKZPYRDE/
- [2026-09-28] Verian offered a second seat scenario assuming all 2023 electorate winners hold their seats; in that version too 
National and Labour are tied on 34 seats each
 — https://www.1news.co.nz/2026/09/28/the-two-ways-of-calculating-seats-for-the-1news-verian-poll/
- [2026-09-28] An independent aggregator (Stuff) framing of the same poll: 
the latest 1News-Verian poll will see the left bloc of Labour (34 seats), the Greens (19 seats), Te Pāti Māori (4 seats) and Mariameno Kapa-Kingi's Te Tai Tokerau party (1 seat), with 58 seats in the house
 versus National's 34 — https://www.stuff.co.nz/politics/361036026/nz-election-2026-live-updates-politics-campaign-news
- [2026-09-15/16] The Kiwi Polls "Quasimodo" aggregator model (last snapshot fetched, dated update 16 September 2026) put average simulated seats at National 37, Labour 36 — a **-1 seat margin for Labour** (National ahead), within a projected 123-seat overhang Parliament — https://www.kiwipolls.co.nz/
- [2026-08-31] Roy Morgan (31 August 2026): 
both blocs held steady, with the National-led parties on 49% and the Labour-led Opposition on 41%; the Opportunity Party rose 1.5 points to a record 9.5%, its fourth straight monthly gain
, with 
Labour falling to 24%, its weakest showing since September 2024
 — https://www.kiwipolls.co.nz/
- [2026-08-12] 1News-Verian (fieldwork 8–11 Aug 2026): 
Labour on 30%, down two percentage points, while National was unchanged on 29%; on those results, Labour would receive 38 seats and National 36
 — a **+2 seat Labour margin** — https://www.newstalkzb.co.nz/news/politics/labour-leads-national-as-opportunity-party-hits-8-in-1news-verian-poll/
- [2026-06-23] 1News-Verian (fieldwork ~June 2026): 
Labour is down five points to 32 percent, but could form a government with the Greens and Te Pati Maori; between them the left bloc would have 64 seats to the right bloc's 60 seats, assuming Te Pāti Māori retains all six electorate seats
 (bloc figures, not a direct Labour-vs-National seat count) — https://www.rnz.co.nz/news/politics/617657/left-bloc-leads-in-latest-poll-despite-labour-slump

**One year earlier (reference point, June 2025 — the closest dated Labour-vs-National seat estimate available):** 
National sat at 34%, Labour dipped 3 points to 29%, and the 1News-Verian polling left Labour with 37 seats
; National's seats were reported elsewhere in the same period at roughly 43 in a similar Verian poll — implying a National lead of roughly 5–9 seats over Labour at that time (i.e., a **negative** Labour-minus-National margin, somewhere around -5 to -9), a much wider gap than today's near-zero margin.

**Trajectory:** Over 2025–2026 the Labour-National gap has narrowed sharply and repeatedly flipped sign between polls/pollsters — Labour led by several seats in June and August 2026 polls, while National edged ahead in some poll aggregations (e.g., Kiwi Polls' mid-September average), and by the newest 28 September 1News-Verian poll the two majors are tied on both vote share (28% each) and modelled seats (34 each). 
Labour's standing marks its lowest result since February 2024, continuing a downward trajectory over recent polls
, while National has also slid to 
its lowest level since the final 1News poll before Judith Collins lost the party's leadership in 2021
.

**Next data point:** Polling agencies (1News-Verian, Taxpayers' Union-Curia, Roy Morgan, Freshwater Strategy/The Post, Reid Research) typically release monthly; with the election on 7 November 2026 and advance voting from 26 October, further polls are expected through October, with the actual seat result only final well after election day given 
the model projects a three-seat overhang, which expands Parliament from 120 seats to 123 and raises the majority threshold from 61 to 62
 and special votes must be counted before the writ is returned.

BOTTOM LINE: The most recent and most authoritative primary data point — the 1News-Verian poll released 28 September 2026 — shows Labour and National dead level on both party vote (28% each) and projected seats (34 each) in both of Verian's modelled scenarios, implying a current-value margin of 0, a dramatic narrowing from the roughly 5–9 seat National lead implied by comparable polling around mid-2025.

NOT FOUND: An official, single "resolution source" data feed (API/CSV) for seat margins was not available to fetch directly — all figures come from pollster press releases and news reporting of poll-derived seat projections; no exact 1News-Verian poll from September 2025 (12 months prior) was located, so the "one year ago" comparison uses the closest available (June 2025) data point instead.

**Note on scope:** New Zealand holds general elections roughly every 3 years, so there is no election-night-to-final-count window in the "prior three years" (2023–2025) — the last three actual instances of this exact calendar dynamic (election day, followed by ~3 weeks of special-vote counting to a final declared result) are the 2017, 2020 and 2023 general elections. These are the closest analogues to the Sept 28–Nov 8 2026 window, since 2026's election day (Oct 17) plus special-vote counting will similarly resolve before Nov 8.

- [2023-11-03] Final 2023 results vs election-night (Oct 14) preliminary count: 
the National Party has 48 seats compared with 50 on election night, and there are no changes for the Labour Party which has 34 seats
. The Labour–National seat gap did not move during the count (stayed at −14), but National's own total fell as seats shifted to Te Pāti Māori and the Greens: 
the National Party has two fewer seats, Te Pāti Māori has gained two, and the Green Party has gained one more
. https://elections.nz/media-and-news/2023/official-results-for-the-2023-general-election

- [2023-11-03] 
The number of special votes was 603,257 - 20.9% of total votes (17% in 2020 & 2017), with turnout of people who were enrolled to vote at 78.2%
, the highest special-vote share of the three cycles reviewed. https://elections.nz/media-and-news/2023/official-results-for-the-2023-general-election

- [2020-11-06] Final 2020 results vs election-night: 
the Labour Party has 65 seats compared with 64 on election night, and the National Party has 33 seats compared with 35 on election night
. The Labour-minus-National margin widened by 3 seats during the count (from +29 on the night to +32 final), driven entirely by National losing 2 seats to Labour on special votes. https://elections.nz/media-and-news/2020/2020-general-election-official-results

- [2020-11-06] Three electorates flipped from National to Labour on specials: 
Labour Party candidate Priyanca Radhakrishnan won Maungakiekie with a majority of 635 votes over National Party candidate Denise Lee; Willow-Jean Prime won Northland with a majority of 163 votes over National candidate Matt King; Emily Henderson won Whangārei with a majority of 431 votes over National candidate Shane Reti
 — all three seats moved from National column to Labour column. https://elections.nz/media-and-news/2020/2020-general-election-official-results

- [2023-11-03] Final 2017 results vs election night (per Elections NZ retrospective in 2017 release): 
the National Party has 56 seats compared with 58 on election night, the Labour Party has 46 seats compared with 45 on election night, and the Green Party has 8 seats compared with 7 on election night
. The Labour-minus-National margin moved from −13 on the night to −10 in the final count, a 3-seat swing toward Labour. https://elections.nz/media-and-news/2017/new-zealand-2017-general-election-official-results

- [Historical pattern] 
Since 1999, the parties that have gained seats between election day and the declaration of the official results have all been left-leaning
, and specifically for National vs Labour: 
in 2017 National's initial seat allocation was reduced by two while Labour and the Greens both gained one seat each, and in 2020 National had two fewer seats than the election night result predicted
 once specials were counted. This is a consistent structural pattern (not random noise) favoring Labour's margin over National by roughly 2–4 seats between election night and final declaration.

- [Largest surprise on record] 
the largest special-vote seat shift was in 1999, when special votes affected five parties and 14 seats, removing the Labour-Alliance coalition's election-night majority and giving the Green Party — which had zero seats on election night — an entitlement to seven seats
, showing that in extreme cases the swing can be far larger than the typical 2–4 seat range seen in 2017/2020/2023.

- [Polling accuracy — 2017] 
the last RNZ poll of polls average before the 2017 election was strikingly close to the final result: National polled at 45.1% vs actual 44.4%, and Labour at 37.2% vs actual 36.9% — both well within 1 percentage point of the actual outcome
, though NZ First's late kingmaker role still produced a change of governing bloc.

- [Polling accuracy — 2023] 
the polls ahead of the 2023 election consistently underestimated the popularity of the right-wing parties; the most accurate polls (Talbot Mills, Curia/Taxpayers' Union, and Verian/Kantar for 1 News) all still understated the right's margin over the left
, indicating a systematic recent polling bias against National/ACT that, if repeated, would push actual results further toward National than current polling implies.

- [2026 baseline] Current 1News-Verian polling (23–27 Sept 2026, n=1,004, MoE 3.4%) has National and Labour essentially tied at the party-vote level (National 28%, Labour 28%), with modeled seat projections putting the left bloc (Labour, Greens, Te Pāti Māori) around 66 seats and the right bloc around 58, per the AskNews summaries already gathered — a far tighter contest than any of the three prior elections reviewed here, where the final Labour-minus-National seat margin was −14 (2023), +32 (2020), and −10 (2017).

BOTTOM LINE: Over the last three NZ elections, the Labour-minus-National seat margin moved by roughly +1 to +3 seats toward Labour between election night and the final declared result (with an extreme historical outlier of far more in 1999), due to a persistent left-lean in special votes; but polling has recently understated National's true strength (2023), while in 2017 it was very accurate — so the historical record suggests only a modest, Labour-favoring adjustment should be expected between now and the Nov 8 resolution date, layered on top of whatever the Oct 17 election night margin turns out to be.

NOT FOUND: No exact analogue exists for a "Sept 28–Nov 8 window in each of the prior three calendar years" since New Zealand does not hold elections annually; I substituted the three most recent general elections (2017, 2020, 2023) as the closest comparable events. Could not find a systematic post-election-night-to-final-count breakdown further back than 1999 in the sources reviewed.

- **[2026-01-21] Election-day/resolution timing mismatch (mechanism, controlled by the Electoral Commission, not campaigns):** The government confirmed 
the House will rise on 24 September, Parliament dissolved on 1 October, writ day on 4 October, and advance voting beginning 26 October
. Crucially, 
official results are declared on November 27, with the writ returned by December 3
 — nearly three weeks *after* this question's November 8 resolution date. This means the resolving figure will almost certainly be the **preliminary election-night count** (7 November), not the specials-inclusive final declaration, cutting against the pro-Labour special-votes shift documented in the base-rate report (2017/2020/2023 all moved 2-4 seats toward Labour only after specials were counted).

- **[2026-07-27/28] Luxon's rule-out of the Opportunity Party (controlled by Luxon/National caucus):** Luxon 
appeared to rule out National working with the Opportunity Party, saying his party didn't want to "do business" with anyone seeking to impose new taxes
, later confirmed by his office as 
a formal rule-out, citing Opportunity's land tax advocacy
. This doesn't move the Labour-vs-National seat count directly, but it structurally caps National's bloc: as one analysis noted, 
a rule-out does come with risks — if National, NZ First and Act don't have the numbers to govern, Luxon will either need an embarrassing about-face, or has unilaterally consigned the right bloc to opposition
. Multiple commentators (Peter Dunne, 2026-07-31) argue the rule-out is "at best conditional" and could be reversed post-election, similar to Luxon's 2023 reversal on NZ First — a live wildcard for late bloc-formation but not the Labour-National seat gap itself.

- **[2026-08-10] Hipkins' reciprocal rule-out of NZ First (controlled by Labour caucus):** 
Labour leader Chris Hipkins has ruled out a governing deal with New Zealand First
, with 
the rule-out discussed at caucus, and not a single member of Labour's caucus disagreeing
. This locks Labour out of NZ First's ~9-11% support pool as a coalition option, meaning Labour's path to government (and thus its incentive/campaign focus) depends almost entirely on Green/TPM/TOP arithmetic rather than any NZ First swing — a factor that shapes campaign strategy but not the two-party seat count directly.

- **[2026-09-28] Māori electorate seat allocation is the single largest swing factor in the seat-margin calculation, not vote share (controlled by ~5 Māori-electorate voters, not national polling):** The two 1News-Verian scenarios published today show identical party-vote-derived seats for Labour and National (34-34) but diverge sharply in total-bloc composition depending solely on 
whether any party that won an electorate in 2023, including Te Pāti Māori, wins at least one again — which still gives National and Labour tied on 34 seats each
 while the overhang shifts between 0 and 4 seats. Because the Labour-vs-National number itself doesn't move across these scenarios, the real "swing" risk for the margin lies in individual electorate battlegrounds (see below), not in TPM/overhang mechanics.

- **[2026-09-28] Poll-to-poll seat volatility over just the past seven weeks (numbers behind the headline, same-day sources):** Compare the 11 August Taxpayers' Union-Curia model — which had 
Labour on 41 seats, National 39, and NZ First 15
 (Labour +2) — against today's tied 1News-Verian 34-34 model and The Post's framing that 
Labour is able to govern with a one-seat majority of 61 while National would have 34 seats, ACT 13 and NZ First 11 — a total of 58 seats
. The Labour-minus-National margin implied by pollster-derived seat models has swung from roughly +2 to 0 in under two months, underscoring that no single poll should be treated as settled ahead of the six-week campaign remaining.

- **[2026-09-28] Undecided/soft vote share remains large and could break either way:** Today's 1News-Verian poll recorded 
a sample of 1004 eligible voters, with a maximum sampling error of about ±3.4% at the 95% confidence level
, and roughly 13% of respondents undecided per the earlier AskNews summary — a bloc large enough, at current 0-2 seat margins, to flip the sign of the Labour-National gap on its own by decision day.

- **[2026-09-28] Electorate-seat mechanics (National's 43 held electorates vs Labour's list-heavy 2023 base) are an under-modelled risk to the margin:** The election2026.nz tracker's own methodology flags that 
the "Full right overhang" preset loads a split-ticket scenario where National's party vote is moved to ACT and NZ First, half each, with the 43 electorates it won in 2023 kept as pure overhang
 — a scenario that would inflate National's total seats (and thus swing the margin against Labour) independent of any change in Labour's or National's headline party-vote share, because New Zealand's MMP overhang rules add seats for electorate wins beyond a party's proportional entitlement.

