Political Roundup: The interesting ways New Zealanders voted in 2017

The final New Zealand general election results for 2017 are in. And, although the media and public focus has swiftly turned to coalition building and negotiations, it's worth looking at some of the more interesting results from election night and the final vote count. Here are 15 of the most notable voting statistics to ponder.

1) The growing urban-rural divide in the election turned out to be a myth

According to journalist Charlie Mitchell, "Election data shows not only is the rural/urban divide not growing, it is shrinking", as "Not only did National fail to increase its influence in regional New Zealand, the opposite happened: Labour made significant gains in rural areas, while National was saved by the country's most urban voters" - see: Election shows rural-urban divide shrinking, not growing. For an alternative view, see National campaigner Jenna Raeburn's Inside the campaigns: how National took the migrant and rural vote.

2) National's party vote increased in 14 electorates

Anna Bracewell-Worrall believes National's increasing vote in these electorates is unlikely to be due to issues such as Labour's controversial water tax plans, because "All but one of them are in Auckland. Just one rural electorate saw an increase in National Party vote. That's Hunua, which is mostly rural on the outskirts of Auckland" - see: Digging through the data: National's urban success story. She reports that "Most of National's biggest losses came from the South Island".

3) National won most of the safe seats

David Farrar looks at the MPs with the safest and most marginal electorates in Biggest and smallest majorities.

Farrar points out that National has eight of the top ten safest seats in the country - led by Amy Adams in Selwyn, with the biggest majority of 19,639. At the other end of the scale, the most marginal seat is Adrian Rurawhe's Te Tai Hauauru, which he holds with a 1,039 majority. Just behind Rurawhe is Greg O'Connor, with a 1,051 majority in Ohariu.

4) National is more of an electorate party and Labour is more of a list party

National won 41 electorates and got 15 list MPs, whereas Labour won 17 electorates and got 29 list MPs. For more, see Colin James' This is the world of MMP - get used to it. James also reports that Labour's 2017 vote was "its highest score since 41.1% in 2005".

5) New Zealand First's support dropped in the Maori seats and Auckland, but picked up in some provinces

According to Claire Trevett, "the seats in which NZ First performed best shows that while it still did better than its 7.2 per cent overall result in the Maori seats, its support dropped from 12-14 per cent in those electorates in 2014 to 7-9 per cent this election" - see: NZ First cops it in Maori seats, held up by regions. In Auckland, the party only received "around 3 to 6 per cent" of the party vote. But in Northland and Whangarei, the party vote was 14 per cent.

6) There were a record number of special votes

The number of special votes cast this year was a record 446,287, up from 328,029 in 2014. National lost particularly badly from those special votes: "National always loses out in special votes, but 2017 was the biggest swing yet. The percentage point drop in its party vote share between the provisional and final vote counts was nearly three times what it was in 2008" - see Michael Wright's Election analysis: Why did the special votes swing so far Left?.

7) Voter turnout was highest in Wellington Central, and lowest in Tamaki Makaurau

According to Anna Bracewell-Worrall, "The electorate with the highest turnout was the one on the Beehive's doorstep - Wellington Central - with 86.6 percent of enrolled voters. The seat was won by Labour's Grant Robertson. Of the 50,234 people enrolled to vote in Wellington Central, 43,166 people cast a vote" - see: NZ Election 2017: Five interesting facts about the vote. In Tamaki Makaurau 35,534 people were enrolled, but only 20,593 voted.

8) The distribution of special votes differed more than usual from regular votes

Caleb Morgan's number crunching shows that this year "Labour did their best ever on special votes: 18.91% better than on preliminary votes. Their previous best was last election, where they did 14.16% better" - see: Special votes are increasingly turning towards Labour, away from National. Similarly, "National did their worst ever on special votes: 21.12% worse than on preliminary votes. Their previous worst was last election, where they did 16.86% worse.

9) Although the National Party vote fell only slightly, the centre-right vote fell much more

According to economist and blogger Michael Reddell, "The centre-right parties did impressively well to increase their total vote share in 2011 and again in 2014. But the fall-off in this election - 6.6 percentage points - is pretty stark" - see: Fourth term government votes. Reddell includes a chart showing the decline of the centre-right. See also Reddell's Fossicking in election statistics.

10) Not all voting booths produced the expected results

The voting booth at St Heliers School in the "beachside east Auckland suburb adjacent to affluent Remuera and Mission Bay" claimed "the title for being the New Zealand People's Party's most successful voting booth, with 11 percent of the party vote in that booth" - see Matt Burrows' Digging through the data: The election's weirdest voting-booth stats. Similar examples are provided for the relative success of the Internet Party, the 1080 Party, and Act.

11) The Labour Party performed very well in and Maori strongly Pacifica electorates

Labour's best party vote was in the South Auckland electorate of Mangere, with 60 per cent Pacifica population - see Anna Bracewell-Worrall's What the data tells us: Maori and Pacific voters throw support behind Labour. Apparently, "Labour's 10 most popular electorates were the seven Māori seats and the three electorates with the highest Pacific populations in the country." And the "biggest swing to Labour was in Te Tai Tokerau... where Labour gained 22.4 points, going from 35.1 percent of the vote in 2014 to 57.4 percent this time around."

12) If only university students had voted, parliament might be made up of 39 per cent Labour MPs, 31 percent National MPs, and 19 per cent Green MPs

This is the way that people voted in the university-based booths, according to Critic magazine - see: National Comfortably Win Party Vote at Two Uni Campuses. Of course, different universities had very different results, with National "winning" at Lincoln University, and the Greens "winning Massey University's Wellington campus and Victoria University".

13) Labour possibly lost votes due to its immigration policies

Looking at where Labour performed well and poorly, Branko Marcetic comes to a number of interesting conclusions, including that "Labour tended to lose the party vote in high-immigrant electorates" and "Labour made most of its gains where there were less immigrants" - see: How did Labour's immigration stance impact its immigrant vote? See also, Jenée Tibshraeny's The extent to which the migrant vote propped up National.

14) Labour did poorly from people splitting their votes

The extent to which people split their party and electorate votes is examined by Andy Fyers in his article, The candidates who outshone their party.

Apparently, "Labour candidates got 1.7 per cent more of the candidate vote than the party vote on average, in the general electorates. National candidates do about 1 per cent worse". And Labour's Stuart Nash is highlighted as a candidate who "won 53.4 per cent of the candidate vote compared to 37 per cent of the party vote that went to Labour." And for the latest on where Labour and National won electorates, but lost the party vote, see Claire Trevett's Election Eye: Hiccups in Nelson but party vote shows Labour reclaiming its traditional turf.

15) There were 30 electorate candidates who got less than 100 votes

Some of the less successful candidates are highlighted by Katie Kenny in her article, Let's pause to consider the losers of this election. Kenny reports "Bob Wessex in Wellington Central was the country's lowest-polling candidate, with just 14 votes." And the article explains how her Not a Party (NAP) caused one voter to regret their support - writing to complain: "So disappointed in myself after being mislead to vote for you in Wellington Central. I didn't have time to research any policies of any party so I just went by the name... To find out that NAP [Not a Party] had nothing to do with taking a nap whenever you wanted means I have wasted my vote. I thought I had found my home, my niche in the political landscape."

Finally, for an effective way to see some of the most interesting voting statistics, you can look at this interactive data-visualisation vote map. Its statistician-creator, Chris McDowall explains how it works in his blog post, Interactive: mapping every booth's votes from the 2017 general election.