The numbers in Mt Albert

First some stats:

• National’s party vote in Mt Albert was only 35.7% – 4.4% lower than any seat National did win.
• Clark’s majority of 10,351 is the third largest of seats not held by National – only Epsom and Manukau East are higher
• The combined party vote for the left (L/G/P) was 54.3%, for the centre (NZF/M/U) 4.1% and for the right (N/A) was 39.2%
• The combined party vote for the right is the 9th lowest out of 63 general seats. For the left their party vote is the 9th highest out of 63 general seats.
• Seats with a smaller gap between right and left are Mana, Wigram, Hutt South and New Lynn

Mt Albert is fundamentally a leftwing seat. This does not mean it is impossible for National to win it, but the focus purely on National vs Labour party vote is misleading.

Also at the end of the day, while the party vote is the best primary indicator of support, this is a winner take all electorate vote contest and we will see tactical voting on the right and left. So how might we calculate how thinsg look. Let us start with 2008 party vote:

1. Labour 14,894 or 42.6%
2. National 12,468 or 35.7%
3. Green 3,846 or 11.0%
4. ACT 1,227 or 3.5%

Now let us adjust these for movement since the election, taking the Curia public poll average:

1. Labour has gone from 34.0% to 30.8% – a relative decrease of 9.4%
2. National has gone from 44.9% to 53.1% – increase 18.2%
3. Green has gone from 6.7% to 7.5% – increase of 11.6%
4. ACT has gone from 3.6% to 2.5% – decrease of 31.4%

So applying this to the party vote, we get

1. Labour 13,497
2. National 14,735
3. Green 4,291
4. ACT 841

So does this mean National is in the lead? Well hold off. Because this is an FPP electorate contest and minor party voters vote tactically. How did votes split their votes in 2008? We can’t use Mt Albert as an example, as that was influenced by the Clark factor. So let us use the overall vote splitting for general seats. This tells us:

1. Green party voters voted 33.9% Green candidate and 47.9% Labour candidate and 10.9% National candidate
2. ACT party voters voted 16.4% ACT candidate, 4.8% Labour candidate and 73.1% National.

What does this then give us:

1. Labour 15,593
2. National 15,818

So on these calculations, it is within 250 votes. But then you look at more qualative factors:

1. Shearer is the only non List MP who can say I will not be in Parliament if you do not vote for me. I think this is easily worth 2,000 votes.
2. Despite high honeymoon polls, there is little reason for people to give the Government an additional seat – no Government has won a by-election for at least 70 years. It is hard to excite pro-Govt supporters to get out and vote as they no this won’t change anything – worth around 1,000 votes
3. Greens and ACT will be mounting vigorous campaigns. As the Greens have more voters to appeal to, I would say this could knock 1,000 votes off Labour and 500 votes off National, so a net loss to Labour of 500
4. Labour’s on the ground organisation in Mt Albert should be worth at least 500 votes
5. Shearer will appeal to some centre-right voters with his backgroudn and views – again maybe 500 votes.

So add all these factors up and I’d say they give Labour an extra 3,500 votes or so, which if nothing else happens would have them win the seat by around 3,250 votes.

Now it is never this simple. Events will occur that change this. But at this stage what I think it shows is that yes it is possible National could win the seat with a very narrow 250 vote majority, but more likely Labour holds on with 3,000 or so votes.

What will be the events that can change this? Well polls and tactical voting. If a poll shows National and Labour neck and neck, and Greens well back, then the Green vote may collapse to Shearer. But likewise if a poll shows the Greens at over 15%, maybe even 20% – then they could make a strong play for Labour voters to vote Greens to give Labour a guaranteed future coalition partner (Labour can probably never govern again without the Greens).

Let the campaigns begin!