While there’s a lot of similarities in how Australians vote between the House and the Senate, they’ve never voted exactly the same. For a start, there are more options on the Senate ballot paper than the House ballot paper. Small parties will not run in all House seats (sometimes they run in very few) and thus can only attract Senate votes in many seats. This means the bigger parties have traditionally done better in the House, where they have less competition.
In contrast, there are sometimes independent candidates who run locally, and don’t have an equivalent option on the Senate ballot. This has become much more of a factor in recent elections.
There is also the question of who can win, and their impact on the result. The major parties until recently were considered the only viable winners in the lower house, and the nature of the House meant the focus was on who would form government. Meanwhile the Senate has been a more viable opportunity for smaller parties to win seats, and a clear history of those electing having influence.
For this post I am going to look at how the difference between House voting and Senate voting has changed over the last half-century, and how it varies between seats across Australia.
Let’s start with the historical lens:
Lines that dip low indicate that the party polled a higher vote for the House than the Senate. The chart dates back to 1974, as there have been no separate House and Senate elections over that period. There were numerous House-only and Senate-only elections in the 1950s, 1960s and early 1970s.
Labor and the combined Coalition parties have consistently had a higher House vote than Senate vote. 2025 was the first time that trend was interrupted, with Labor polling more votes in the Senate than the House, just slightly.
The Greens also polled more in the House than the Senate for just the third time in the party’s history. Traditionally the Greens (like the Democrats before them) did better in the upper house, but as the party has been more competitive in the House and had more competition from smaller left-leaning parties in the Senate, that advantage has been reversed.
The vote for all other parties is also intriguing. It has always been higher in the Senate, but the gap has varied quite a bit. It shrunk tremendously from 1987 to 1990, not because the Senate others vote went down, but because the House mostly caught up, with a vote of over 15% in the House for minor parties and independents excluding the Greens. The Senate vote kept growing, but the House vote fell back for some time.
The gap for other parties and independents widened to a new peak in 2013-2016, with the last GVT election and a double dissolution election making a wide range of parties viable in the Senate.
But more recently the others vote in the House has caught up on the Senate, primarily thanks to a surge in independents running in the House. It’s possible One Nation’s efforts in running candidates almost everywhere has helped.
Unsurprisingly, these national trends have not played out everywhere equally. This table breaks down the gap for the main groups by state:
The ACT stands out tremendously, thanks to the impact of David Pocock. Pocock’s presence in the Senate significantly depresses the Labor and Greens votes compared to the House. We’ll see later on that the same phenomenon can be seen in many House seats in reverse, but those only make up small parts of large states, whereas the Pocock phenomenon covers the whole territory (although it is dampened in Bean by Jessie Price’s candidacy, with Price’s vote correlating strongly with Pocock’s vote at the booth level).
Labor’s overperformance in the Senate is focused in NSW, Victoria and Western Australia – three states with incumbent independent MPs. Interestingly Andrew Wilkie’s candidacy leads the Greens to have a big Senate bias, but not Labor. When I look at the individual seats, you can see that Labor did poll about 15% better in Clark in the Senate, but that is cancelled out by the other Tasmanian seats having quite strong House biases. I believe this is due to Jacqui Lambie’s presence across Tasmania cancelling out Wilkie’s big impact in Clark.
The Coalition’s Senate vote was generally lower everywhere. The ACT stands out (presumably reflecting that some Liberal voters have switched to Pocock), but also Queensland. After investigating the case of Queensland, I believe this reflects the vote for Gerard Rennick. His party polled 1.9% in the House and 4.7% in the Senate. If you subtract that gap from the LNP’s 4% Senate deficit, it brings the state roughly back in line with other states.
For each of these four blocs, I’ve also made a scatterplot showing each individual seat, and how the House and Senate vote compares.
A majority of seats have a close relationship between the Labor vote in the two houses. The seats which stand out are quite obvious.
There’s a big cluster of seats where Labor’s House vote is repressed to around 10-20%, often with Senate votes above 30%. Quite a lot of these seats are in NSW, which has been marked in orange.
There are 14 seats where Labor polled 10% or more worse in the House than the Senate. An independent won eleven of those 14 seats. The only exceptions were Goldstein, Wannon and Cowper, where independents had serious contests against the Coalition. This dynamic has always existed, but usually on a much smaller scale.
On the other hand, the ACT seats of Canberra and Fenner stand out with a much higher Labor vote in the House than the Senate, due to David Pocock’s candidacy. The effect is muted in Bean due to Jessie Price.
There’s much less deviation in the Coalition vote, although those independent seats do have a Senate vote that implies a higher House vote than actually results. Standouts include Calare, Fowler, Calwell, Mayo, Indi, Wentworth, Mackellar and Curtin. I think the rest of the chart implies that the “natural” Liberal House vote in these seats would be about 5% higher.