BOTTOM LINE: The Labour-minus-National margin is currently pinned near zero by tied party-vote polling, but the resolution date (Nov 8) falls before New Zealand's official specials-inclusive count (due Nov 27), so the number that resolves this question is very likely the preliminary election-night tally — removing the modest pro-Labour "specials bump" seen in 2017/2020/2023 and leaving the outcome hostage to (a) whichever way ~13% undecided voters break, and (b) electorate-level overhang effects tied to National's 43 held seats and the fluid Māori electorates.

NOT FOUND: No official Electoral Commission or party statement confirming whether this specific question's compilers intend to use the preliminary (election-night) count or wait for a later confirmed count published after Nov 8; no seat-by-seat marginal-electorate list (e.g., the closest 5-10 seats decided by <1,000 votes in 2023) was located to quantify overhang risk precisely.

- [2026-09-28] The aggregator election2026.nz, combining 1News–Verian, RNZ–Reid Research, Taxpayers' Union–Curia, Talbot Mills, The Post/Freshwater Strategy and Roy Morgan polls with house-effect adjustment, projects 
a house-effect-adjusted polling average, MMP seat projection, coalition builder and election-day forecast for 7 November 2026
, with the table showing National on 29.6% party vote (36 projected seats) and Labour on 29.4% (also 36 seats) — a projected margin of roughly zero seats, essentially a dead heat. https://election2026.nz/

- [2026-09-28] The same tracker's Monte Carlo-style simulation states the right bloc (National-ACT-NZ First) wins outright in "38.6% of 10,000 simulated elections and the left bloc (Labour, Greens, Te Pāti Māori) in 16.0%", with a further "left, only with TOP" scenario at 31.6%, reflecting how sensitive the majority is to the Opportunities Party (TOP) rather than to Labour's own seat margin over National. https://election2026.nz/

- [2026-09 fetched] Polymarket's "NZ Election: Labour Party # of seats?" market (opened 29 Apr 2026, $23,905 traded) currently prices "30-34" at 31% and "35-39" at 34%, with only 7% for "45-49" and negligible weight above 50, i.e. traders see Labour most likely finishing in the low-to-mid 30s. https://polymarket.com/event/nz-election-labour-party-of-seats

- [2026-09 fetched] Polymarket's parallel "NZ Election: National Party # of seats?" market ($21,608 traded) prices "35-39" at 37% and "30-34" at 34%, i.e. National is priced very slightly higher than Labour on average, but the two distributions heavily overlap in the 30-39 seat band. https://polymarket.com/event/nz-election-national-party-of-seats

- [2026-09 fetched] Polymarket's own market-context note describes 
"Recent polling averages place Labour near 29% party vote, projecting roughly 35 seats in New Zealand's 122-seat MMP parliament ahead of the November 7 election, with individual surveys showing the party between 28% and 32%. National sits at comparable levels around 28-29%"
, underscoring a razor-thin implied seat margin between the two majors. https://polymarket.com/event/nz-election-labour-party-of-seats

- [2026-09-28 search snippet] A separate Polymarket market, "NZ Election: Popular Vote Margin of Victory?", currently leads with "Labour 0-5%" at 35% ahead of "National 0-5%" at 18%, implying traders lean toward Labour narrowly winning the party-vote share, though this does not directly translate into seat margin given coalition/overhang dynamics. https://polymarket.com/event/nz-election-popular-vote-margin-of-victory

- [2026-06-13 dated] A lower-volume, less-reliable Polymarket market tracked by aggregator site Lines.com ("Will Labour win 55+ seats") shows only 46% for a 55+ outcome, but this market has "Total volume is only $757 with zero traded in the last 24 hours," making it a weak signal versus the higher-volume seat-bracket markets. https://www.lines.com/prediction-markets/elections/nz-election-labour-party-of-seats

- [2026-08-12] The Taxpayers' Union-Curia poll's own seat projection (Aug 1-4 fieldwork) gave the National-ACT-NZ First bloc 61 seats (a bare one-seat majority) with National on 31.0% and Labour on 27.8%, a poll-based scenario implying a modest National seat lead over Labour at that time, though this was two months before the current, much tighter polling. https://sundayguardianlive.com/world/new-zealand-election-polls-2026-pm-christopher-luxons-national-coalition-leads-with-61-seats-labour-loses-support-opportunity-party-crosses-5-heres-what-opinion-polls-say-curia-vs-roy-morgan-258705/

- [2026-08-04] Roy Morgan's July 2026 poll-based seat model instead had the National-led bloc's seats falling and 
the Labour-led Parliamentary Opposition winning 51 seats (down four seats)
 against a shrunken National-led bloc, illustrating how volatile pollster-specific seat models have been over just the past two months — a sign that any single-source projection should be treated cautiously this cycle.

- [2026-09-28] Independent tracker predictor.nz also runs an MMP seat-projection model assuming 
"Te Pāti Māori is the only party that may enter overhang and that the ACT Party will win at least one electorate seat... All other parties are assumed to be bound by the 5% MMP threshold"
, but as fetched it does not surface a single current headline Labour-minus-National number distinct from election2026.nz's near-zero projection. https://predictor.nz/

- [historical accuracy] election2026.nz's methodology section states its simulation approach was "calibrated by replaying this exact method against the 2014, 2017, 2020 and 2023 elections" and that "at this range, history says a major party's polling can miss by five points or more," meaning even the best current aggregator projection (~0 seat margin) carries wide uncertainty bands of several seats in either direction six weeks out. https://election2026.nz/

BOTTOM LINE: The best current, higher-volume market/aggregator evidence (election2026.nz's polling-average seat model and both Polymarket seat-bracket markets, all dated 2026-09-28/fetched today) converges on a near-tied outcome, with Labour and National both projected in the mid-30s seats and a margin close to zero (possibly single digits in either direction); a minority popular-vote-margin market leans slightly toward Labour, but historical volatility in NZ seat forecasts (5+ point misses per the tracker's own backtesting) means the sign and size of the final Labour-minus-National margin remains genuinely uncertain.

NOT FOUND: No single, high-confidence "expert consensus" point-forecast for the exact Labour-minus-National seat margin (e.g., from a major NZ psephologist or bank) was found; no evidence of a dedicated Kalshi or iPredict market on this specific margin; could not confirm whether the Lines.com aggregator's 46% "55+ seats" figure reflects a real, current Polymarket price given its very low volume and possible staleness.

- [2026-04-19] The single largest Labour-over-National gap recorded anywhere in the 2026 cycle came from 1News-Verian: 
support for the National Party and Prime Minister Christopher Luxon crashed to new lows, while Labour overtook National for the first time in the poll since before the 2023 election, climbing into a seven-point lead
. 
The left bloc of Labour, the Greens and Te Pāti Māori would have had 66 seats compared to 58 for the right bloc, with National losing 12 seats
 on those numbers — the evidence-backed template for how far the gap can widen when National is in crisis (this poll came "amid the fuel crisis and after the PM's Cabinet reshuffle," per the same report). This is the ceiling case the "high scenario" needs to re-emerge.
- [2025-12-05/10] An earlier still-wider party-vote gap appeared in The Post/Freshwater Strategy poll (5–10 Dec 2025), which had National on 30% against Labour's 38% — an 8-point lead, per the Wikipedia-compiled polling table — showing double-digit-point Labour leads are not unprecedented this term, even if none have persisted more than a month.
- [2026-08-14/21] The most recent poll giving a direct Labour-vs-National seat count (rather than a bloc figure) is RNZ-Reid Research: 
Labour would have 39 seats… National would have 35 seats
 — a +4 Labour margin, the highest concrete Labour-minus-National seat number found in the dataset outside the April Verian bloc-only estimate.
- [2026-03-12/20] The same RNZ-Reid Research series in March had 
Labour with the biggest party-vote share at 35.6%, while National sat on 30.8%
, a ~4.8-point party-vote lead, but the seat model still produced a 60-60 bloc tie because 
New Zealand First and the Greens picked up party vote while everyone else suffered drops
 — illustrating that a high Labour-National margin requires Labour's gains to come disproportionately at National's direct expense, not merely a general anti-government swing that also lifts NZ First/Greens.
- [2026-09-27] A structural tailwind for a wide Labour-National gap: Labour's own campaign launch coverage notes the party unveiled 
a new cost-of-living policy – wiping 10% of student loan debt for those who didn't benefit from Labour's previous fees-free policy
, a retail policy explicitly targeted at price-sensitive, youth-leaning voters who are also the demographic most likely to swing directly from National-curious "soft" voters to Labour rather than to a minor party.
- [2026-09-28] Working against the widest-gap scenario, the newest 1News-Verian poll shows the anti-government swing is currently being captured by the Greens, not Labour: 
the Greens climbed up four points in the latest poll, while Labour was on 28%, down two points
 — meaning momentum that in April flowed to Labour is now diffusing to a third party, which caps rather than widens the direct Labour-National seat margin even if the overall left bloc grows.
- [2026-04-08] A Taxpayers' Union-Curia snapshot cited by an aggregator found 
Labour leading National for the first time since 2022, with the left bloc (Labour, Green, Te Pāti Māori) holding a majority over the right bloc
, corroborating that April 2026 was a genuine trough for National across multiple pollsters simultaneously, not a single-pollster outlier — raising the credibility of the ~7-9 point gap as a real, replicable "high" state rather than noise.
- [2026-04-19] The same April period produced record-setting personal numbers for Hipkins: an aggregator noted 
Hipkins leads by 8 points — no NZ PM has ever recovered from trailing the opposition leader by this margin 7 months from election day in modern polling history
 (referring to Luxon's deficit), indicating that if Luxon's personal ratings deteriorate again to April levels, the party-vote gap that produces a high Labour-National seat margin has historical precedent within this very cycle.
- [2026-09-28] Countervailing mechanism: even in today's most left-favourable poll read, 
National and Labour are tied on 34 seats each… the Opportunity Party, on 6% of the party vote, takes seven seats
, and 
neither bloc reaches a majority in either scenario without Opportunity's MPs
 — showing the direct Labour-National margin has structurally compressed toward zero as the election nears, and a jump back to the April-style 7-9 point gap would require a fresh shock (economic, scandal, or campaign misstep) rather than simple trend continuation.
- [2026-09-28] Special-votes precedent (already noted by colleagues) adds a modest, reliably left-leaning tailwind on top of any election-night margin, which in a high scenario would stack on an already-wide Labour lead rather than create one from nothing.

BOTTOM LINE: The evidence-backed high end of the range is a Labour-National seat margin in roughly the +8 to +12 seat band, precedented by the 19 April 2026 1News-Verian poll (7-point party-vote lead, National down 12 seats, left bloc 66–58) and by the 5–10 December 2025 Post/Freshwater poll (8-point lead), both of which would require a comparable National-specific collapse (economic shock, leadership crisis, or a decisive cost-of-living backlash) to recur between now and 7 November, since the current (28 September) polling has the two majors dead level and the anti-government swing is currently flowing to the Greens rather than Labour directly.

NOT FOUND: No poll in the 2026 cycle has shown a direct Labour-minus-National seat margin above roughly +4-5 (RNZ-Reid Research August) using an explicit per-party seat count; the wider +7 to +12 figures come from bloc-level (Labour+Greens+TPM vs National+ACT+NZF) comparisons or party-vote-only leads, so a precise historical "high" for the pure two-party seat margin in this cycle could not be confirmed beyond the RNZ August figure.

- [2026-09-07] The Sept 1–3 Taxpayers' Union–Curia poll — the most National-favourable poll of the whole cycle — put National at 29.0% (down 2.0) and Labour at 25.9% (down 1.9), translating into a seat model of 
National down 2.0 points to 29.0%, while Labour is down 1.9 points to 25.9%
 and, per the Curia seat table, National 39 seats vs Labour 35, i.e. a **−4 seat margin for Labour** under this poll's methodology (no overhang assumed). https://www.taxpayers.org.nz/sept26_pollnztughjkuhn

- [2026-09-07] The same Curia release shows Labour's 25.9% is 
its lowest result in this poll since July 2024, and even lower than the 26.9 point thrashing it got at the 2023 election
, while National's 29% is 
also significantly lower than the 38 percentage points it won at the last election
 — evidence a genuinely bad Labour showing is achievable within the current polling range even as both majors fall from 2023 levels. https://www.rnz.co.nz/news/politics_election-2026/1278458/labour-and-national-polling-collapsing-both-under-30-percent

- [2026-09-07] Under this poll's bloc arithmetic, 
the combined projected seats for National, NZ First and ACT is 64, up three seats, while the combined projected seats for Labour, the Greens, Opportunity and Te Pāti Māori was 56 (down three seats)
 — an 8-seat bloc gap, showing the low-scenario mechanism is not just a Labour-National swing but Opportunity collapsing below threshold (from 6.1% in August to 4.5%), which pushes its ~8 "wasted" seats disproportionately toward National's bloc. https://www.interest.co.nz/public-policy/140143/taxpayers-union-curia-poll-shows-labour-slumping-its-lowest-result-july-2024

- [2026-08-24] RNZ–Reid Research (14–21 Aug 2026) also shows a mid-cycle low-scenario data point: 
Labour would have 39 seats, the Greens 12, and Te Pāti Māori 4, making 55 seats for the left bloc in total. National would have 35 seats, New Zealand First 15, and ACT 9, or 59 in total for the right bloc
 — a +4 Labour seat lead individually, but the right bloc still 4 seats ahead overall, illustrating that even polls with Labour narrowly ahead of National can still produce a low/negative outcome for the broader left coalition. https://www.rnz.co.nz/news/politics_election-2026/1093636/rnz-reid-research-poll-opportunity-in-kingmaker-position-luxon-and-hipkins-approval-slumps

- [2026-09-19] Substack analyst Elliot Crossan's three-month rolling average (to late Sept) has 
both parties sitting on a three-month average of 30.4% each, while combined support for the Greens, ACT, NZ First, TPM and Opportunity is sitting at 39.2%
, implying the "true" underlying race is closer to dead-even than any single poll, but also that 
Labour is headed for its 4th worst result in nearly 100 years; National is headed for its 3rd worst result since the party was founded in 1936
 — both majors are historically weak, so small campaign shocks could tip the margin sharply in either direction. https://elliotcrossan.substack.com/p/election-2026-welcome-to-the-unpopularity

- [2026-09-15] RNZ's Freshwater Strategy/The Post poll (4–11 Sept) shows a mechanism for a further Labour slide: 
Since the last of this poll in June, Labour has slumped by 7 points. It remains in the lead, but only just
 — a seven-point one-poll drop for Labour is the single largest swing recorded this cycle and, if repeated by other pollsters, would move the margin decisively toward National. https://www.rnz.co.nz/news/politics_election-2026/1382772/possible-hung-parliament-as-labour-and-national-slump-in-latest-poll

- [2023-11-03 precedent, reinforcing downside risk] The 2023 election is the direct precedent for a real-world "low scenario": polls in the final weeks had National and Labour roughly tied or Labour trailing only modestly, yet 
Seats won 48 / 34 / Seat change +15 / −31
 for National vs Labour respectively — a final margin of National +14, showing final-campaign consolidation of the right vote (undecideds breaking right, minor-right-party voters returning to National) can produce a much larger gap than mid-campaign polling implies.

