Silly buggers: what game is New Zealand First playing in the electorates?
- Grant McLachlan
- 2 hours ago
- 11 min read
New Zealand First is parachuting household names into seats it has little hope of winning outright. Work out why, and you start to see how Winston Peters could decide the next government one electorate at a time.
Few political commentators will stick their necks out far enough to make an audacious prediction. When they do, it is usually because they hold inside information, or because they have the influence to turn a forecast into a self-fulfilling prophecy.
Look closely at New Zealand First’s electorate candidacies for the 7 November election, though, and a third possibility presents itself: that the prediction is being engineered in plain sight, and almost no one is writing it down.
New Zealand First is running marquee candidates in Hawke’s Bay seats it has almost no hope of winning outright.
The lazy reading is vanity — Winston Peters has always liked a household name.
The more interesting reading is that winning the seat was never the point. The point is what a strong third candidate does to everyone else’s arithmetic.
And in a Parliament heading for an overhang, that arithmetic decides who governs.
Contents
The seats that suddenly matter
Under MMP, the party vote decides how many seats a party gets; the electorate vote decides only who fills them.
For most of the MMP era that has made individual electorate races a sideshow to the national party-vote contest. 2026 breaks that rule for one party in particular.
I set out the mechanism in The overhang trap. At around 30 per cent of the party vote but 40-something electorate seats, National’s proportional entitlement is swallowed by its own electorate wins, its list empties out, and Parliament expands seat-for-seat to seat the surplus.
Those seats are allocated by the Sainte-Laguë method under the Electoral Act 1993. The consequence is brutal and counter-intuitive: every electorate National holds above its proportional share makes the right bloc more dependent on its coalition partners, not less.
In that world an electorate is no longer a sideshow. It is the whole game.
In a Parliament heading for an overhang, every electorate is a bargaining chip — and New Zealand First has worked out where the chips are.
The Hawke’s Bay formula
I was once a campaign director for the National Party in Hawke’s Bay, so I know the territory and I know the formula.
National’s revival in the Bay in the 2000s did not rest on one strong candidate. It rested on two, running in tandem.
Chris Tremain — the son of All Black and local legend Kel Tremain — took Napier in 2005. His campaign was run by the strategist Simon Lusk — while former investment banker Craig Foss prised Tukituki off Labour’s Rick Barker the same year.
Two recognisable local men, two neighbouring seats, each reinforcing the other and feeding a regional swing. The pairing held until Tremain stood down in 2014 and Labour’s Stuart Nash took Napier back.
National rebuilt the same formula in 2023 with two fresh household names. In Napier it ran Katie Nimon, whose surname has been on the side of the region’s buses for more than a century. In Tukituki it ran Catherine Wedd, a former TVNZ reporter married into a well-known local family.
Both turned their seats comfortably blue — Nimon by around 8,000 votes, Wedd by close to 9,000. The Bay was National’s again.
Two names, one bay
Now look at what New Zealand First has done.
It has run the National formula straight back at National — with two of the biggest names it could find, in the same two seats.
In Tukituki it has selected Taine Randell, the former All Black captain, born and schooled in Hastings, campaigning on energy reform and mass immigration.
In Napier it has Stuart Nash — three-term former Labour MP for the seat, sacked from Cabinet in 2023, now standing under Peters’ banner and openly tipped by Peters as a future minister.
Even Chris Hipkins acknowledged the threat, joking that Randell would “soon discover there’s only one captain in New Zealand First.”
And look closely at who each man is built to wound.
Nash is not just any defector. The surname is Labour scripture — he is the great-grandson of former Labour Prime Minister Walter Nash, and he held Napier for Labour across three terms. A Nash on the Napier ballot does not frighten National; he frightens Labour, drawing off the provincial, blue-collar Labour voters who sent him to Parliament in the first place.
Randell is the mirror image: an All Black captain running on rural grievance, energy and immigration, aimed squarely at the conservative country vote that is National’s to lose.
New Zealand First has not picked two candidates. It has picked one to bleed each major party.
The 2025 boundary review cuts in opposite directions for the two seats.
Napier has been fortified for National: the Representation Commission shifted its southern boundary down to the Ngaruroro River, handing Nimon a clutch of rural communities from Tukituki — a conservative buffer she publicly welcomed.
Tukituki, losing that same rural north, has been nudged left, stripping conservative votes out of Wedd’s base just as the Wattie’s and McCain closures hammer the local economy.
