SA2026 – Does the SA Ticket Vote Savings Provision Matter

One Nation was accused of using South Australia’s ticket vote savings provision to harvest preferences at the state election. In fact One Nation’s tickets were entirely in line with the party’s promise of no deals on preferences. But it appears that registered how-to-votes should be clearer in saying number all squares.

As I explained earlier this week, South Australia has a unique provision in its Electoral Act that applies registered preference tickets to ‘save’ votes that would otherwise be informal.

Misunderstanding of the provision generated commentary during the election campaign about parties controlling preferences through the tickets. In particular there was criticism of One Nation’s how-to-votes that did not recommend preferences. The how-to-votes recommended voting ‘1’ for the One Nation candidate before saying to number all the other squares.

Was this just a way of encouraging One Nation voters to vote ‘1’ and then grab control of the vote’s preferences through the application of the ticket vote?

I’ll get on to the mechanics of the One Nation how-to-vote in 2026 inside the post, but first want to look at how the provision has worked in the past .

The purpose of the registered tickets is not to grab preferences. It is to ‘save’ votes votes that would otherwise be informal, to count votes that would be informal at a Federal election.

Some confusion is understandable as any mention of ‘ticket voting’ invokes memory of the now abolished group voting ticket system that distorted Senate elections until abolished in 2016.

The two systems are not comparable. Senate ticket voting turned preferences into a tradeable commodity allowing the engineering of election results with parties electing Senators despite tiny first preference vote shares.

That was possible because Senate ticket votes represented around 95% of all votes. In stark contrast, the South Australian ticket system applies only to a small percentage of votes, votes that would otherwise have been informal. The percentage of formal votes that have had tickets applied has been, wait for it, between 3% and 6% over the past 40 years.

And of those ballot papers that were saved, more than two-thirds were never counted out as preferences. They were saved by the ticket vote provision and only their first preferences were counted.

(For more information on the tickets, read this backgrounder produced by the Electoral Commission of South Australia.

Why Does the Ticket Provision Exists

The 1984 Federal election was the first to use the divided Senate ballot paper still used today. The ballot paper instructions were to mark a single square for a party ticket ‘above the line’ or to number all the squares below. There were no changes to voting for the House of Representatives.

The new Senate system saw informal voting plummet from 9.9% in 1983 to 4.7% in 1984. But there was a tripling of informal voting for the House, from 2.1% in 1983 to 6.3%. Research identified numbering errors as the cause, in particular a big rise in voters marking only a single ‘1’ first preference.

In 1985, South Australia debated and introduced a new Electoral Act. It included adopting the new Senate ballot paper for Legislative Council elections. To avoid duplicating the increase in lower house informal voting, the new Act included a savings provision that would allow a vote with insufficient preferences to be saved by the application of a registered ticket of preferences.

Parties were required to register how-to-vote recommendations, and these were to be displayed in a standard format in the partitions where voters cast their votes. Several states have followed South Australia in requiring the registration of campaign material, and in particular how-to-vote recommendations. No state has followed South Australia in displaying the how-to-votes.

Nor has any state followed the South Australian idea of registering a preference ticket as a savings provision. The ticket did not have to be the same as the how to vote. In the case of candidates that did not make preference recommendations on their how-to-vote, they were required to register two tickets to spit preferences in line with having made no recommendation.

The full mechanism of how the tickets are applied is set out in the backgrounder linked above.

How Many Votes are Saved by the Ticket Provision

The table below shows the total number of saved votes at elections since 1985. The table is taken from the backgrounder linked above. Ticket vote data is not available for the 2022 election.

As the table shows, the number of votes saved is not large. Also, because most votes saved are for candidates who finish in the final pairing after preferences, most ballot papers are effectively saved for only their first preference. Further preferences were not required.

Let me pull apart the number of votes saved using the ticket vote provision in 2018.

• There were a total of 40,067 otherwise informal votes saved at the 2018 election. These represented 3.7% of the total vote, or once admitted to the count, 3.8% of the formal vote.
• Of the saved votes, 28,242 were for one of the final two candidates in each district. These votes counted only for their first preference. The application of the ticket merely saved to vote from being informal and never needed to have preferences applied. These votes represented 2.7% of the formal vote or 70.5% of the total saved votes.
• Another 2,440 votes were saved in seats where the winning candidate polled above 50% of the formal vote. While ticket preferences were applied in these contests, they played no part in determining who won. This represented 0.2% of the formal vote or 6.1% of the saved votes.
• In total there were 9,385 saved votes where ticket preferences were applied and counted. No seats were close enough for the saved votes to impact the result. These votes represented 0.9% of the formal vote or 23.4% of the saved votes.
• There were 8 districts where more than 1,000 votes were saved, the highest being 1,340 in Kaurna representing 5.9% of the formal vote. Of these Kaurna saved votes, 1,176 or 88% were cast for the two candidates who finished in the final pairing after preferences.

The question is whether any particular party is helped by the provision. As 2018 data below shows, Nick Xenophon’s SA-Best received greater benefit from the operation of the savings provision. The reason why is relevant to the fuss over One nation in 2026.