- [2026-09 fetched] Curia's own methodology note states seat projections 
assumed all current parliamentary parties with electorate seats will win at least one electorate seat and be eligible for List MPs
 but explicitly 
no overhang seats are assumed or projected
 for National or Te Pāti Māori — a modelling choice that, if incorrect (i.e., if Te Pāti Māori wins Māori seats beyond its party-vote share, creating overhang), could shrink Labour's practical seat count relative to National even at similar vote shares, since overhang expands the House rather than reallocating National/Labour seats. https://www.taxpayers.org.nz/sept26_pollnztughjkuhn

BOTTOM LINE: A genuine low-scenario path exists and has already been observed mid-cycle (the Sept 1–3 Curia poll's National 39–Labour 35 seat model, a −4 Labour margin) driven by Opportunity falling back below the 5% threshold and right-bloc consolidation; combined with the 2023 precedent of late-campaign swings toward National well beyond mid-campaign polling and Labour's historically weak underlying vote share, a final result with Labour trailing National by several seats (roughly −4 to −10) is a credible bottom-of-range outcome, distinct from the tied-race scenario implied by the newest (Sept 28) 1News-Verian poll.

NOT FOUND: No single pollster or aggregator has published a scenario explicitly modelling "Opportunity fails threshold AND late-campaign right consolidation" combined; no NZ-specific analysis quantifying how much of the 2023 late swing was due to undecided-voter breakage versus differential turnout was located in this search.

13 AskNews articles (newest first; 6 from the last 48 h, 10 from the archive). Summaries are AskNews's own; dates are publication dates.

[2026-09-28] Left Bloc in Contention if National Rules Out Opportunity Party in Latest 1News-Verian Poll
The Brief — https://www.thebrief.news/en/standard/article/24750/left-bloc-in-contention-on-latest-1news-verian-poll-if-national-rules-out-opportunity-party
  - According to a 1News-Verian poll published on 28 September 2026 and conducted from 23 to 27 September with 1,004 eligible voters (margin of error 3.4 percent), the left bloc of Labour, the Greens, and Te Pāti Māori could form a government if National Party leader Christopher Luxon maintains his refusal to work with the Opportunity Party.
  - The poll records a lift in Green Party support, with 13 percent of respondents undecided or refusing to answer.
  - Verian modeled two scenarios based on Māori electorate results: one where Te Pāti Māori wins four seats and Mariameno Kapa-Kingi's Te Tai Tokerau party wins one, and another with different splits; in both, the Opportunity Party holds the balance of power, as neither main grouping can reach the 61 seats required for a majority on its own.
  - On preferred prime minister, Christopher Luxon, Green co-leader Chlöe Swarbrick, and ACT leader David Seymour were each up two points, while Labour leader Chris Hipkins was down one point and NZ First leader Winston Peters remained steady.
  - Previous polls show Labour support at a 30-year low in June, with the June poll projecting 66 seats for the left bloc against 58 for the right bloc, while the 19 April poll projected 66 seats for the left and 58 for the right.
  - The article notes that with 13 percent undecided and a 3.4-point margin, the party-vote order could shift before writs are returned, and the 'veto' by Luxon on working with the Opportunity Party remains a critical factor in post-election coalition talks.

[2026-09-28] Greens Surge in Latest Poll as Left Bloc Eyes Government
Otago Daily Times Online News — https://www.odt.co.nz/news/national/greens-surge-in-latest-poll-opportunity-hold-balance-of-power-vlfhsmnc
  - According to new polling from TVNZ-Verian published on September 28 2026, the Green Party has surged to 16 percent support, up 4 points, while the left bloc (Labour and Greens combined) is positioned to form a government if the National Party refuses to work with the Opportunity Party.
  - The poll, conducted between September 23 and 27 2026 with 1,004 eligible voters, shows the National Party at 28 percent (down 1 point) and the Labour Party at 28 percent (down 2 points).
  - The Opportunity Party holds the balance of power in both modeled scenarios, regardless of whether Te Pāti Māori wins four Māori seats or five seats including one for Mariameno Kapa-Kingi's Te Tai Tokerau party.
  - Christopher Luxon's preferred prime minister rating rose to 19 percent, up 2 points, while Chris Hipkins fell to 14 percent, down 1 point.
  - Chlöe Swarbrick and David Seymour each gained 2 points, reaching 8 percent and 7 percent respectively.
  - Winston Peters remained steady at 7 percent.
  - The poll has a margin of error of 3.4 percent at the 95 percent confidence interval and is described as a snapshot rather than a prediction.

[2026-09-27] National and Labour Election Campaign Launches in Auckland
The Spinoff — https://thespinoff.co.nz/analysis/28-09-2026/field-notes-from-national-and-labours-campaign-launches
  - On Sunday, September 27, 2026, New Zealand's two major parties launched their election campaigns in Auckland.
  - National Party held its event at the New Zealand International Convention Center, drawing a crowd of around 1,000 people, though the large venue left several hundred seats empty.
  - Labour Party launched at the Due Drop Events Centre, a smaller venue with a capacity of 690 seats that was full to capacity and described as far more boisterous.
  - National Party leader Christopher Luxon emphasized a pledge of 'no new taxes' and received strong support from senior figures including Simeon Brown and John Key, who urged against strategic voting.
  - Luxon spoke about his mother, Kathleen, and criticized Labour for a lack of policies.
  - In contrast, Labour leader Chris Hipkins focused on the rising cost of living, accused National of lying about tax policy, and announced a new policy to write off 10% of eligible student loan balances for New Zealand-based graduates.
  - The article notes that by measuring enthusiasm, Labour was the clear winner, with refreshed faithful compared to the determined but less enthusiastic National supporters.

[2026-09-27] NZ Election 2026: Extreme 20-Day Wait for Results Predicted (opinion)
TheDailyBlogNZ — https://thedailyblog.co.nz/the-20-days-between-election-and-final-results-will-be-the-most-extreme-in-nz-political-history/
  - According to an article published in TheDailyBlogNZ on September 27, 2026, New Zealand's upcoming election is projected to create a highly contested political environment.
  - The author predicts that the National Party will lose approximately 40 electorates and secure a party vote of under 27%, which will generate an Overhang of between 6 to 8 seats while placing ACT and NZF in double digits.
  - The article argues that Labour is chasing a non-existent centrist voter in a social media hate algorithm culture, whereas the winner will turn out their base tactically without watering down their appeal.
  - Consequently, the Overhang will mean that any side winning 51% of the vote will require four political parties to form a government.
  - The author states they do not expect the election result to be known on November 7, as the final outcome must wait for Special votes to arrive on November 27, creating a 20-day period described as 'the most extreme in NZ Political History.' The article warns that 'Huge money' will be spent arguing that the TOP party must cut a deal with National, ACT, and NZF, and cautions that if the Left is not ready to battle the ensuing media war, they will be 'swallowed up and spat out by Big Money interests setting the agenda.'.

[2026-09-27] National and Labour Launch 2026 Election Campaigns with Contrasting Events and Policies
The Spinoff — https://thespinoff.co.nz/the-bulletin/28-09-2026/pledge-cards-and-bullshit-national-and-labour-officially-launch-their-campaigns
  - On Sunday, September 27, 2026, New Zealand's two major political parties officially launched their election campaigns in Auckland, marking the start of the 2026 general election.
  - National Party, led by Prime Minister Christopher Luxon, held its launch at the New Zealand International Convention Centre, which drew a headcount of roughly 750 people but left rows of empty seats visible in a venue capable of holding more.
  - The party's central message was 'no new taxes,' with Luxon being introduced as 'Mr No New Taxes' by media outlets including The Herald and RNZ.
  - National unveiled a 10-point pledge card including economic growth, compulsory KiwiSaver with higher contributions, a $1,500 Baby Boost, extended paid parental leave from 26 to 30 weeks, replacing NCEA with regular assessment, tougher consequences for criminals, faster healthcare, infrastructure upgrades, supermarket competition, and returning the books to surplus.
  - Finance Minister Nicola Willis teased the upcoming pre-election economic update on Tuesday, promising 'smaller deficits, bigger surpluses, less debt.' Meanwhile, Labour Party, led by Prime Minister Chris Hipkins, launched its campaign at the Due Drop Events Centre in Manukau, filling its 690-seat auditorium to capacity with some attendees turned away.
  - Hipkins focused his speech on the cost of living and directly challenged Luxon's recent 'cry me a river' remark about rising prices, calling it 'out-of-touch' and stating, 'It made me so angry.' Labour announced a new policy offering a 10% write-off of student loan balances for eligible graduates, with loans of $2,000 or less cleared entirely.
  - The policy faces criticism from National's Simeon Brown, who called it 'yet another expensive handout,' and ACT's David Seymour, who labeled it a 'bribe.' Hipkins also accused National of lying about Labour's tax intentions, calling it 'bullshit.' According to The Spinoff's Joel MacManus, who attended both events, Labour's launch was more energetic and enthusiastic, while National's supporters appeared 'determined, but nowhere near as enthusiastic as they were three years ago.'.

[2026-09-08] Taxpayers' Union-Curia Poll Projects Coalition Retention Despite National Decline
THE GOOD OIL — https://goodoil.news/poll-puts-coalition-on-64-seats-as-national-falls-below-30-2/
  - A Taxpayers' Union-Curia poll conducted September 8, 2026, projects that the governing coalition of National, ACT, and New Zealand First would retain 64 of Parliament's 120 seats, despite National's party vote falling 2 points to 29.0%.
  - ACT rose 3.2 points to 10.5% (winning 14 seats), while New Zealand First dropped 0.7 points to 8.4% (winning 11 seats); National is projected to win 39 seats, giving the coalition a combined party vote of 47.9%.
  - The opposition is projected to hold 56 seats: Labour fell 1.9 points to 25.9% (its lowest result since 2024 and below its 2023 election result) with 35 seats, the Greens gained 3.8 points to 13.9% with 18 seats, and Te Pāti Māori polled 2.0% with 3 seats.
  - Labour leader Chris Hipkins stated that the party's internal Talbot Mills polling places Labour above 30%, though he declined to provide the exact figure, and cited a recent RNZ-Reid Research poll showing Labour at 32.2%, while acknowledging the broader downward trend.
  - The poll methodology assumes all current parliamentary parties with an electorate seat win at least one again and assumes no overhang seats.
  - The Opportunity party fell 1.6 points to 4.5%, below the 5% threshold and assumed not to win an electorate, resulting in zero seats; the article notes that crossing 5% or winning an electorate could materially change the arithmetic.
  - Christopher Luxon remains ahead of Chris Hipkins as preferred prime minister with 21.2% to 16.9%.
  - Curia surveyed 1,000 eligible and likely voters over three days, with a maximum margin of error of ±3.1 points and 4.1% undecided.
  - The article emphasizes this is a snapshot, not a forecast.

[2026-09-06] Latest Poll Shows Coalition Leading as TOP Slips Below Threshold
The Brief — https://www.thebrief.news/en/standard/article/17935/latest-poll-puts-coalition-ahead-as-top-slips-below-threshold
  - A Taxpayers' Union-Curia poll conducted 1-3 September 2026 projects the National, ACT, and NZ First coalition will hold power with 64 seats, while the opposition Labour, Greens, and Te Pāti Māori bloc secures 56 seats.
  - The poll shows The Opportunities Party (TOP) slipping to 4.5 percent, falling below the 5 percent threshold required under the MMP system to enter Parliament without winning an electorate seat, which would cost TOP all its parliamentary seats.
  - This marks a reversal from previous polls in July and August where TOP was polling at 6.1 percent and approaching the threshold.
  - National party support fell to 29 percent, Labour dropped to 25.9 percent, while the Greens rose to 13.9 percent and ACT increased to 10.5 percent.
  - New Zealand First slipped to 8.4 percent, and Te Pāti Māori gained 0.5 points to 2 percent, though it is assumed to win three electorates, keeping it in Parliament.
  - Christopher Luxon remains the preferred prime minister at 21.2 percent, down 2 points, while Chris Hipkins stands at 16.9 percent, down 2.9 points.
  - The poll surveyed 1,000 adult New Zealanders with a maximum margin of error of ±3.1 percent.
  - The results raise questions about whether TOP's decline is genuine or within the margin of error, as their true support could range from 1.4 to 7.6 percent.
  - If TOP clears the threshold, their seats would likely come from the coalition, whereas falling below it secures the coalition's majority.
  - Te Pāti Māori aims to use the MMP overhang mechanism by winning three electorate seats to increase total seats and reduce the coalition's governing requirement.

[2026-09-06] Labour and National Polling Collapse in New Zealand
RNZ — https://www.rnz.co.nz/news/politics_election-2026/1278458/labour-and-national-polling-collapsing-both-under-30-percent
  - According to a Taxpayers' Union-Curia Poll conducted between Tuesday 01 September and Thursday 03 September 2026, New Zealand's major parties are polling below 30 percent, with the current coalition government still projected to retain power.
  - The poll, conducted by Curia Market Research for the Taxpayers' Union, found that the National Party fell two points to 29 percent, while the Labour Party dropped 1.9 points to 25.9 percent, its lowest result since July 2024.
  - The Quilae Wong-led Opportunity party fell from 6.1 to 4.5 percent, remaining below the 5 percent threshold required to enter parliament.
  - Conversely, the Greens gained 3.8 percentage points to 13.9 percent, and ACT increased by 3.2 points to 10.5 percent.
  - In terms of projected seats, the coalition (National, ACT, and NZ First) would hold 64 seats compared to the opposition (Labour, Greens, and Te Pāti Māori) on 56 seats, assuming Te Pāti Māori wins three seats.
  - Christopher Luxon remained the preferred prime minister at 21.2 percent, ahead of Chris Hipkins at 16.9 percent.
  - The poll surveyed 1000 adult New Zealanders with a maximum margin of error of +/- 3.1%.

