We are now coming towards the end of the Victorian council elections. For the councils I have been covering, final primary votes have been recorded for all wards, and in the vast majority the winners have been determined and preferences distributed.
For this post I am looking at 310 contests over 32 urban councils that use single-member wards. I am excluding the City of Melbourne, which uses quite a different electoral system, but otherwise cover all councils with a population over 85,000.
For this post I wanted to summarise a bunch of interesting data points for these races. Then I’ll come back with another post going through each council’s results, probably out on Tuesday. I will update these stats when the final winners are known.
At the time of writing, using data as of the end of Sunday, the race has been determined in all but 39 seats. Those 39 wards include three wards of Ballarat, six in Frankston, six in Manningham, five in Maribyrnong, three in Merri-bek, one in Moonee Valley, two in Mornington Peninsula, three in Port Phillip, three in Whitehorse and seven in Yarra. I assume the preferences and the final outcomes in these wards will come through soon – they may have been decided by the time you read this post.
The winner won a majority of the primary vote in 37 contests. In another 41 contests, there were only two candidates so there was no need for a preference distribution. There is also 26 contests where a candidate achieved a majority of the primary vote during the distribution of preferences but before reaching the final two, and the distribution then ended. Another 9 wards were uncontested, so had no voting. This leaves 158, a little over half, that had a full two-candidate-preferred count.
If you include the 2-candidate races, there are 199 contests where the vote can be distributed between just two candidates. In just under half of them, the final margin of victory was under 5% (85/199). A further 68 had margins of 5-10%, leaving less than a quarter with a greater margin. The Terrara Ward of Whitehorse had the biggest margin of 29.95%. But it’s worth bearing in mind that the other 63 contests without a 2CP would have likely been safer.
Of the 298 wards decided so far, just over 90% of seats went to the candidate leading on the primary votes. There were 27 come-from-behind wins.
There were a lot of concerns about turnout, but in the end turnout, at least in these councils, was almost identical to 2020. The turnout was 71.5% in 2016, 81.8% in 2020 and 81.9% in 2024. You could do more analysis into where it might have gone up or down, and where the change happened from 2016 to 2020. I did see a chart showing a very strong relationship between turnout rates and the proportion of renters, although there are other census demographics that would have strong correlations (such as density and closeness to the CBD) that could also explain that trend.
The informal rate has continued to drop as these councils have been forced to single-member wards. This means smaller ballots, and under the compulsory preferential system used in Victoria, lower informal rates.
There is a clear relationship between the informal rate and the number of candidates on the ballot. Interestingly, it looks like 2-candidate races had higher informal rates than those with more candidates, but once the number of candidates gets above six the informal rate climbs steeply.
This map shows the informal rate, turnout and share of the primary vote for the winners, by council area.
The turnout rate tends to be higher to the south-east of the Yarra. The informal rate generally was higher in the north-west, where ballot papers were longer, but Bayside had the highest rate. Bayside had a number of wards with enormous ballot papers.
The final layer shows what share of the primary vote was cast for the winner. For this analysis I’ve included the primary vote share for the candidate in the lead where preferences are yet to be distributed, so those numbers may go down slightly.
Generally the smaller ballots of the south-east have led to candidates winning with a higher vote share, with Bayside standing out. Casey also stands out – that council had no incumbents and this seems to have led to a more fractured vote and councillors elected on fewer votes. Casey did have long ballot papers, also likely caused by the lack of incumbents, but had smaller ballots than Wyndham or Bayside, yet the primary vote for winners was lower. I suspect this reflects that a larger share of Casey’s candidates were serious contenders, not paper candidates supporting others.
Finally I wanted to look at how incumbents have done.
In these 310 contests, 209 incumbent councillors were running. 132 of those incumbents were re-elected, or are currently leading. That’s a rate of 63.2%. In contrast, just 14% of non-incumbent candidates won. So incumbency is obviously an advantage, or at least an indicator of seriousness, but it’s not overwhelming. 77 councillors were defeated for re-election.
One factor that is important here is that the electoral system has changed in a majority of these councils. In 20 wards, two incumbents both contested the ward. In three of those wards, both lost, but in the other 17 an incumbent was still elected. So just 60 councillors lost to non-incumbents.
I did split up the statistics based on whether the council was already using single-member wards, but it didn’t make much of a difference. 64.2% of incumbents were re-elected in councils which switched to single-member, and 61.6% were re-elected in those which already had single-member. Incumbents were much more likely to run where the electoral system didn’t change – there were 123 councillors running for 196 wards in changed councils (62.3%) and 86 running for 114 wards in unchanged councils (75.4%).
Incumbents generally got a higher vote than non-incumbents, even those who were elected. Re-elected councillors polled 47.4% of the vote on average – if you include losers, their average vote was 39.3%. For non-incumbent candidates, winners averaged 36.8%. I suspect that incumbents generally won their wards by wider margins, and fewer of the close races involved incumbents. Indeed all nine uncontested wards were won by an incumbent.
This map shows how an incumbent did in each ward. There’s a sea of blue (re-elected) and purple (no incumbent) with a smattering of red (defeated).
I’ll be back tomorrow with a summary of how the parties did and an assessment of the results in each council, hopefully with more complete results.