Ticket Votes by Party – 2018 South Australian Election

Party	% of Preferences Completed Votes	% of Ticket Votes	% of Total Votes
Labor Party	38.1	33.9	38.0
Liberal Party	32.9	30.4	32.8
Nick Xenophon SA-Best	13.8	22.6	14.1
The Greens	6.7	6.3	6.7
Others	8.5	6.8	8.4

Nick Xenophon’s How-to-Votes

The image below shows the major party and SA-Best how-to-votes displayed on voting screen in the electorate of Hartley in 2018. Hartley was the seat that Nick Xenophon chose to contest after resigning from the Senate.

In most seats the Labor and Liberal parties buried the SA-Best candidate towards the end of the preference ticket and varied whether they received preferences before the other major party. The how-to-votes were designed to offer little assistance to SA-Best candidates.

As shown above, the high profile of Xenophon in Hartley saw the two major parties adopt a different tactic. Both published split tickets that divided preferences between the other major party and Xenophon. The structure of these split tickets offered little help for Xenophon who finished third anyway.

But look at Xenophon’s how-to-vote. It says vote ‘1’ Xenophon with text in a red box to say number all the squares. This was the structure of the Xenophon how-to-votes in all districts. I think there is a good argument that the SA-Best how-to-vote induced 1-only votes and is why there were more SA-Best votes saved by the ticket votes.

In conjunction with his open ticket, Xenophon’s registered tickets with a 50:50 split between the major parties in all seats. This is exactly what One Nation did in 2026.

In Hartley there were 1,061 votes saved by the ticket votes, 4.9% of the formal vote. Of these, 195 were Labor first preferences, 374 Liberal and 389 Xenophon. With the major parties finishing top two, their saved votes only counted as first preferences. Xenophon’s votes were implied to have preferences and were roughly evenly split between Labor and Liberal.

Labor’s vote in Hartley was 24.0% of all votes compared to 18.4% of the saved votes. The same comparison figures for other parties were Liberal 40.4% versus 35.2%, and for Xenophon 24.9% versus 36.7%.

If the high save vote percentage for Xenophon were just the case in Hartley, you could argue personal vote was the reason. But as the difference is there for SA-Best across the state, it is a wider issue. Does it just reflect a vote for SA-Best, or is it to do with the displayed 1-only how-to-vote?

I strongly suspect the how-to-vote is the cause.

But why was there no fuss about SA-Best’s how-to-votes in 2018 where it has been so controversial with One Nation in 2026?

Why the Controversy with One Nation in 2026?

Before the election One Nation made it clear the party had done no preference “deals”. They made no recommendation of preferences to the Liberal Party despite the Liberals listing One Nation ahead of Labor across the state.

In line with its statement on preferences, One Nation lodged 1-only how-to-votes for display, along with registered tickets with preferences split between the major parties.

The 2018 data suggests that 1-only how-to-votes increases the number of 1-only completed votes that need to be saved by registered tickets. Had One Nation lodged tickets sending preferences to the Liberal Party, that would have been controversial. It would have encouraged 1-only votes and sent preferences to the Liberal Party.

But One Nation didn’t do that. They lodged split tickets in-line with the party’s how-to-vote recommendations. In seats like Finniss and Mount Gambier the split tickets favoured the Independent, but everywhere there was an equal split between Labor and Liberal.

Compare that to Independent Geoff Brock in Stuart who had an open how-to-vote but his registered ticket sent the preferences of any saved vote to Labor. I should stress that Brock’s preferences were never distributed so only the first preference of saved votes counted.

If One Nation had lodged a ticket that directed preferences to Liberal then an outcry would have been justified. But One Nation didn’t. They lodged a how-to-vote in line with what they said before the election, no deals on preferences.

If there is a criticism of One Nation it would probably be over the text about further preferences on the displayed how-to-vote. The examples below are from the seat of Flinders and show the how-to-votes for One Nation and two Independents.

In my view, the text on the One Nation how-to-vote encouraging further preferences was harder to read than the Independent examples. It was also less clear than the Nick Xenophon example from 2018.

While the savings provision exists, it is not meant as an option for voters. The clear intent of the Electoral Act is for full preferential voting. The tickets are a savings provision, there to avoid a vote with clear intent being declared informal.

The evidence from 2018 is that 1-only how-to-votes induce more voters to number only one square.

In my opinion, the small typeface on the One Nation how-to-vote may have induced more 1-only votes.

But One Nation finished in the top two in 31 of the 47 districts so most saved votes didn’t need preferences. Even where preferences were needed, One Nation’s split ticket was entirely in-line with the party’s how-to-vote and wasn’t some “preference harvesting” operation. It didn’t favour either major party.

What One Nation did with its how-to-votes and registered preference tickets is no different to Nick Xenophon’s SA-Best in 2018. Both how-to-vote and ticket corresponded to statements that there were no preference deals.

But, both Labor and Liberal people I spoke to in Adelaide described the registered tickets as the final proof that One Nation were not prepared to lift a finger to help the Liberal Party. Refusing to recommend preferences to the Liberal Party at the election was decision that proved One Nation were setting out to destroy the Liberal Party’s chances at the election.

The registered tickets were just the final proof of the strategy.

But they were not a matter of misleading voters or stealing preferences. The tickets exactly matched what One Nation said, no deals on preferences.