[2026-08-21] Victorian State Election 2026: Coalition, Labor, and One Nation Battle for 45 Seats
Australian Broadcasting Corporation — https://www.abc.net.au/news/2026-08-22/coalition-path-to-victory-victoria-state-election-2026/107063196
  - With the Victorian state election scheduled for November 28, less than 100 days away, the political landscape is shifting as the Liberal and National Coalition, after 12 years in opposition, attempts to topple the Labor government.
  - According to the article, the lower house comprises 88 seats, requiring 45 to govern; Labor currently holds 56 seats (with the Greens holding 3), while the Coalition holds 29 (20 Liberals, 9 Nationals).
  - Recent opinion polls suggest One Nation will be a significant player, potentially preventing any single party from reaching the 45-seat majority, making a minority government a 'genuine prospect'.
  - Labor insiders fear the party could lose upwards of 20 seats, particularly in Melbourne's west, north-west, and eastern suburbs, as well as regional areas.
  - New Premier Ben Carroll, who replaced Jacinta Allan last month, has made crime a centrepiece of his leadership to appeal to voters in at-risk electorates like Sunbury, Sydenham, Melton, and Yan Yean.
  - Labor faces specific risks in seats like Niddrie (Carroll's seat, held with a 6.7 per cent margin), Pakenham, Bayswater, and Bass.
  - The Coalition, led by Jess Wilson, must win at least 16 additional seats to govern alone, a task made 'steep' by One Nation's presence.
  - Research briefed to Labor ministers in June claimed One Nation could take as many as 17 regional and outer-suburban seats from the Coalition, with six Nationals-held seats identified as 'easy One Nation gains'.
  - The Greens are also challenging Labor in inner-city seats like Richmond, Northcote, and Footscray, though Labor believes the announcement of a royal commission and a recent teacher pay deal will help them hold ground.
  - Early polls by Resolve and Redbridge showed a slump in Labor's primary support following the leadership change, though this was within the margin of error.

[2026-08-07] Taxpayers' Union Poll: National Party Leads with Narrow Margin in 2026 Election Forecast
RNZ — https://www.rnz.co.nz/news/chinese/927715
  - According to a new poll conducted by the Taxpayers' Union and commissioned to Curia, released on August 7, 2026, the National Party is projected to form a coalition government with a slim majority.
  - The National Party's support rose by 0.5 percentage points to 31%, making it the largest party in parliament.
  - In contrast, the Labour Party's support fell by 3.7 percentage points to 27.8%.
  - The poll indicates that the ACT Party increased its support to 7.3%, while the Green Party declined slightly to 10.1% and the NZ First Party dropped 1.7 percentage points to 9.1%.
  - Notably, the Future New Zealand Party (Opportunity Party) is projected to enter parliament with 6.1% support, though National has confirmed it will not cooperate with them, leaving them as part of the opposition.
  - The Māori Party recorded its worst performance since March 2023, with support falling 1.9 percentage points to 1.5%.
  - Based on these figures, the coalition of National, ACT, and NZ First would hold 61 seats, compared to 59 seats for the opposition bloc (including the Opportunity Party).
  - In terms of prime ministerial preference, Christopher Luxon of the National Party leads with 23.2% personal support, up 4.1 percentage points, against Labour's Chris Hipkins at 19.8%, who saw a slight decline.
  - The survey was conducted between last Saturday and Tuesday with a sample size of 1,000 voters and a margin of error of ±3.1%.

[2026-08-07] National Leads in New Poll as Opportunity Party Enters Parliament
Otago Daily Times Online News — https://www.odt.co.nz/news/national/coalition-just-holding-on-opportunity-could-enter-parliament-poll-shows-f6xw42lf
  - A new Taxpayers' Union Curia poll released on Friday indicates the National-led coalition remains in a position to form a government, albeit by a narrow margin.
  - The poll shows the Opportunity Party has broken the 5% threshold with 6.1%, though it would join the Opposition as National ruled out working with them.
  - National's support rose 0.5 points to 31%, making it the most popular party, while Labour dropped 3.7 points to 27.8%.
  - The coalition parties (National, ACT, and New Zealand First) would secure 61 seats, compared to the Opposition's 59 seats.
  - ACT gained slightly to 7.3%, while Te Pati Maori fell to 1.5%, its lowest since March 2023.
  - In preferred Prime Minister polling, National's Christopher Luxon leads with 23.2%, up 4.1 points, against Labour's Chris Hipkins at 19.8%.
  - The general election is scheduled for November 7.
  - The poll surveyed 1,000 respondents between Saturday and Tuesday with a margin of error of +/- 3.1%.

[2026-06-14] Labour Holds Lead Over National in Latest NZ Poll as ACT and NZ First Gain
RNZ — https://www.rnz.co.nz/news/political/598196/national-slides-below-30-percent-in-latest-poll-act-and-nz-first-gain
  - According to the latest The Post/Freshwater Strategy poll conducted between June 5-11, 2026, with a sample of 1,038 eligible voters and a margin of error of +/- 3%, New Zealand's major parties have seen shifts in support.
  - Labour leader Chris Hipkins leads with 44% support as the preferred Prime Minister, down two points since February, while National leader Christopher Luxon has gained ground to 37%, up three points.
  - In terms of party support, Labour stands at 35%, down two points, maintaining a lead over National, which has dropped one point to 29%.
  - Meanwhile, New Zealand First has risen to 12% (up one point) and ACT has increased to 8% (up two points).
  - The Greens remain steady at 10%, and Te Pāti Māori holds 2%.
  - Based on these figures, the coalition parties would hold 68 seats in parliament, compared to 62 for the left bloc.

[… older articles omitted]

## Tool outputs

### resolution_source (FAILED)
  Error: Unknown

### wikipedia_lookup (SUCCESS)
Wikipedia (3 lookups):

  Opinion polling for the 2026 New Zealand general election:
  


Eight polling firms have conducted opinion polls during the term of the 54th New Zealand Parliament (2023–present) for the 2026 New Zealand general election. The regular polls are the quarterly polls produced by Television New Zealand conducted by Verian and Radio New Zealand (RNZ) conducted by Reid Research, along with monthly polls by Roy Morgan and by Curia. Curia Market Research is no longer a member of the Research Association of New Zealand, following complaints and the resignation from RANZ by its principal, David Farrar. The sample size, margin of error, and confidence interval of each poll varies by organisation and date. The parliament was elected on 14 October 2023, and the election is scheduled to be held on 7 November 2026.

  2026 New Zealand general election:
  The 2026 New Zealand general election is scheduled to be held on 7 November 2026. Voters will elect 120 members to the House of Representatives under New Zealand's mixed-member proportional (MMP) voting system, comprising 71 members elected from single-member electorates and 49 members elected from closed party lists to achieve proportionality.

  2023 New Zealand general election:
  A general election took place in New Zealand on 14 October 2023 to determine the composition of the 54th New Zealand Parliament. Voters elected 122 members to the unicameral New Zealand House of Representatives under the mixed-member proportional (MMP) voting system, with 71 members elected from single-member electorates and the remaining 51 members elected from closed party lists. Of the 72 electorates, only 71 seats were filled, with the remaining electorate MP determined in the 2023 Port Waikato by-election, due to the death of one of the general election candidates. Two overhang seats were added due to Te Pāti Māori winning six electorate seats when the party vote only entitled them to four seats, with an additional overhang seat added after the National party won the Port Waikato by-e

### web_search (SUCCESS)
Web search summary (20 results):
Here's a comprehensive summary of the latest polling data for each query:

## 1. 1News Verian Poll — September 28, 2026

The most recent 1News Verian poll, released September 28, 2026, showed a major shift with the **Green Party surging to a record high**. 
The Green Party climbed to its highest level ever, recording support that would give them 19 MPs if repeated in November's election, with the Greens' surge to 16% coming as National and Labour continue to drop, with both major parties on just 28%.


**Seat projections:** 
National and Labour are tied on 34 seats each, the Greens have 19, ACT 14 and NZ First 11.
 
The Opportunity Party, on 6% of the party vote, takes seven seats.


**Bloc totals:** 
The bloc of Labour, the Greens, Te Pāti Māori and Opportunity would have 65 seats, compared to 59 for the right bloc of National, NZ First and ACT.
 This calculation 
is based on Te Pāti Māori holding its four seats and Mariameno Kapa-Kingi, running for her freshly launched Te Tai Tokerau Party, retaining the Te Tai Tokerau seat.
 Notably, 
neither bloc reaches a majority in either scenario without Opportunity's MPs.


**Preferred PM:** 
Christopher Luxon was preferred as prime minister by 19% and Chris Hipkins by 14%.


**Methodology:** 
Between September 23 and September 27 2026, 1004 eligible voters were polled by mobile phone (503) and online using online panels (501), with a maximum sampling error of approximately ±3.4%-points at the 95% confidence level.


Source: [1News](https://www.1news.co.nz/2026/09/28/poll-greens-surge-to-best-result-ever-as-national-labour-slide-further/) and [NZ Herald](https://www.nzherald.co.nz/nz/politics/election-2026-green-party-surges-to-record-high-in-1news-verian-poll/LU6JE4FXSZAETCMYWNKKZPYRDE/)

---

## 2. RNZ Reid Research Poll — Most Recent: August 24, 2026

The most recent RNZ-Reid Research poll (published August 24, 2026) put **Opportunity in a kingmaker position**. 
The Opportunity Party was in the kingmaker position, with the poll showing neither the coalition nor the opposition would have the numbers to form a government without its help, having tipped over the 5% threshold for the first time.


**Seat projections:** 
Labour would have 39 seats, the Greens 12, and Te Pāti Māori 4, making 55 seats for the left bloc; National would have 35 seats, New Zealand First 15, and ACT 9, or 59 in total for the right bloc.
 
If replicated on polling day, the right bloc would have 59 seats and the left bloc 55 — both needing Opportunity's 6 to get over the line.


**Party support:** 
Labour was still in the lead, on 32.2%, though its drop of 1.8 points since July put Labour on its lowest result in this series since the election.
 
National emerged from its aborted coup with a slight bump in support, remaining stuck in the twenties, on 29.1%, up 0.4 points.


**Methodology:** 
The poll of 1000 people was conducted by Reid Research using quota sampling and weighting, through online interviews between 14-21 August 2026

You are an elite superforecaster. Produce a probability distribution over the answer to this Metaculus numeric question.

## Question
What will be the margin of seats won by Labour over National in the 2026 New Zealand general election?

## Description / Resolution Criteria
## Description
Radio New Zealand: [The most unpredictable yet most dull election campaign in recent history](https://www.rnz.co.nz/news/politics_election-2026/1319079/the-most-unpredictable-yet-most-dull-election-campaign-in-recent-history)

`{"format": "bot_tournament_question", "info": {"hash_id": "7a2f1ac16e820d42", "sheet_id": "1"}}`

## Resolution Criteria
This question will resolve as the number of seats won by the Labour Party minus the number of seats won by the National Party in the 2026 New Zealand general election.

## Range
The question's displayed range is [-15.5, 15.5] (units: seats).
The LOWER bound is OPEN: outcomes below -15.5 are possible and resolve as "below range".
The UPPER bound is OPEN: outcomes above 15.5 are possible and resolve as "above range".
Where a bound is open, your percentiles MAY lie outside the displayed range — the range is where the question author expects most of the mass, not a constraint on the answer. Do not compress your distribution to fit inside it.

## Research tool outputs (raw; the resolution-source read is the status-quo anchor)
## Tool outputs

### resolution_source (FAILED)
  Error: Unknown

### wikipedia_lookup (SUCCESS)
Wikipedia (3 lookups):

  Opinion polling for the 2026 New Zealand general election:
  


Eight polling firms have conducted opinion polls during the term of the 54th New Zealand Parliament (2023–present) for the 2026 New Zealand general election. The regular polls are the quarterly polls produced by Television New Zealand conducted by Verian and Radio New Zealand (RNZ) conducted by Reid Research, along with monthly polls by Roy Morgan and by Curia. Curia Market Research is no longer a member of the Research Association of New Zealand, following complaints and the resignation from RANZ by its principal, David Farrar. The sample size, margin of error, and confidence interval of each poll varies by organisation and date. The parliament was elected on 14 October 2023, and the election is scheduled to be held on 7 November 2026.

  2026 New Zealand general election:
  The 2026 New Zealand general election is scheduled to be held on 7 November 2026. Voters will elect 120 members to the House of Representatives under New Zealand's mixed-member proportional (MMP) voting system, comprising 71 members elected from single-member electorates and 49 members elected from closed party lists to achieve proportionality.

  2023 New Zealand general election:
  A general election took place in New Zealand on 14 October 2023 to determine the composition of the 54th New Zealand Parliament. Voters elected 122 members to the unicameral New Zealand House of Representatives under the mixed-member proportional (MMP) voting system, with 71 members elected from single-member electorates and the remaining 51 members elected from closed party lists. Of the 72 electorates, only 71 seats were filled, with the remaining electorate MP determined in the 2023 Port Waikato by-election, due to the death of one of the general election candidates. Two overhang seats were added due to Te Pāti Māori winning six electorate seats when the party vote only entitled them to four seats, with an additional overhang seat added after the National party won the Port Waikato by-e

### web_search (SUCCESS)
Web search summary (20 results):
Here's a comprehensive summary of the latest polling data for each query:

## 1. 1News Verian Poll — September 28, 2026

The most recent 1News Verian poll, released September 28, 2026, showed a major shift with the **Green Party surging to a record high**. 
The Green Party climbed to its highest level ever, recording support that would give them 19 MPs if repeated in November's election, with the Greens' surge to 16% coming as National and Labour continue to drop, with both major parties on just 28%.


