Four election night scenarios - featuring Christopher Luxon’s worst nightmare

Analysis: In election after MMP election, the formula has been simple: the party vote decides the shape of Parliament, and who can form a government. You could generally look at the party vote share on election night, divide it up between 120-odd seats and get a good sense for who won.But it's the '120-odd seats' bit of that equation that is likely to be different this time. The way the votes fall could mean as many as 130 seats - or even more - to share between parties.How many more than 120-odd seats we'll get, and how many National and Te Pāti Māori get of them are two key things to watch on election night, alongside the party vote, which is still decisive. And there's one more thing - if it's really close, the way special votes fall could shift the final result.An overhang for National - or notIn the current Parliament, National has 49 seats, 44 of them electorates and five from the list. Its party vote is almost certain to fall, most likely to somewhere between 25 and 32 per cent. It is also likely to lose some of its electorate seats, most likely ending up with somewhere between 32 and 40 of these.If National does very poorly on party vote (closer to 25 per cent) and very well in electorate seats (closer to 40 electorate seats), there will be an overhang for the party. It will have won more electorate seats than its party vote alone would entitle it to. The larger any overhang for National, the better its chances of being able to get the coalition band back together again for a return to power. (This is the thrust of the Harry Mowbray-led vote splitting campaign, which National rejects as a strategy.)National doesn't need an overhang in order to lead the next government. If it can get a high enough party vote, it won't matter how many electorates it wins. This is its response to the vote splitting idea - just vote for National if you want a National-led government.The trouble is, the polls tend to suggest it may not have enough support to safely claim victory on Saturday night. There would need to be a polling error. This has happened before and it could happen again. There are too many other contingencies to give a decisive threshold at which National could lead the coalition back to power without an overhang, but it probably needs at least 30 percent of the party vote to feel confident.The strength of support for the junior coalition partners could be its saviour, however. Even if National falls in to the 20s and doesn't win enough seats for an overhang, ACT and NZ First could make up the numbers with a strong showing between them. A combined 18 percentage points for ACT and NZ First plus National at 28 could conceivably just be enough for a one seat majority, even with Opportunity in Parliament.National's electorate wins tallyIf National is clearly in the 20s and looks like it will need an overhang to return the coalition, the number of electorate seat wins will become more of a focus than in any previous MMP election.It's hard to know if National will shed many of its 43 electorate seats - there are even some seats it might pick up, like Epsom or Mount Albert. There are different dynamics at play in several of these seats which make individual predictions only a little better than guesswork. But given National currently holds a number of traditionally marginal or swing seats, it is likely to lose at least some races.The four most likely to fall are probably: West Coast Tasman, Banks Peninsula, Wairarapa, and Mount Roskill.Another five on the watchlist: Hutt South, Northland, East Cape, Kāpiti and Maungakiekie.There could be others in play too. If Labour flips all nine of those seats and you throw in a Hamilton East or Whanganui as surprise loses, National would still have more than 30 electorate seats. At say 32 electorates, it would need a dramatic collapse of the party vote to the low 20s to achieve an overhang for National.If you split the difference and say National sheds a net five electorate seats, to 38, then an overhang is likely if National ends the night at something like 25 to 30 per cent of the party vote.A Te Pāti Māori overhangThe Māori seats, where Te Pāti Māori is entirely focused, are also hard to predict. Polling error rates are typically high and most analysis relies on assumptions about incumbency.It's unlikely, but Te Pāti Māori could theoretically be wiped out and get no seats at all, which would obviously mean no overhang for the party.More likely, it will hold at least two seats, quite possibly three or four. It has explicitly asked its supporters to only back its electorate candidates and give their party vote to any other party that would change the government. That makes it close to certain that any more than 1 seat for Te Pāti Māori will create an overhang. If it wins four, Parliament will grow to 123 (or even 124 if its party vote is microscopic) seats and a majority to govern is at least 62.The electorates to watch for the Te Pāti Māori tally are: Waiariki (Rawiri Waititi), Te Tai Hauāuru (Debbie Ngarewa-Packer), Hauraki-Waikato (Hana Maipi-Clarke), and Tāmaki Makaura (Oriini Kaipara).If it holds all four of these seats, and National's overhang is five, that's a 128-seat Parliament and 65 seats are needed for a majority.There's only a few combinations of parties that can realistically be pieced together in order to get that majority. The biggest variable in that calculation is the newcomer: Opportunity.If Opportunity makes it, or notIn poll after poll, Opportunity comes out over the 5 per cent threshold and in a position to decide which of the left or right bloc forms a government.But even moreso than National, its sensitivity to a polling error is huge. It does not have an electorate seat it is likely to win, and in many polls it is at 5 or 6 percent. It would not be a big surprise for it to end up below 5 percent on election night, relying on special votes to squeak over the line and make it into Parliament, or worse. Equally, it could conceivably hit 7 percent and romp in.If it is clearly over 5 percent and headed for Parliament, it's likely but not certain that its votes will be needed to form the government. The size of the overhang - and how much of it National holds - is the first place to look for a sense of how likely.One more potentially important data point about Opportunity: its leader Qiulae Wong has said she will talk first to whichever party wins the most seats, which could differ from which has the most votes. If Opportunity does hold the balance of power, Wong's position could favour National as having first crack at forming a government. The major caveat, of course, is that National Party leader Christopher Luxon would have to do an enormous U-turn and bin his countless pledges not to work with the party.This is probably the absolute worst case scenario for Luxon - even worse, in a way, than if Labour has the most seats and it's a clean result for the left bloc plus Opportunity. The idea that Luxon would stick to his position of not working with Opportunity and send National to the opposition benches even when it needs only a single seat more to govern - compared with Labour needing eight more - is difficult to conceive of, to put it gently. He would be condemning not only his own caucus, but also an enlarged ACT and NZ First to opposition.How much special votes might matterHistorically, the final make up of Parliament has changed between election night and the final count, which includes special votes.There are four things to watch out for to know how much they're likely to matter this time.First, how close it is. If the coalition has a three-seat buffer on election night, it will probably have cause to believe it can't be caught by any change to the seat allocation from special votes. It might lose some vote, but it may gain some too. That might be complicated, however, by two other things: the shape and size of any overhang and how close to the 5 percent threshold Opportunity is.Special votes could shift a seat that increases or decreases the overhang, which could have an impact on coalition calculations by not only changing the seat tally but also the majority needed to govern.If Opportunity is close to 5 percent on either side, it will have to wait for special votes to be counted before it can be sure of making it. If it doesn't, there will be more 'wasted' vote in the final calculation of seats. If it does, it has a strong chance of being part of the government.Exactly how many special votes there are to be counted is also something to consider if its close. A lot of specials means more runway to shift the seat counts, fewer specials means less room to move.