As for the Greens, there is again a close relationship which seems to naturally produce a slightly higher House vote than Senate vote. But that trend is up-ended in seats with prominent independents who cut into the Greens primary vote: Clark and Franklin in Tasmania stand out with a particularly high Greens Senate vote, but also teal seats like Curtin.
As for the “others” vote, the majority of seats follow a fairly tight pattern where the Senate vote is usually slightly higher than the House vote. But seats with prominent independents (win or lose) deviate from this pattern. And again, the ACT seats have the opposite pattern. Bean actually looks like it fits the standard pattern, just at the extreme end, with a teal independent falling just short in the House and a strong independent senator.
Kennedy also stands out with a huge vote for parties outside of the major parties. There was a 20% primary vote for Gerard Rennick and another 10% for One Nation.
Finally, I’ve included the same data in a map for Labor, the Coalition, the Greens and One Nation.
“where independents had serious contests against the independent”
Fascinating result in Wentworth. With a winning 2 party preferred vote of + 8.24 that would indicate that Allegra Spender draws only + 2.7% of her 58.4% two party preferred vote from otherwise rusted on Liberal voters. It is not her winning margin this year. & her broadly speaking Liberal agenda on taxation seems to cut no ice with these Liberal when it comes to polling booth time?
@Ben re: Kennedy. Kennedy stands out as the KAP ran a joint ticket with Gerard Rennick’s PFP. That would explain the high 20% there.
There is an interesting correlation between KAP’s HOR and SEN vote. Kennedy accounted for 30% (2016), 31% (2019) and 46% (2022) of the KAP Senate Share. Herbert was the second best seat for KAP Senate vote over the same period, averaging around 10% of the KAP Senate Vote. Having Bob on the ballot in Kennedy and trying to get him re-elected, quite a few voters then flow that through to the Senate. In other electorates, as there’s not that strong focus (resources/name recognition/etc) the KAP Senate vote drops and just scatters. 2022 was interesting in that 70% of the KAP QLD Senate vote came from the 4 electorates of FNQ that KAP focuses on, Leichhardt, Kennedy, Herbert and Dawson.
The 10% share for ONP is interesting…it could be that the joint ticket meant some didn’t see Katter’s name and went to the next best thing, ONP, while others like Katter in HOR and like ONP in SEN. One of those unknown unknowns but would interesting to see the Senate vote in 2028 for this electorate if Bob isn’t running and whether ONP absorbs the rest of the voters or if it spreads back to ALP/LNP.
As an aside, KAP have succeeded where JLN failed, having a very regional based party with long-term members at both Federal and State Level, while having it originally based on one strong personality. (With Lambie saying this is her last term and not running in State Elections, I’m surprised she even keeps the JLN party machine and not just revert to a pure independent.) Then the results in Tassie will deviate differently as people come back to choose between LIB vs LAB for the last seat…unless Tammy gets her brand going.
The Sydney seats show how much personal votes helped even when Dutton was toxic.
@ NP
Dutton may not have Been toxic in peri-urban and rural areas For example Hume, Hawkesbury etc also Casey in Victoria i posted there where i think Dutton may have been liked.
The shambolic Liberal campaign, combined with Labor’s overperformance in securing 94 seats in the House, suggests a volatile electorate to me. The difference between the aggregate polling and the final result indicates a higher-than-average number of undecided voters who broke heavily in favor of Labor late in the election. The crossbench is growing in the House at the expense (mostly) of the Liberals. The chart showing the difference between House and Senate votes by party group is eerily reminiscent of a similar outcome in the recent Canadian election. Much of the minor party votes in Canada shifted to the centre-left Liberals in response to Trump. True Canada was buffeted by Trump much more directly, but I wonder if the Dutton-Trump comparison had an impact here. Perhaps voters decided that Labor was going to form the government, and there may have been a tendency by some % to move to Labor. On a side note, it would be interesting if Simon Holms à Court decided to make a play in the Senate.
It has to be remembered that the KAP actually has a philosophy – conservative socially mixed with rural socialism and economic regulation with a regional Queensland twist. In this way it is unlike One Nation or Palmer which are just knee jerk populism. Katter is the product of being a biological ideological love child of Vince Gair and John McEwen mixed with Bjelke-Petersen nurture. Who knows if it can continue post Bob even with Robbie Katter in the mix. Even if it somehow coalesced with One Nation and Rennick, the mix might not work outside of regional Queensland.