**Seat projections:** 
National and Labour are tied on 34 seats each, the Greens have 19, ACT 14 and NZ First 11.
 
The Opportunity Party, on 6% of the party vote, takes seven seats.


**Bloc totals:** 
The bloc of Labour, the Greens, Te Pāti Māori and Opportunity would have 65 seats, compared to 59 for the right bloc of National, NZ First and ACT.
 This calculation 
is based on Te Pāti Māori holding its four seats and Mariameno Kapa-Kingi, running for her freshly launched Te Tai Tokerau Party, retaining the Te Tai Tokerau seat.
 Notably, 
neither bloc reaches a majority in either scenario without Opportunity's MPs.


**Preferred PM:** 
Christopher Luxon was preferred as prime minister by 19% and Chris Hipkins by 14%.


**Methodology:** 
Between September 23 and September 27 2026, 1004 eligible voters were polled by mobile phone (503) and online using online panels (501), with a maximum sampling error of approximately ±3.4%-points at the 95% confidence level.


Source: [1News](https://www.1news.co.nz/2026/09/28/poll-greens-surge-to-best-result-ever-as-national-labour-slide-further/) and [NZ Herald](https://www.nzherald.co.nz/nz/politics/election-2026-green-party-surges-to-record-high-in-1news-verian-poll/LU6JE4FXSZAETCMYWNKKZPYRDE/)

---

## 2. RNZ Reid Research Poll — Most Recent: August 24, 2026

The most recent RNZ-Reid Research poll (published August 24, 2026) put **Opportunity in a kingmaker position**. 
The Opportunity Party was in the kingmaker position, with the poll showing neither the coalition nor the opposition would have the numbers to form a government without its help, having tipped over the 5% threshold for the first time.


**Seat projections:** 
Labour would have 39 seats, the Greens 12, and Te Pāti Māori 4, making 55 seats for the left bloc; National would have 35 seats, New Zealand First 15, and ACT 9, or 59 in total for the right bloc.
 
If replicated on polling day, the right bloc would have 59 seats and the left bloc 55 — both needing Opportunity's 6 to get over the line.


**Party support:** 
Labour was still in the lead, on 32.2%, though its drop of 1.8 points since July put Labour on its lowest result in this series since the election.
 
National emerged from its aborted coup with a slight bump in support, remaining stuck in the twenties, on 29.1%, up 0.4 points.


**Methodology:** 
The poll of 1000 people was conducted by Reid Research using quota sampling and weighting, through online interviews between 14-21 August 2026

## Question dates and conventions
Dates: opened 2026-09-28; forecasting closes 2026-09-28; resolves 2026-11-08. Only events AFTER the open date can satisfy the criteria; assume the event has NOT happened unless a source shows it has. A question phrased 'before <date>' or 'by <date>' asks whether it happens between now and that date. The deadline for the outcome is the RESOLUTION date, not the close date.

## Recent news (AskNews, raw article summaries, newest first — read these directly)
13 AskNews articles (newest first; 6 from the last 48 h, 10 from the archive). Summaries are AskNews's own; dates are publication dates.

[2026-09-28] Left Bloc in Contention if National Rules Out Opportunity Party in Latest 1News-Verian Poll
The Brief — https://www.thebrief.news/en/standard/article/24750/left-bloc-in-contention-on-latest-1news-verian-poll-if-national-rules-out-opportunity-party
  - According to a 1News-Verian poll published on 28 September 2026 and conducted from 23 to 27 September with 1,004 eligible voters (margin of error 3.4 percent), the left bloc of Labour, the Greens, and Te Pāti Māori could form a government if National Party leader Christopher Luxon maintains his refusal to work with the Opportunity Party.
  - The poll records a lift in Green Party support, with 13 percent of respondents undecided or refusing to answer.
  - Verian modeled two scenarios based on Māori electorate results: one where Te Pāti Māori wins four seats and Mariameno Kapa-Kingi's Te Tai Tokerau party wins one, and another with different splits; in both, the Opportunity Party holds the balance of power, as neither main grouping can reach the 61 seats required for a majority on its own.
  - On preferred prime minister, Christopher Luxon, Green co-leader Chlöe Swarbrick, and ACT leader David Seymour were each up two points, while Labour leader Chris Hipkins was down one point and NZ First leader Winston Peters remained steady.
  - Previous polls show Labour support at a 30-year low in June, with the June poll projecting 66 seats for the left bloc against 58 for the right bloc, while the 19 April poll projected 66 seats for the left and 58 for the right.
  - The article notes that with 13 percent undecided and a 3.4-point margin, the party-vote order could shift before writs are returned, and the 'veto' by Luxon on working with the Opportunity Party remains a critical factor in post-election coalition talks.

[2026-09-28] Greens Surge in Latest Poll as Left Bloc Eyes Government
Otago Daily Times Online News — https://www.odt.co.nz/news/national/greens-surge-in-latest-poll-opportunity-hold-balance-of-power-vlfhsmnc
  - According to new polling from TVNZ-Verian published on September 28 2026, the Green Party has surged to 16 percent support, up 4 points, while the left bloc (Labour and Greens combined) is positioned to form a government if the National Party refuses to work with the Opportunity Party.
  - The poll, conducted between September 23 and 27 2026 with 1,004 eligible voters, shows the National Party at 28 percent (down 1 point) and the Labour Party at 28 percent (down 2 points).
  - The Opportunity Party holds the balance of power in both modeled scenarios, regardless of whether Te Pāti Māori wins four Māori seats or five seats including one for Mariameno Kapa-Kingi's Te Tai Tokerau party.
  - Christopher Luxon's preferred prime minister rating rose to 19 percent, up 2 points, while Chris Hipkins fell to 14 percent, down 1 point.
  - Chlöe Swarbrick and David Seymour each gained 2 points, reaching 8 percent and 7 percent respectively.
  - Winston Peters remained steady at 7 percent.
  - The poll has a margin of error of 3.4 percent at the 95 percent confidence interval and is described as a snapshot rather than a prediction.

[2026-09-27] National and Labour Election Campaign Launches in Auckland
The Spinoff — https://thespinoff.co.nz/analysis/28-09-2026/field-notes-from-national-and-labours-campaign-launches
  - On Sunday, September 27, 2026, New Zealand's two major parties launched their election campaigns in Auckland.
  - National Party held its event at the New Zealand International Convention Center, drawing a crowd of around 1,000 people, though the large venue left several hundred seats empty.
  - Labour Party launched at the Due Drop Events Centre, a smaller venue with a capacity of 690 seats that was full to capacity and described as far more boisterous.
  - National Party leader Christopher Luxon emphasized a pledge of 'no new taxes' and received strong support from senior figures including Simeon Brown and John Key, who urged against strategic voting.
  - Luxon spoke about his mother, Kathleen, and criticized Labour for a lack of policies.
  - In contrast, Labour leader Chris Hipkins focused on the rising cost of living, accused National of lying about tax policy, and announced a new policy to write off 10% of eligible student loan balances for New Zealand-based graduates.
  - The article notes that by measuring enthusiasm, Labour was the clear winner, with refreshed faithful compared to the determined but less enthusiastic National supporters.

[2026-09-27] NZ Election 2026: Extreme 20-Day Wait for Results Predicted (opinion)
TheDailyBlogNZ — https://thedailyblog.co.nz/the-20-days-between-election-and-final-results-will-be-the-most-extreme-in-nz-political-history/
  - According to an article published in TheDailyBlogNZ on September 27, 2026, New Zealand's upcoming election is projected to create a highly contested political environment.
  - The author predicts that the National Party will lose approximately 40 electorates and secure a party vote of under 27%, which will generate an Overhang of between 6 to 8 seats while placing ACT and NZF in double digits.
  - The article argues that Labour is chasing a non-existent centrist voter in a social media hate algorithm culture, whereas the winner will turn out their base tactically without watering down their appeal.
  - Consequently, the Overhang will mean that any side winning 51% of the vote will require four political parties to form a government.
  - The author states they do not expect the election result to be known on November 7, as the final outcome must wait for Special votes to arrive on November 27, creating a 20-day period described as 'the most extreme in NZ Political History.' The article warns that 'Huge money' will be spent arguing that the TOP party must cut a deal with National, ACT, and NZF, and cautions that if the Left is not ready to battle the ensuing media war, they will be 'swallowed up and spat out by Big Money interests setting the agenda.'.

[2026-09-27] National and Labour Launch 2026 Election Campaigns with Contrasting Events and Policies
The Spinoff — https://thespinoff.co.nz/the-bulletin/28-09-2026/pledge-cards-and-bullshit-national-and-labour-officially-launch-their-campaigns
  - On Sunday, September 27, 2026, New Zealand's two major political parties officially launched their election campaigns in Auckland, marking the start of the 2026 general election.
  - National Party, led by Prime Minister Christopher Luxon, held its launch at the New Zealand International Convention Centre, which drew a headcount of roughly 750 people but left rows of empty seats visible in a venue capable of holding more.
  - The party's central message was 'no new taxes,' with Luxon being introduced as 'Mr No New Taxes' by media outlets including The Herald and RNZ.
  - National unveiled a 10-point pledge card including economic growth, compulsory KiwiSaver with higher contributions, a $1,500 Baby Boost, extended paid parental leave from 26 to 30 weeks, replacing NCEA with regular assessment, tougher consequences for criminals, faster healthcare, infrastructure upgrades, supermarket competition, and returning the books to surplus.
  - Finance Minister Nicola Willis teased the upcoming pre-election economic update on Tuesday, promising 'smaller deficits, bigger surpluses, less debt.' Meanwhile, Labour Party, led by Prime Minister Chris Hipkins, launched its campaign at the Due Drop Events Centre in Manukau, filling its 690-seat auditorium to capacity with some attendees turned away.
  - Hipkins focused his speech on the cost of living and directly challenged Luxon's recent 'cry me a river' remark about rising prices, calling it 'out-of-touch' and stating, 'It made me so angry.' Labour announced a new policy offering a 10% write-off of student loan balances for eligible graduates, with loans of $2,000 or less cleared entirely.
  - The policy faces criticism from National's Simeon Brown, who called it 'yet another expensive handout,' and ACT's David Seymour, who labeled it a 'bribe.' Hipkins also accused National of lying about Labour's tax intentions, calling it 'bullshit.' According to The Spinoff's Joel MacManus, who attended both events, Labour's launch was more energetic and enthusiastic, while National's supporters appeared 'determined, but nowhere near as enthusiastic as they were three years ago.'.

[2026-09-08] Taxpayers' Union-Curia Poll Projects Coalition Retention Despite National Decline
THE GOOD OIL — https://goodoil.news/poll-puts-coalition-on-64-seats-as-national-falls-below-30-2/
  - A Taxpayers' Union-Curia poll conducted September 8, 2026, projects that the governing coalition of National, ACT, and New Zealand First would retain 64 of Parliament's 120 seats, despite National's party vote falling 2 points to 29.0%.
  - ACT rose 3.2 points to 10.5% (winning 14 seats), while New Zealand First dropped 0.7 points to 8.4% (winning 11 seats); National is projected to win 39 seats, giving the coalition a combined party vote of 47.9%.
  - The opposition is projected to hold 56 seats: Labour fell 1.9 points to 25.9% (its lowest result since 2024 and below its 2023 election result) with 35 seats, the Greens gained 3.8 points to 13.9% with 18 seats, and Te Pāti Māori polled 2.0% with 3 seats.
  - Labour leader Chris Hipkins stated that the party's internal Talbot Mills polling places Labour above 30%, though he declined to provide the exact figure, and cited a recent RNZ-Reid Research poll showing Labour at 32.2%, while acknowledging the broader downward trend.
  - The poll methodology assumes all current parliamentary parties with an electorate seat win at least one again and assumes no overhang seats.
  - The Opportunity party fell 1.6 points to 4.5%, below the 5% threshold and assumed not to win an electorate, resulting in zero seats; the article notes that crossing 5% or winning an electorate could materially change the arithmetic.
  - Christopher Luxon remains ahead of Chris Hipkins as preferred prime minister with 21.2% to 16.9%.
  - Curia surveyed 1,000 eligible and likely voters over three days, with a maximum margin of error of ±3.1 points and 4.1% undecided.
  - The article emphasizes this is a snapshot, not a forecast.

[2026-09-06] Latest Poll Shows Coalition Leading as TOP Slips Below Threshold
The Brief — https://www.thebrief.news/en/standard/article/17935/latest-poll-puts-coalition-ahead-as-top-slips-below-threshold
  - A Taxpayers' Union-Curia poll conducted 1-3 September 2026 projects the National, ACT, and NZ First coalition will hold power with 64 seats, while the opposition Labour, Greens, and Te Pāti Māori bloc secures 56 seats.
  - The poll shows The Opportunities Party (TOP) slipping to 4.5 percent, falling below the 5 percent threshold required under the MMP system to enter Parliament without winning an electorate seat, which would cost TOP all its parliamentary seats.
  - This marks a reversal from previous polls in July and August where TOP was polling at 6.1 percent and approaching the threshold.
  - National party support fell to 29 percent, Labour dropped to 25.9 percent, while the Greens rose to 13.9 percent and ACT increased to 10.5 percent.
  - New Zealand First slipped to 8.4 percent, and Te Pāti Māori gained 0.5 points to 2 percent, though it is assumed to win three electorates, keeping it in Parliament.
  - Christopher Luxon remains the preferred prime minister at 21.2 percent, down 2 points, while Chris Hipkins stands at 16.9 percent, down 2.9 points.
  - The poll surveyed 1,000 adult New Zealanders with a maximum margin of error of ±3.1 percent.
  - The results raise questions about whether TOP's decline is genuine or within the margin of error, as their true support could range from 1.4 to 7.6 percent.
  - If TOP clears the threshold, their seats would likely come from the coalition, whereas falling below it secures the coalition's majority.
  - Te Pāti Māori aims to use the MMP overhang mechanism by winning three electorate seats to increase total seats and reduce the coalition's governing requirement.

