Late on Monday, the AEC transitioned the election results website to its final archive form. You can now find the results at results.aec.gov.au, and the previous link no longer works.
There is probably enough fresh data to fuel a whole week of blog posts. I am planning to return to the question of the national 2PP and 2PP preference flows by party, as well as some deeper analysis of the 3CP trends across the country.
For this post, I am going to focus on close races, in particular those involving independents against the major parties.
Let’s start by going back to the seat of Bradfield. Last week I calculated an estimate for the Labor 2PP here, but I also flagged that I would revisit the figure once we had preference flows for similar seats. Those have now come in, and it turns out that the rate of preferences flowing from independents to Labor (from the 3CP to the 2PP) has increased in similar urban “teal” seats.
The proportion of independent 3CP votes flowing to Labor on the 2PP went up substantially in every urban teal seat, by an average of 5.3%. Mackellar stands out as having a weaker preference flow, but the average is 5.3%.
If I apply that 5.3% increased preference flow to Bradfield, that produces a Labor 2PP of 51.1%, up from the 49.4% previously estimated. It’s a bit higher than Kevin Bonham’s estimate of 50.6%. Either way, that would mean that Labor would win the 2PP in Bradfield, giving them 101 out of 150 seats.
This would mean that the 2PP swing in Bradfield would be 7.25%, which would have been Labor’s sixteenth-best in the country.
It also means our estimate of the Labor 2PP nationally ticks up to 55.26%, a swing of 3.13%.
I will use another blog post to look at the number of seats with particular 3CP combinations, but there were 37 seats where an independent made the 3CP. This is an increase in raw numbers, but there was also an increase in the size of the 3CP vote those independents achieved – they are getting bigger shares of the vote than in the past.
In 35 of these seats, it was a 3CP between Labor, the Coalition and an independent.
Calare and Indi are in a category of their own, seemingly unique for at least 20 years. In Calare, two different independents made the 3CP. In Indi, the 3CP included independent Helen Haines, the Liberal Party and One Nation.
In my 3CP dataset dating back to 2004, there have only been five cases where a major party missed out on the 3CP, but in the other three cases there was a Liberal and National candidate. These seats are the only cases where two different independents or minor parties made the 3CP.
In 21 of those 37 seats, the independent made the 2CP – 14 vs the Coalition, 7 vs Labor. In the other 16, Labor was stronger in six and the Coalition was stronger in ten. So despite Labor dominating the federal election, a majority of independent 3CP contests had the Coalition as the main rival to the independent.
This next chart shows what proportion of preferences from the lower-polling major party flowed to the independent (or to the other major party) in the 19 seats where the independent made the 2CP (excluding Calare and Indi).
The independents always dominated preference flows in seats where they ran against the Coalition. There was a gradual increase in preferences as you move towards the strongest independent seats, usually those in the cities, but there are some surprises. The flow of Labor preferences to Nicolette Boele was relatively low in Bradfield, while Flinders ranked higher than two of the incumbent teals.
The Independent vs Labor seats were more dispersed. The Muslim independent in Watson barely gained a majority of Liberal preferences, while Liberal preferences slightly favoured Labor over Peter George, the independent in Franklin. Rates in the other five seats look similar to preference flows to independents in IND vs LIB seats.
So this brings us to the races where the 3CP was particularly close.
The gap between second and third seems to be about the same as 2022, on average. It looked closer when calculated on primary votes, but it’s about even using 3CP data.
But the number of seats with a very close 3CP margin is higher than ever before, with 19 seats reporting a gap of less than 5%.
Of those 19 close races, seven of them had a Greens candidate in the top 3, and eight were amongst the 35 Labor-Coalition-independent races.
For those where an independent narrowly missed out on the 2CP, I have applied the preference flow for a similar seat to come up with an estimate of the 2CP.
For the first five seats on this list, they are vaguely “teal” independents who challenged the Coalition. I applied a preference flow from Labor to independent of 75.8%. For Blaxland, I applied the preference flow seen in Watson.
The most interesting element here is that I am projecting that Forrest and Grey would have very narrowly had the independent in front if they had made the 2CP. In Forrest, the independent missed out on the top two by just 0.8%.
I should express some caution here – if the preference flow was more like Wannon or Cowper (other regional seats) they would have fallen short. But it is possible those independents may have won.
In Blaxland, the independent fell just 22 votes short of the 2CP. Ouf actually polled higher on the primary vote, 3CP and my estimated 2CP than the other independent in Watson, but a higher Liberal vote stopped him from making the final count.
So finally, I have taken my estimates of 2CP in seats where independents made the 3CP but not the 2CP, as well as my Bradfield estimate, to produce these charts comparing 2PP to 3CP for independents, split between those competing against the Coalition and those competing against Labor. I have not included all of the Labor vs Independent races because they are all so different to each other – I don’t think Paterson, Moore or McMahon necessarily fit the mould of other Labor vs Independent seats.
Labor’s 2PP trailed the independent 2CP in most Coalition vs independent seats, but not exclusively. It was higher in Berowra and Casey, and in the case of Solomon (one where an independent was a distant third) the independent would have theoretically lost if they’d overtaken Labor.
As for the Labor vs Independent seats, the independent always does better than the Coalition, but by how much varies enormously. The Muslim independents in Blaxland and Watson don’t do much better, while Andrew Wilkie in Clark does much much better.
That’s it for today, but I’ll be working on another post tomorrow afternoon.