Of the two, Tukituki is the soft target; Napier is the harder ask.
Neither candidate is likely to win outright. Wedd and Nimon are sitting on majorities of roughly nine and eight thousand, and New Zealand First took just 6 per cent of the Tukituki party vote in 2023.
Asked about Nash, Luxon simply insisted Nimon would hold Napier. He is probably right about the headline result in both seats.
Luxon may be missing the point — because in 2026, what these candidates do to the seats around them matters even when they lose.
The Bradbury manoeuvre
Here is the point. You do not have to win a seat to decide it.
Remember Steven Bradbury, the Australian speed skater who won his country’s first Winter Olympic gold in 2002 by being the only man still upright when the entire field crashed on the final corner?
Here’s a reminder, in pure Ozzy Man Reviews style (contains colourful language:
A strong third candidate can do to an electorate what that pile-up did to the race — but in Hawke’s Bay the crashes happen in different lanes.
In Tukituki the pile-up is on National’s side of the track: if Randell peels 10 to 15 per cent of the conservative vote off Wedd, and a national swing is already eating into a majority the boundary review has just trimmed, a Labour candidate could come up the middle and take a seat National would otherwise hold easily.
In Napier, it will be chaos.
Napier was a safe Labour seat from 1954 to 2005, then swung to National behind a recognisable surname from 2005 to 2014, and again from 2023. That points to a large bloc of swing voters loyal to no party.
Stuart Nash was both a recognisable surname and a Labour candidate from 2014 to 2023. How much of his support was the name, and how much was loyal Labour, is impossible to disentangle.
It is that unknown that should worry both National and Labour.
The tidy reading — one candidate to bleed each party — is the one I started with. Napier may not be that obedient. If Nash’s vote is as personal as his three wins suggest, a joint New Zealand First campaign with Randell could pull as hard on Nimon’s centre-right as on Labour’s base.
Nash is less a scalpel than a wildcard — which is exactly what makes the seat dangerous for both major parties.
Every National seat that falls this way — whether to Labour or to New Zealand First — is one fewer brick in the overhang, and a seat National can no longer count on. The party that gains most from the wreckage is the one that engineered it.
You do not have to win a seat to decide it. You only have to crash into it on the final corner.
Insurance against the threshold
There is a second, colder reading.
Start with where it stands now. New Zealand First holds no electorate seat at all, and it is polling near 11 per cent — every one of its MPs a list MP, every one safe on those numbers.
New Zealand First has been wiped out of Parliament twice in living memory — in 2008 and again in 2020 — each time by falling below the 5 per cent threshold.
It has not won an electorate since Peters took Northland at a 2015 by-election, and Shane Jones came third trying to reclaim it in 2023, as Dan Brunskill noted in The Spinoff.
An electorate win is the insurance policy against another wipe-out.
A party that holds a single electorate stays in Parliament and brings its list MPs with it, whether or not it clears 5 per cent.
Running genuine contenders in seats that are winnable on a good night is exactly what you would do if you wanted a hedge against your own party vote collapsing in the final month.
Napier shows the insurance logic in miniature. Nash does not need to win the seat.
If New Zealand First is returned to Parliament, he comes back on the list whether or not he takes Napier — so the candidacy wounds Labour at no cost to Peters.
New Zealand First’s vote has a habit of doing the opposite — surging late — but Peters did not build the longest career in New Zealand politics by trusting a habit.
The sanctuary on the right
The Hawke’s Bay double act is not the whole of it.
New Zealand First has spent this term turning itself into a sanctuary for the politically displaced:
- Former National minister Alfred Ngaro is now a New Zealand First candidate.
- So is Elliot Ikilei, the former New Conservative leader and Hobson’s Pledge spokesman.
- Former National MP Harete Hipango has crossed over.
- On the West Coast, former Buller mayor Jamie Cleine is carrying the flag in West Coast–Tasman.
Read individually, these are unremarkable candidate announcements. Read together, they describe a party methodically planting recognisable local names in marginal and symbolic seats up and down the country — the Hawke’s Bay manoeuvre, repeated at scale.
This is not one trick. And it is certainly not Peters’ first rodeo.
The horse-trade
Imagine being in a position where Labour and National both have to play nice on the campaign trail with a party that did not — knowing it could be the kingmaker.
But follow the arithmetic to its strangest conclusion.