[2026-09-06] Labour and National Polling Collapse in New Zealand
RNZ — https://www.rnz.co.nz/news/politics_election-2026/1278458/labour-and-national-polling-collapsing-both-under-30-percent
  - According to a Taxpayers' Union-Curia Poll conducted between Tuesday 01 September and Thursday 03 September 2026, New Zealand's major parties are polling below 30 percent, with the current coalition government still projected to retain power.
  - The poll, conducted by Curia Market Research for the Taxpayers' Union, found that the National Party fell two points to 29 percent, while the Labour Party dropped 1.9 points to 25.9 percent, its lowest result since July 2024.
  - The Quilae Wong-led Opportunity party fell from 6.1 to 4.5 percent, remaining below the 5 percent threshold required to enter parliament.
  - Conversely, the Greens gained 3.8 percentage points to 13.9 percent, and ACT increased by 3.2 points to 10.5 percent.
  - In terms of projected seats, the coalition (National, ACT, and NZ First) would hold 64 seats compared to the opposition (Labour, Greens, and Te Pāti Māori) on 56 seats, assuming Te Pāti Māori wins three seats.
  - Christopher Luxon remained the preferred prime minister at 21.2 percent, ahead of Chris Hipkins at 16.9 percent.
  - The poll surveyed 1000 adult New Zealanders with a maximum margin of error of +/- 3.1%.

[2026-08-21] Victorian State Election 2026: Coalition, Labor, and One Nation Battle for 45 Seats
Australian Broadcasting Corporation — https://www.abc.net.au/news/2026-08-22/coalition-path-to-victory-victoria-state-election-2026/107063196
  - With the Victorian state election scheduled for November 28, less than 100 days away, the political landscape is shifting as the Liberal and National Coalition, after 12 years in opposition, attempts to topple the Labor government.
  - According to the article, the lower house comprises 88 seats, requiring 45 to govern; Labor currently holds 56 seats (with the Greens holding 3), while the Coalition holds 29 (20 Liberals, 9 Nationals).
  - Recent opinion polls suggest One Nation will be a significant player, potentially preventing any single party from reaching the 45-seat majority, making a minority government a 'genuine prospect'.
  - Labor insiders fear the party could lose upwards of 20 seats, particularly in Melbourne's west, north-west, and eastern suburbs, as well as regional areas.
  - New Premier Ben Carroll, who replaced Jacinta Allan last month, has made crime a centrepiece of his leadership to appeal to voters in at-risk electorates like Sunbury, Sydenham, Melton, and Yan Yean.
  - Labor faces specific risks in seats like Niddrie (Carroll's seat, held with a 6.7 per cent margin), Pakenham, Bayswater, and Bass.
  - The Coalition, led by Jess Wilson, must win at least 16 additional seats to govern alone, a task made 'steep' by One Nation's presence.
  - Research briefed to Labor ministers in June claimed One Nation could take as many as 17 regional and outer-suburban seats from the Coalition, with six Nationals-held seats identified as 'easy One Nation gains'.
  - The Greens are also challenging Labor in inner-city seats like Richmond, Northcote, and Footscray, though Labor believes the announcement of a royal commission and a recent teacher pay deal will help them hold ground.
  - Early polls by Resolve and Redbridge showed a slump in Labor's primary support following the leadership change, though this was within the margin of error.

[2026-08-07] Taxpayers' Union Poll: National Party Leads with Narrow Margin in 2026 Election Forecast
RNZ — https://www.rnz.co.nz/news/chinese/927715
  - According to a new poll conducted by the Taxpayers' Union and commissioned to Curia, released on August 7, 2026, the National Party is projected to form a coalition government with a slim majority.
  - The National Party's support rose by 0.5 percentage points to 31%, making it the largest party in parliament.
  - In contrast, the Labour Party's support fell by 3.7 percentage points to 27.8%.
  - The poll indicates that the ACT Party increased its support to 7.3%, while the Green Party declined slightly to 10.1% and the NZ First Party dropped 1.7 percentage points to 9.1%.
  - Notably, the Future New Zealand Party (Opportunity Party) is projected to enter parliament with 6.1% support, though National has confirmed it will not cooperate with them, leaving them as part of the opposition.
  - The Māori Party recorded its worst performance since March 2023, with support falling 1.9 percentage points to 1.5%.
  - Based on these figures, the coalition of National, ACT, and NZ First would hold 61 seats, compared to 59 seats for the opposition bloc (including the Opportunity Party).
  - In terms of prime ministerial preference, Christopher Luxon of the National Party leads with 23.2% personal support, up 4.1 percentage points, against Labour's Chris Hipkins at 19.8%, who saw a slight decline.
  - The survey was conducted between last Saturday and Tuesday with a sample size of 1,000 voters and a margin of error of ±3.1%.

[2026-08-07] National Leads in New Poll as Opportunity Party Enters Parliament
Otago Daily Times Online News — https://www.odt.co.nz/news/national/coalition-just-holding-on-opportunity-could-enter-parliament-poll-shows-f6xw42lf
  - A new Taxpayers' Union Curia poll released on Friday indicates the National-led coalition remains in a position to form a government, albeit by a narrow margin.
  - The poll shows the Opportunity Party has broken the 5% threshold with 6.1%, though it would join the Opposition as National ruled out working with them.
  - National's support rose 0.5 points to 31%, making it the most popular party, while Labour dropped 3.7 points to 27.8%.
  - The coalition parties (National, ACT, and New Zealand First) would secure 61 seats, compared to the Opposition's 59 seats.
  - ACT gained slightly to 7.3%, while Te Pati Maori fell to 1.5%, its lowest since March 2023.
  - In preferred Prime Minister polling, National's Christopher Luxon leads with 23.2%, up 4.1 points, against Labour's Chris Hipkins at 19.8%.
  - The general election is scheduled for November 7.
  - The poll surveyed 1,000 respondents between Saturday and Tuesday with a margin of error of +/- 3.1%.

[2026-06-14] Labour Holds Lead Over National in Latest NZ Poll as ACT and NZ First Gain
RNZ — https://www.rnz.co.nz/news/political/598196/national-slides-below-30-percent-in-latest-poll-act-and-nz-first-gain
  - According to the latest The Post/Freshwater Strategy poll conducted between June 5-11, 2026, with a sample of 1,038 eligible voters and a margin of error of +/- 3%, New Zealand's major parties have seen shifts in support.
  - Labour leader Chris Hipkins leads with 44% support as the preferred Prime Minister, down two points since February, while National leader Christopher Luxon has gained ground to 37%, up three points.
  - In terms of party support, Labour stands at 35%, down two points, maintaining a lead over National, which has dropped one point to 29%.
  - Meanwhile, New Zealand First has risen to 12% (up one point) and ACT has increased to 8% (up two points).
  - The Greens remain steady at 10%, and Te Pāti Māori holds 2%.
  - Based on these figures, the coalition parties would hold 68 seats in parliament, compared to 62 for the left bloc.

[… older articles omitted]

## Research reports (one per dimension: web search + page reads, dated, with sources)
### CURRENT STATE
**Current value (as of 2026-09-28):** The latest published poll, the 1News-Verian survey (fieldwork 23–27 September 2026, n=1,004, released 28 September 2026), shows 
National and Labour level at 28 percent
 in party vote, translating into 
National and Labour tied on 34 seats each
. On this poll's modelled seat allocations the margin (Labour minus National) is **0 seats** in both of Verian's scenarios (with or without an overhang from the Māori electorates) — 
National and Labour are tied on 34 seats each, the Greens have 19, ACT 14 and NZ First 11, and the Opportunity Party, on 6% of the party vote, takes seven seats in each version
.

- [2026-09-28] Latest 1News-Verian poll: 
Labour on 28%, down two points since the last poll, and National down one point, also on 28%; the Green Party is on 16%, up four points
 — https://www.thepost.co.nz/politics/361088727/1news-verian-poll-has-both-major-parties-28-greens-sky-rocket
- [2026-09-28] On the 34-34 party-vote-only reading: 
the coalition of National, Act and NZ First would gather 59 seats between them if an election was held when polling took place, while the bloc of Labour, the Green Party, Te Pāti Māori and Opportunity would trump that with 65 seats
 (this bloc figure includes Green/TPM/Opportunity seats, not just Labour vs National) — https://www.nzherald.co.nz/nz/politics/election-2026-green-party-surges-to-record-high-in-1news-verian-poll/LU6JE4FXSZAETCMYWNKKZPYRDE/
- [2026-09-28] Verian offered a second seat scenario assuming all 2023 electorate winners hold their seats; in that version too 
National and Labour are tied on 34 seats each
 — https://www.1news.co.nz/2026/09/28/the-two-ways-of-calculating-seats-for-the-1news-verian-poll/
- [2026-09-28] An independent aggregator (Stuff) framing of the same poll: 
the latest 1News-Verian poll will see the left bloc of Labour (34 seats), the Greens (19 seats), Te Pāti Māori (4 seats) and Mariameno Kapa-Kingi's Te Tai Tokerau party (1 seat), with 58 seats in the house
 versus National's 34 — https://www.stuff.co.nz/politics/361036026/nz-election-2026-live-updates-politics-campaign-news
- [2026-09-15/16] The Kiwi Polls "Quasimodo" aggregator model (last snapshot fetched, dated update 16 September 2026) put average simulated seats at National 37, Labour 36 — a **-1 seat margin for Labour** (National ahead), within a projected 123-seat overhang Parliament — https://www.kiwipolls.co.nz/
- [2026-08-31] Roy Morgan (31 August 2026): 
both blocs held steady, with the National-led parties on 49% and the Labour-led Opposition on 41%; the Opportunity Party rose 1.5 points to a record 9.5%, its fourth straight monthly gain
, with 
Labour falling to 24%, its weakest showing since September 2024
 — https://www.kiwipolls.co.nz/
- [2026-08-12] 1News-Verian (fieldwork 8–11 Aug 2026): 
Labour on 30%, down two percentage points, while National was unchanged on 29%; on those results, Labour would receive 38 seats and National 36
 — a **+2 seat Labour margin** — https://www.newstalkzb.co.nz/news/politics/labour-leads-national-as-opportunity-party-hits-8-in-1news-verian-poll/
- [2026-06-23] 1News-Verian (fieldwork ~June 2026): 
Labour is down five points to 32 percent, but could form a government with the Greens and Te Pati Maori; between them the left bloc would have 64 seats to the right bloc's 60 seats, assuming Te Pāti Māori retains all six electorate seats
 (bloc figures, not a direct Labour-vs-National seat count) — https://www.rnz.co.nz/news/politics/617657/left-bloc-leads-in-latest-poll-despite-labour-slump

**One year earlier (reference point, June 2025 — the closest dated Labour-vs-National seat estimate available):** 
National sat at 34%, Labour dipped 3 points to 29%, and the 1News-Verian polling left Labour with 37 seats
; National's seats were reported elsewhere in the same period at roughly 43 in a similar Verian poll — implying a National lead of roughly 5–9 seats over Labour at that time (i.e., a **negative** Labour-minus-National margin, somewhere around -5 to -9), a much wider gap than today's near-zero margin.

**Trajectory:** Over 2025–2026 the Labour-National gap has narrowed sharply and repeatedly flipped sign between polls/pollsters — Labour led by several seats in June and August 2026 polls, while National edged ahead in some poll aggregations (e.g., Kiwi Polls' mid-September average), and by the newest 28 September 1News-Verian poll the two majors are tied on both vote share (28% each) and modelled seats (34 each). 
Labour's standing marks its lowest result since February 2024, continuing a downward trajectory over recent polls
, while National has also slid to 
its lowest level since the final 1News poll before Judith Collins lost the party's leadership in 2021
.

**Next data point:** Polling agencies (1News-Verian, Taxpayers' Union-Curia, Roy Morgan, Freshwater Strategy/The Post, Reid Research) typically release monthly; with the election on 7 November 2026 and advance voting from 26 October, further polls are expected through October, with the actual seat result only final well after election day given 
the model projects a three-seat overhang, which expands Parliament from 120 seats to 123 and raises the majority threshold from 61 to 62
 and special votes must be counted before the writ is returned.

BOTTOM LINE: The most recent and most authoritative primary data point — the 1News-Verian poll released 28 September 2026 — shows Labour and National dead level on both party vote (28% each) and projected seats (34 each) in both of Verian's modelled scenarios, implying a current-value margin of 0, a dramatic narrowing from the roughly 5–9 seat National lead implied by comparable polling around mid-2025.

NOT FOUND: An official, single "resolution source" data feed (API/CSV) for seat margins was not available to fetch directly — all figures come from pollster press releases and news reporting of poll-derived seat projections; no exact 1News-Verian poll from September 2025 (12 months prior) was located, so the "one year ago" comparison uses the closest available (June 2025) data point instead.

### BASE RATE
**Note on scope:** New Zealand holds general elections roughly every 3 years, so there is no election-night-to-final-count window in the "prior three years" (2023–2025) — the last three actual instances of this exact calendar dynamic (election day, followed by ~3 weeks of special-vote counting to a final declared result) are the 2017, 2020 and 2023 general elections. These are the closest analogues to the Sept 28–Nov 8 2026 window, since 2026's election day (Oct 17) plus special-vote counting will similarly resolve before Nov 8.

- [2023-11-03] Final 2023 results vs election-night (Oct 14) preliminary count: 
the National Party has 48 seats compared with 50 on election night, and there are no changes for the Labour Party which has 34 seats
. The Labour–National seat gap did not move during the count (stayed at −14), but National's own total fell as seats shifted to Te Pāti Māori and the Greens: 
the National Party has two fewer seats, Te Pāti Māori has gained two, and the Green Party has gained one more
. https://elections.nz/media-and-news/2023/official-results-for-the-2023-general-election

- [2023-11-03] 
The number of special votes was 603,257 - 20.9% of total votes (17% in 2020 & 2017), with turnout of people who were enrolled to vote at 78.2%
, the highest special-vote share of the three cycles reviewed. https://elections.nz/media-and-news/2023/official-results-for-the-2023-general-election

- [2020-11-06] Final 2020 results vs election-night: 
the Labour Party has 65 seats compared with 64 on election night, and the National Party has 33 seats compared with 35 on election night
. The Labour-minus-National margin widened by 3 seats during the count (from +29 on the night to +32 final), driven entirely by National losing 2 seats to Labour on special votes. https://elections.nz/media-and-news/2020/2020-general-election-official-results

- [2020-11-06] Three electorates flipped from National to Labour on specials: 
Labour Party candidate Priyanca Radhakrishnan won Maungakiekie with a majority of 635 votes over National Party candidate Denise Lee; Willow-Jean Prime won Northland with a majority of 163 votes over National candidate Matt King; Emily Henderson won Whangārei with a majority of 431 votes over National candidate Shane Reti
 — all three seats moved from National column to Labour column. https://elections.nz/media-and-news/2020/2020-general-election-official-results

- [2023-11-03] Final 2017 results vs election night (per Elections NZ retrospective in 2017 release): 
the National Party has 56 seats compared with 58 on election night, the Labour Party has 46 seats compared with 45 on election night, and the Green Party has 8 seats compared with 7 on election night
. The Labour-minus-National margin moved from −13 on the night to −10 in the final count, a 3-seat swing toward Labour. https://elections.nz/media-and-news/2017/new-zealand-2017-general-election-official-results

- [Historical pattern] 
Since 1999, the parties that have gained seats between election day and the declaration of the official results have all been left-leaning
, and specifically for National vs Labour: 
in 2017 National's initial seat allocation was reduced by two while Labour and the Greens both gained one seat each, and in 2020 National had two fewer seats than the election night result predicted
 once specials were counted. This is a consistent structural pattern (not random noise) favoring Labour's margin over National by roughly 2–4 seats between election night and final declaration.

- [Largest surprise on record] 
the largest special-vote seat shift was in 1999, when special votes affected five parties and 14 seats, removing the Labour-Alliance coalition's election-night majority and giving the Green Party — which had zero seats on election night — an entitlement to seven seats
, showing that in extreme cases the swing can be far larger than the typical 2–4 seat range seen in 2017/2020/2023.

- [Polling accuracy — 2017] 
the last RNZ poll of polls average before the 2017 election was strikingly close to the final result: National polled at 45.1% vs actual 44.4%, and Labour at 37.2% vs actual 36.9% — both well within 1 percentage point of the actual outcome
, though NZ First's late kingmaker role still produced a change of governing bloc.

- [Polling accuracy — 2023] 
the polls ahead of the 2023 election consistently underestimated the popularity of the right-wing parties; the most accurate polls (Talbot Mills, Curia/Taxpayers' Union, and Verian/Kantar for 1 News) all still understated the right's margin over the left
, indicating a systematic recent polling bias against National/ACT that, if repeated, would push actual results further toward National than current polling implies.

- [2026 baseline] Current 1News-Verian polling (23–27 Sept 2026, n=1,004, MoE 3.4%) has National and Labour essentially tied at the party-vote level (National 28%, Labour 28%), with modeled seat projections putting the left bloc (Labour, Greens, Te Pāti Māori) around 66 seats and the right bloc around 58, per the AskNews summaries already gathered — a far tighter contest than any of the three prior elections reviewed here, where the final Labour-minus-National seat margin was −14 (2023), +32 (2020), and −10 (2017).

BOTTOM LINE: Over the last three NZ elections, the Labour-minus-National seat margin moved by roughly +1 to +3 seats toward Labour between election night and the final declared result (with an extreme historical outlier of far more in 1999), due to a persistent left-lean in special votes; but polling has recently understated National's true strength (2023), while in 2017 it was very accurate — so the historical record suggests only a modest, Labour-favoring adjustment should be expected between now and the Nov 8 resolution date, layered on top of whatever the Oct 17 election night margin turns out to be.

NOT FOUND: No exact analogue exists for a "Sept 28–Nov 8 window in each of the prior three calendar years" since New Zealand does not hold elections annually; I substituted the three most recent general elections (2017, 2020, 2023) as the closest comparable events. Could not find a systematic post-election-night-to-final-count breakdown further back than 1999 in the sources reviewed.

### KEY FACTORS
- **[2026-01-21] Election-day/resolution timing mismatch (mechanism, controlled by the Electoral Commission, not campaigns):** The government confirmed 
the House will rise on 24 September, Parliament dissolved on 1 October, writ day on 4 October, and advance voting beginning 26 October
. Crucially, 
official results are declared on November 27, with the writ returned by December 3
 — nearly three weeks *after* this question's November 8 resolution date. This means the resolving figure will almost certainly be the **preliminary election-night count** (7 November), not the specials-inclusive final declaration, cutting against the pro-Labour special-votes shift documented in the base-rate report (2017/2020/2023 all moved 2-4 seats toward Labour only after specials were counted).

- **[2026-07-27/28] Luxon's rule-out of the Opportunity Party (controlled by Luxon/National caucus):** Luxon 
appeared to rule out National working with the Opportunity Party, saying his party didn't want to "do business" with anyone seeking to impose new taxes
, later confirmed by his office as 
a formal rule-out, citing Opportunity's land tax advocacy
. This doesn't move the Labour-vs-National seat count directly, but it structurally caps National's bloc: as one analysis noted, 
a rule-out does come with risks — if National, NZ First and Act don't have the numbers to govern, Luxon will either need an embarrassing about-face, or has unilaterally consigned the right bloc to opposition
. Multiple commentators (Peter Dunne, 2026-07-31) argue the rule-out is "at best conditional" and could be reversed post-election, similar to Luxon's 2023 reversal on NZ First — a live wildcard for late bloc-formation but not the Labour-National seat gap itself.

- **[2026-08-10] Hipkins' reciprocal rule-out of NZ First (controlled by Labour caucus):** 
Labour leader Chris Hipkins has ruled out a governing deal with New Zealand First
, with 
the rule-out discussed at caucus, and not a single member of Labour's caucus disagreeing
. This locks Labour out of NZ First's ~9-11% support pool as a coalition option, meaning Labour's path to government (and thus its incentive/campaign focus) depends almost entirely on Green/TPM/TOP arithmetic rather than any NZ First swing — a factor that shapes campaign strategy but not the two-party seat count directly.

- **[2026-09-28] Māori electorate seat allocation is the single largest swing factor in the seat-margin calculation, not vote share (controlled by ~5 Māori-electorate voters, not national polling):** The two 1News-Verian scenarios published today show identical party-vote-derived seats for Labour and National (34-34) but diverge sharply in total-bloc composition depending solely on 
whether any party that won an electorate in 2023, including Te Pāti Māori, wins at least one again — which still gives National and Labour tied on 34 seats each
 while the overhang shifts between 0 and 4 seats. Because the Labour-vs-National number itself doesn't move across these scenarios, the real "swing" risk for the margin lies in individual electorate battlegrounds (see below), not in TPM/overhang mechanics.

- **[2026-09-28] Poll-to-poll seat volatility over just the past seven weeks (numbers behind the headline, same-day sources):** Compare the 11 August Taxpayers' Union-Curia model — which had 
Labour on 41 seats, National 39, and NZ First 15
 (Labour +2) — against today's tied 1News-Verian 34-34 model and The Post's framing that 
Labour is able to govern with a one-seat majority of 61 while National would have 34 seats, ACT 13 and NZ First 11 — a total of 58 seats
. The Labour-minus-National margin implied by pollster-derived seat models has swung from roughly +2 to 0 in under two months, underscoring that no single poll should be treated as settled ahead of the six-week campaign remaining.

- **[2026-09-28] Undecided/soft vote share remains large and could break either way:** Today's 1News-Verian poll recorded 
a sample of 1004 eligible voters, with a maximum sampling error of about ±3.4% at the 95% confidence level
, and roughly 13% of respondents undecided per the earlier AskNews summary — a bloc large enough, at current 0-2 seat margins, to flip the sign of the Labour-National gap on its own by decision day.

- **[2026-09-28] Electorate-seat mechanics (National's 43 held electorates vs Labour's list-heavy 2023 base) are an under-modelled risk to the margin:** The election2026.nz tracker's own methodology flags that 
the "Full right overhang" preset loads a split-ticket scenario where National's party vote is moved to ACT and NZ First, half each, with the 43 electorates it won in 2023 kept as pure overhang
 — a scenario that would inflate National's total seats (and thus swing the margin against Labour) independent of any change in Labour's or National's headline party-vote share, because New Zealand's MMP overhang rules add seats for electorate wins beyond a party's proportional entitlement.

BOTTOM LINE: The Labour-minus-National margin is currently pinned near zero by tied party-vote polling, but the resolution date (Nov 8) falls before New Zealand's official specials-inclusive count (due Nov 27), so the number that resolves this question is very likely the preliminary election-night tally — removing the modest pro-Labour "specials bump" seen in 2017/2020/2023 and leaving the outcome hostage to (a) whichever way ~13% undecided voters break, and (b) electorate-level overhang effects tied to National's 43 held seats and the fluid Māori electorates.

NOT FOUND: No official Electoral Commission or party statement confirming whether this specific question's compilers intend to use the preliminary (election-night) count or wait for a later confirmed count published after Nov 8; no seat-by-seat marginal-electorate list (e.g., the closest 5-10 seats decided by <1,000 votes in 2023) was located to quantify overhang risk precisely.

### EXPERT VIEW
- [2026-09-28] The aggregator election2026.nz, combining 1News–Verian, RNZ–Reid Research, Taxpayers' Union–Curia, Talbot Mills, The Post/Freshwater Strategy and Roy Morgan polls with house-effect adjustment, projects 
a house-effect-adjusted polling average, MMP seat projection, coalition builder and election-day forecast for 7 November 2026
, with the table showing National on 29.6% party vote (36 projected seats) and Labour on 29.4% (also 36 seats) — a projected margin of roughly zero seats, essentially a dead heat. https://election2026.nz/

- [2026-09-28] The same tracker's Monte Carlo-style simulation states the right bloc (National-ACT-NZ First) wins outright in "38.6% of 10,000 simulated elections and the left bloc (Labour, Greens, Te Pāti Māori) in 16.0%", with a further "left, only with TOP" scenario at 31.6%, reflecting how sensitive the majority is to the Opportunities Party (TOP) rather than to Labour's own seat margin over National. https://election2026.nz/

- [2026-09 fetched] Polymarket's "NZ Election: Labour Party # of seats?" market (opened 29 Apr 2026, $23,905 traded) currently prices "30-34" at 31% and "35-39" at 34%, with only 7% for "45-49" and negligible weight above 50, i.e. traders see Labour most likely finishing in the low-to-mid 30s. https://polymarket.com/event/nz-election-labour-party-of-seats

- [2026-09 fetched] Polymarket's parallel "NZ Election: National Party # of seats?" market ($21,608 traded) prices "35-39" at 37% and "30-34" at 34%, i.e. National is priced very slightly higher than Labour on average, but the two distributions heavily overlap in the 30-39 seat band. https://polymarket.com/event/nz-election-national-party-of-seats

- [2026-09 fetched] Polymarket's own market-context note describes 
"Recent polling averages place Labour near 29% party vote, projecting roughly 35 seats in New Zealand's 122-seat MMP parliament ahead of the November 7 election, with individual surveys showing the party between 28% and 32%. National sits at comparable levels around 28-29%"
, underscoring a razor-thin implied seat margin between the two majors. https://polymarket.com/event/nz-election-labour-party-of-seats

- [2026-09-28 search snippet] A separate Polymarket market, "NZ Election: Popular Vote Margin of Victory?", currently leads with "Labour 0-5%" at 35% ahead of "National 0-5%" at 18%, implying traders lean toward Labour narrowly winning the party-vote share, though this does not directly translate into seat margin given coalition/overhang dynamics. https://polymarket.com/event/nz-election-popular-vote-margin-of-victory

- [2026-06-13 dated] A lower-volume, less-reliable Polymarket market tracked by aggregator site Lines.com ("Will Labour win 55+ seats") shows only 46% for a 55+ outcome, but this market has "Total volume is only $757 with zero traded in the last 24 hours," making it a weak signal versus the higher-volume seat-bracket markets. https://www.lines.com/prediction-markets/elections/nz-election-labour-party-of-seats

- [2026-08-12] The Taxpayers' Union-Curia poll's own seat projection (Aug 1-4 fieldwork) gave the National-ACT-NZ First bloc 61 seats (a bare one-seat majority) with National on 31.0% and Labour on 27.8%, a poll-based scenario implying a modest National seat lead over Labour at that time, though this was two months before the current, much tighter polling. https://sundayguardianlive.com/world/new-zealand-election-polls-2026-pm-christopher-luxons-national-coalition-leads-with-61-seats-labour-loses-support-opportunity-party-crosses-5-heres-what-opinion-polls-say-curia-vs-roy-morgan-258705/

- [2026-08-04] Roy Morgan's July 2026 poll-based seat model instead had the National-led bloc's seats falling and 
the Labour-led Parliamentary Opposition winning 51 seats (down four seats)
 against a shrunken National-led bloc, illustrating how volatile pollster-specific seat models have been over just the past two months — a sign that any single-source projection should be treated cautiously this cycle.

- [2026-09-28] Independent tracker predictor.nz also runs an MMP seat-projection model assuming 
"Te Pāti Māori is the only party that may enter overhang and that the ACT Party will win at least one electorate seat... All other parties are assumed to be bound by the 5% MMP threshold"
, but as fetched it does not surface a single current headline Labour-minus-National number distinct from election2026.nz's near-zero projection. https://predictor.nz/

- [historical accuracy] election2026.nz's methodology section states its simulation approach was "calibrated by replaying this exact method against the 2014, 2017, 2020 and 2023 elections" and that "at this range, history says a major party's polling can miss by five points or more," meaning even the best current aggregator projection (~0 seat margin) carries wide uncertainty bands of several seats in either direction six weeks out. https://election2026.nz/

BOTTOM LINE: The best current, higher-volume market/aggregator evidence (election2026.nz's polling-average seat model and both Polymarket seat-bracket markets, all dated 2026-09-28/fetched today) converges on a near-tied outcome, with Labour and National both projected in the mid-30s seats and a margin close to zero (possibly single digits in either direction); a minority popular-vote-margin market leans slightly toward Labour, but historical volatility in NZ seat forecasts (5+ point misses per the tracker's own backtesting) means the sign and size of the final Labour-minus-National margin remains genuinely uncertain.