National’s two most senior figures — Finance Minister Nicola Willis and Chris Bishop, the Housing Minister now also serving as Attorney-General — survive only if the party’s list comes back to life.
Willis has no electorate at all; she is standing list-only.
Bishop holds Hutt South by 1,332 votes on a redrawn map that has bolted a left-leaning suburb onto his seat, and if he loses it, he too falls back on the list.
But the list only exists if National stops winning so many electorates: at around 30 per cent of the party vote, every seat National holds above roughly thirty-six is an overhang seat that keeps its list stuck at zero.
So the people at the very top of National’s list are rescued not when National wins, but when National loses electorates — and losing electorates is precisely what New Zealand First is trying to make happen.
If Randell takes Tukituki and Nash takes Napier — or even if Labour wins those seats — those are two National seats gone, two steps back toward the point where National’s list, and Willis and Bishop with it, returns from the dead.
And here is the asymmetry that makes it leverage rather than mischief. The candidates New Zealand First has sent to do the damage are risking nothing themselves.
Nash and Randell are senior recruits, not back-of-the-list makeweights — and on its current polling the party comfortably returns its top names. They walk into Parliament whether or not they win a seat.
Willis and Bishop have no such cushion. Peters’ people are untouchable; National’s two most senior ministers are not — and their fate now runs through his candidates.
The Finance Minister and the Attorney-General could owe their place in Parliament to the two men Winston Peters sent to beat their own colleagues.
Ministers who owe their survival to Winston Peters are ministers in Winston Peters’ debt — and the two most likely to lead National next would owe him everything.
That is a different and deeper kind of leverage than the usual kingmaker story.
Peters would not merely hold the balance of power between two blocs; he would hold a quiet claim on the loyalty of the very people most likely to replace Luxon.
A Prime Minister whose caucus is restless, whose finance minister and Attorney-General sit in Parliament courtesy of his coalition partner, and who needs Peters to govern at all, is negotiating from weakness on every front at once.
Remember, Bishop and Willis were two of National’s four negotiators (the others being Luxon and Chief of Staff Cam Burrows) that sat down with Peters’ team to form the current government. Look at what Peters gained from that deal.
This is not a prediction that Peters becomes Prime Minister. It is an observation that he would not need to be — because the people who might take the job would already be his.
The blind spot
All of which depends on something nobody is measuring.
Electorate-level polling is scarce and expensive, and the national party-vote polls that dominate coverage tell you almost nothing about who wins a given seat.
In 2023, the only published poll in Tamaki was a Taxpayers’ Union–Curia survey, run through Curia — the firm of National-aligned pollster David Farrar, who also co-founded the Taxpayers’ Union. It showed a statistical dead heat with National marginally ahead. ACT’s Brooke van Velden won the seat anyway.
This is not a one-off. Through 2020, National’s internal Curia numbers repeatedly reached the public showing ACT polling higher than the broadcaster surveys did — on one occasion teased by the Taxpayers’ Union’s own director while a rival poll was still in the field. From 2021 the Herald dropped its own pollster in favour of the Taxpayers’ Union–Curia poll.
The lesson, however, was plain: the seats decided by a few thousand split votes are the seats no one is watching.
The Opportunity Party’s new leader, Qiulae Wong, is running in Mount Albert — a Labour seat held by just 18 votes in 2023.
When the main source of electorate polling is a single partisan-aligned outfit, the choice of what to publish — and when — becomes a weapon in itself.
The Tamaki poll could have been withheld, leaving the public none the wiser. The fact that it was released signalled van Velden was a credible threat, and arguably created the momentum that carried her to victory.
If the Taxpayers’ Union commission a poll for any of the Hawke’s Bay seats — then decide to release it — all it demonstrates is that it fits with their strategic interests. If they don’t commission a poll, then the fog of war that it creates achieves the same outcome.
Three-way contests are forming across Hawke’s Bay. If commentators keep reading the national polls and ignoring the electorates, they will be as surprised on election night as the voters will be furious.
Winston Peters has never needed to win the most seats. He has only ever needed to be the man everyone else cannot do without — and in 2026 he is building that position one electorate at a time, while the people whose job is to keep score are all staring at the wrong number.
I was the only commentator to predict that Winston Peters could form a government with Labour in 2017.
I was also the only commentator to predict the overhang in the 2023 election, which would make Winston Peters the kingmaker.
I am predicting an even larger overhang this time if the current situation persists.