NOT FOUND: No single, high-confidence "expert consensus" point-forecast for the exact Labour-minus-National seat margin (e.g., from a major NZ psephologist or bank) was found; no evidence of a dedicated Kalshi or iPredict market on this specific margin; could not confirm whether the Lines.com aggregator's 46% "55+ seats" figure reflects a real, current Polymarket price given its very low volume and possible staleness.

### HIGH CASE
- [2026-04-19] The single largest Labour-over-National gap recorded anywhere in the 2026 cycle came from 1News-Verian: 
support for the National Party and Prime Minister Christopher Luxon crashed to new lows, while Labour overtook National for the first time in the poll since before the 2023 election, climbing into a seven-point lead
. 
The left bloc of Labour, the Greens and Te Pāti Māori would have had 66 seats compared to 58 for the right bloc, with National losing 12 seats
 on those numbers — the evidence-backed template for how far the gap can widen when National is in crisis (this poll came "amid the fuel crisis and after the PM's Cabinet reshuffle," per the same report). This is the ceiling case the "high scenario" needs to re-emerge.
- [2025-12-05/10] An earlier still-wider party-vote gap appeared in The Post/Freshwater Strategy poll (5–10 Dec 2025), which had National on 30% against Labour's 38% — an 8-point lead, per the Wikipedia-compiled polling table — showing double-digit-point Labour leads are not unprecedented this term, even if none have persisted more than a month.
- [2026-08-14/21] The most recent poll giving a direct Labour-vs-National seat count (rather than a bloc figure) is RNZ-Reid Research: 
Labour would have 39 seats… National would have 35 seats
 — a +4 Labour margin, the highest concrete Labour-minus-National seat number found in the dataset outside the April Verian bloc-only estimate.
- [2026-03-12/20] The same RNZ-Reid Research series in March had 
Labour with the biggest party-vote share at 35.6%, while National sat on 30.8%
, a ~4.8-point party-vote lead, but the seat model still produced a 60-60 bloc tie because 
New Zealand First and the Greens picked up party vote while everyone else suffered drops
 — illustrating that a high Labour-National margin requires Labour's gains to come disproportionately at National's direct expense, not merely a general anti-government swing that also lifts NZ First/Greens.
- [2026-09-27] A structural tailwind for a wide Labour-National gap: Labour's own campaign launch coverage notes the party unveiled 
a new cost-of-living policy – wiping 10% of student loan debt for those who didn't benefit from Labour's previous fees-free policy
, a retail policy explicitly targeted at price-sensitive, youth-leaning voters who are also the demographic most likely to swing directly from National-curious "soft" voters to Labour rather than to a minor party.
- [2026-09-28] Working against the widest-gap scenario, the newest 1News-Verian poll shows the anti-government swing is currently being captured by the Greens, not Labour: 
the Greens climbed up four points in the latest poll, while Labour was on 28%, down two points
 — meaning momentum that in April flowed to Labour is now diffusing to a third party, which caps rather than widens the direct Labour-National seat margin even if the overall left bloc grows.
- [2026-04-08] A Taxpayers' Union-Curia snapshot cited by an aggregator found 
Labour leading National for the first time since 2022, with the left bloc (Labour, Green, Te Pāti Māori) holding a majority over the right bloc
, corroborating that April 2026 was a genuine trough for National across multiple pollsters simultaneously, not a single-pollster outlier — raising the credibility of the ~7-9 point gap as a real, replicable "high" state rather than noise.
- [2026-04-19] The same April period produced record-setting personal numbers for Hipkins: an aggregator noted 
Hipkins leads by 8 points — no NZ PM has ever recovered from trailing the opposition leader by this margin 7 months from election day in modern polling history
 (referring to Luxon's deficit), indicating that if Luxon's personal ratings deteriorate again to April levels, the party-vote gap that produces a high Labour-National seat margin has historical precedent within this very cycle.
- [2026-09-28] Countervailing mechanism: even in today's most left-favourable poll read, 
National and Labour are tied on 34 seats each… the Opportunity Party, on 6% of the party vote, takes seven seats
, and 
neither bloc reaches a majority in either scenario without Opportunity's MPs
 — showing the direct Labour-National margin has structurally compressed toward zero as the election nears, and a jump back to the April-style 7-9 point gap would require a fresh shock (economic, scandal, or campaign misstep) rather than simple trend continuation.
- [2026-09-28] Special-votes precedent (already noted by colleagues) adds a modest, reliably left-leaning tailwind on top of any election-night margin, which in a high scenario would stack on an already-wide Labour lead rather than create one from nothing.

BOTTOM LINE: The evidence-backed high end of the range is a Labour-National seat margin in roughly the +8 to +12 seat band, precedented by the 19 April 2026 1News-Verian poll (7-point party-vote lead, National down 12 seats, left bloc 66–58) and by the 5–10 December 2025 Post/Freshwater poll (8-point lead), both of which would require a comparable National-specific collapse (economic shock, leadership crisis, or a decisive cost-of-living backlash) to recur between now and 7 November, since the current (28 September) polling has the two majors dead level and the anti-government swing is currently flowing to the Greens rather than Labour directly.

NOT FOUND: No poll in the 2026 cycle has shown a direct Labour-minus-National seat margin above roughly +4-5 (RNZ-Reid Research August) using an explicit per-party seat count; the wider +7 to +12 figures come from bloc-level (Labour+Greens+TPM vs National+ACT+NZF) comparisons or party-vote-only leads, so a precise historical "high" for the pure two-party seat margin in this cycle could not be confirmed beyond the RNZ August figure.

### LOW CASE
- [2026-09-07] The Sept 1–3 Taxpayers' Union–Curia poll — the most National-favourable poll of the whole cycle — put National at 29.0% (down 2.0) and Labour at 25.9% (down 1.9), translating into a seat model of 
National down 2.0 points to 29.0%, while Labour is down 1.9 points to 25.9%
 and, per the Curia seat table, National 39 seats vs Labour 35, i.e. a **−4 seat margin for Labour** under this poll's methodology (no overhang assumed). https://www.taxpayers.org.nz/sept26_pollnztughjkuhn

- [2026-09-07] The same Curia release shows Labour's 25.9% is 
its lowest result in this poll since July 2024, and even lower than the 26.9 point thrashing it got at the 2023 election
, while National's 29% is 
also significantly lower than the 38 percentage points it won at the last election
 — evidence a genuinely bad Labour showing is achievable within the current polling range even as both majors fall from 2023 levels. https://www.rnz.co.nz/news/politics_election-2026/1278458/labour-and-national-polling-collapsing-both-under-30-percent

- [2026-09-07] Under this poll's bloc arithmetic, 
the combined projected seats for National, NZ First and ACT is 64, up three seats, while the combined projected seats for Labour, the Greens, Opportunity and Te Pāti Māori was 56 (down three seats)
 — an 8-seat bloc gap, showing the low-scenario mechanism is not just a Labour-National swing but Opportunity collapsing below threshold (from 6.1% in August to 4.5%), which pushes its ~8 "wasted" seats disproportionately toward National's bloc. https://www.interest.co.nz/public-policy/140143/taxpayers-union-curia-poll-shows-labour-slumping-its-lowest-result-july-2024

- [2026-08-24] RNZ–Reid Research (14–21 Aug 2026) also shows a mid-cycle low-scenario data point: 
Labour would have 39 seats, the Greens 12, and Te Pāti Māori 4, making 55 seats for the left bloc in total. National would have 35 seats, New Zealand First 15, and ACT 9, or 59 in total for the right bloc
 — a +4 Labour seat lead individually, but the right bloc still 4 seats ahead overall, illustrating that even polls with Labour narrowly ahead of National can still produce a low/negative outcome for the broader left coalition. https://www.rnz.co.nz/news/politics_election-2026/1093636/rnz-reid-research-poll-opportunity-in-kingmaker-position-luxon-and-hipkins-approval-slumps

- [2026-09-19] Substack analyst Elliot Crossan's three-month rolling average (to late Sept) has 
both parties sitting on a three-month average of 30.4% each, while combined support for the Greens, ACT, NZ First, TPM and Opportunity is sitting at 39.2%
, implying the "true" underlying race is closer to dead-even than any single poll, but also that 
Labour is headed for its 4th worst result in nearly 100 years; National is headed for its 3rd worst result since the party was founded in 1936
 — both majors are historically weak, so small campaign shocks could tip the margin sharply in either direction. https://elliotcrossan.substack.com/p/election-2026-welcome-to-the-unpopularity

- [2026-09-15] RNZ's Freshwater Strategy/The Post poll (4–11 Sept) shows a mechanism for a further Labour slide: 
Since the last of this poll in June, Labour has slumped by 7 points. It remains in the lead, but only just
 — a seven-point one-poll drop for Labour is the single largest swing recorded this cycle and, if repeated by other pollsters, would move the margin decisively toward National. https://www.rnz.co.nz/news/politics_election-2026/1382772/possible-hung-parliament-as-labour-and-national-slump-in-latest-poll

- [2023-11-03 precedent, reinforcing downside risk] The 2023 election is the direct precedent for a real-world "low scenario": polls in the final weeks had National and Labour roughly tied or Labour trailing only modestly, yet 
Seats won 48 / 34 / Seat change +15 / −31
 for National vs Labour respectively — a final margin of National +14, showing final-campaign consolidation of the right vote (undecideds breaking right, minor-right-party voters returning to National) can produce a much larger gap than mid-campaign polling implies.

- [2026-09 fetched] Curia's own methodology note states seat projections 
assumed all current parliamentary parties with electorate seats will win at least one electorate seat and be eligible for List MPs
 but explicitly 
no overhang seats are assumed or projected
 for National or Te Pāti Māori — a modelling choice that, if incorrect (i.e., if Te Pāti Māori wins Māori seats beyond its party-vote share, creating overhang), could shrink Labour's practical seat count relative to National even at similar vote shares, since overhang expands the House rather than reallocating National/Labour seats. https://www.taxpayers.org.nz/sept26_pollnztughjkuhn

BOTTOM LINE: A genuine low-scenario path exists and has already been observed mid-cycle (the Sept 1–3 Curia poll's National 39–Labour 35 seat model, a −4 Labour margin) driven by Opportunity falling back below the 5% threshold and right-bloc consolidation; combined with the 2023 precedent of late-campaign swings toward National well beyond mid-campaign polling and Labour's historically weak underlying vote share, a final result with Labour trailing National by several seats (roughly −4 to −10) is a credible bottom-of-range outcome, distinct from the tied-race scenario implied by the newest (Sept 28) 1News-Verian poll.

NOT FOUND: No single pollster or aggregator has published a scenario explicitly modelling "Opportunity fails threshold AND late-campaign right consolidation" combined; no NZ-specific analysis quantifying how much of the 2023 late swing was due to undecided-voter breakage versus differential turnout was located in this search.

## Required pre-forecast walkthrough

Before giving percentiles, address these explicitly in your rationale:
  (0) Resolution check: what the resolution source currently shows, and whether the value is already known or fixed as of today.
  (a) The time left until the question resolves.
  (b) The outcome if NOTHING changes from today (the status quo value).
  (c) The outcome if the CURRENT TREND continues.
  (d) The expectations of experts / markets / base rates.
  (e) A plausible scenario that produces a LOW outcome (near p10).
  (f) A plausible scenario that produces a HIGH outcome (near p90).

## Calibration guidance

- **Be humble about tails.** Good forecasters set WIDE 90/10 intervals to account for unknown unknowns. Narrow tails get punished by the log score far more than slightly-biased medians.
- **Status quo anchoring.** The p50 should be close to the status quo value unless you have strong evidence of a trend.
- Don't pile mass at one value — if you're tempted, widen the spread by 20-50%.
- **Markets and experts are priors, not anchors.** If a SAME QUESTION market, analyst consensus, or community percentiles appear in the evidence, treat them as a strong prior — centre near them and widen — and name the specific evidence they lack whenever you move away. RELATED markets are context only.
- **Open bounds carry real mass.** Where a bound is OPEN, the outcome can land beyond it and the question then resolves "below range" / "above range". Put honest probability there (p_below_min / p_above_max in the output) instead of squeezing the distribution inside the displayed range; the season's worst losses were outcomes that landed where the distribution had ~2%.
- **Relative-return / spread questions ("how much will X's return exceed Y's").** A near-zero median is usually right, but size the TAILS to the more VOLATILE leg, not to a generic 2-3pp spread. Two broad equity indices (e.g. Nasdaq-100 vs S&P 500) do stay within roughly ±2-3pp over a two-week window. But when one leg is a commodity (crude oil, gold) or a single high-beta stock (e.g. Nvidia), the two-week realized spread regularly reaches ±10pp or more — crude-vs-equity spreads have hit ±18pp. In those cases set p10/p90 to at least ±10pp (±15pp or wider for oil) while keeping the median near the status-quo spread; the 2-3pp intuition applies only to index-vs-index.
- **Heavy-tailed one-off magnitudes (opening-weekend box office, product-launch or debut numbers).** Pre-event tracking for these routinely misses by ±40% or more, and results land both far above and far below the consensus. Do NOT treat the displayed [min, max] range as the plausible range or clamp p10/p90 to its edges: the outcome can fall outside the range in either direction (recent opening weekends resolved ~7% above a displayed ceiling in one case and roughly half the displayed floor in another). Set p10/p90 wide enough that a 2x-or-0.5x surprise versus your central estimate still lands inside the interval, and when concrete current tracking disagrees with historical/franchise base rates, weight the tracking.

## Output

Return ONLY valid JSON, no markdown fences:

{
  "rationale": "<address (a) through (f) above, then state your distribution — 6-10 sentences total>",
  "p10": <number>,
  "p25": <number>,
  "p50": <number>,
  "p75": <number>,
  "p90": <number>,
  "p_below_min": <probability in [0, 1] that the outcome is BELOW -15.5>,
  "p_above_max": <probability in [0, 1] that the outcome is ABOVE 15.5>
}

Percentiles MUST be in non-decreasing order (p10 <= p25 <= p50 <= p75 <= p90). p_below_min is P(outcome < -15.5); p_above_max is P(outcome > 15.5). These must agree with your percentiles: a p10 below the lower bound implies p_below_min > 0.10, a p90 above the upper bound implies p_above_max > 0.10. A mass below ~0.05 on an open side needs a specific reason.