It has to be remembered that the KAP actually has a philosophy – conservative socially mixed with rural socialism and economic regulation with a regional Queensland twist. In this way it is unlike One Nation or Palmer which are just knee jerk populism. Katter is the product of being a biological ideological love child of Vince Gair and John McEwen mixed with Bjelke-Petersen nurture. Who knows if it can continue post Bob even with Robbie Katter in the mix. Even if it somehow coalesced with One Nation and Rennick, the mix might not work outside of regional Queensland.
For Climate 200 to make a play for the senate they actually have to have a political party structure as otherwise it is just too hard. That would require some of the Teals (at least) to coalesce into a cohesive political structure – and if that happens their Greens and Labor tactical voters will either have to make choice or return from whence they came.
Where does Dorinda Cox fit, is she in the Greens or Labor senate count?
@Nimalan this would be due to regional areas prioritising economic and agricultural issues even if they don’t agree with his social and environmental policies whereas social and environmental views dominant in the city when the views of the other candidate are extreme.
The Liberals are not wholly unpopular, or they would have disappeared. Their current offer is unlikely to garner them a majority. Dutton himself, along with the gaffes related to work-from-home arrangements and comments about making government workers redundant, were merely symptoms of a bigger problem. The centre of the electorate was here, and the Liberals were over there, and Trump was in the middle. Of course, there is always a certain percentage of inelastic party voters who will mark their ballots accordingly, regardless. The problem now is that there is an ever-growing percentage of elastic voters in urban areas who are willing to move, and they are forming the foundation of the teal vote.
A most interesting analysis, thanks Ben.
However, I still don’t understand why you suddenly switch from ‘independents’ to ‘teals. I really am interested why someone like you, recognising the community independent phenomenon, still do this. Perhaps, at some stage, you could a blog on this issue?
It’s pretty simple Louis. “Teal” and “independent” mean different things. I say independent when I refer to all independents or a broader range, or in some cases to an independent who is not a teal.
I think it is a useful description for a particular kind of seat or a particular kind of independent which is how I’ve used the term. Unlike “community independent” which is meaningless nonsense.
Calling someone a “teal” is not an insult. It’s just a description of a broad identifiable group of people with commonalities. This is very tedious.
The seat by seat analysis is very interesting. Two obvious things stand out to me in WA:
1. Andrew Hastie in Canning has a HUGE personal vote – the coalition vote in Canning (all him – there was no Nat) is almost 12% higher than the coalition senate vote in Canning.
2. Kate Chaney (Teal Ind.) in Curtin derives almost all her support from people who otherwise vote Greens or ALP.
Andrew Hastie will be running the numbers after July 1 when at least 2 maybe 3 of sussans backers will be out of parliament. thentake into account those who voted for Susan because they didnt want Taylor and he should be able to easily win should he choose to challenge
3 of my children voted in their first federal election on this occasion.
One has finished his degree and paid his HECS outright and the two siblings are still studying.
All three voted ALP on the promise of the fee relief.
They seemed surprised / incredulous when I said a lot of people voted for electricity bill relief last time and that wasn’t delivered and until you see the bill passed neither will the fees go down.
Same tactics different mugs?
Still doesn’t explain why ALP deserved the result save it was lost by the Libs – and perhaps, justifiably so !
The promised electricity bill relief was delivered.
Sounds like a lying parent has been found out by their kids.
Senate votes are interesting and helpful as a hypothetical result when no independents contest seats as well as to show the floor of a parties popularity
@watsonwatch really? Do you know anybody’s electricity bill that fell by $275?
By this metric, the most personally popular MHR (Labor, Coalition or Greens that is) in Australia is Barnaby Joyce – outperforming his own team in the Senate by 15%!
Could be some Kate Chaney style tactical voting from Greens/Labor helping ol’ Barnaby over the line there, imo.
Allan
In Canning, Andrew Hastie had an 8.63% premium over Senate Coalition vote.
(Nats were running in Senate)
Susan Templeman also a very good local member confirmed.
Alan,
The National Energy Bill relief delivered as a credit on domestic electricity bills.
The line item is labelled ‘Australian Government Energy Bill Relief’. It was $150 per year split into two instalments 6 months apart. The most recent one I received was on 13 April 2025.
Agree Watson, I also received that benefit too.
I believe Alan and J knight are referring to the base price without rebate not reducing which would make sense.
Gympie why on earth would Greens/Labor voters support Barnaby?
@redistributed: the KAP antecedents are even more complex. The Katters (original father & son) were ALP members energetically trying to “knock off” the local communists who included Bob’s uncle. It is long forgotten that North Queensland was the only part of Australia where the communist Party reached an electoral mass to elect a state MP.
@Alan: you can do it more widely & you discern quickly that Labor voters, whether in Calare, Wentworth, Indi, Curtin or Cowper are skilled tactical voters. The Labor primary vote, where & when it matters, is clearly higher than 34 – 35% commented on derogatively by some commentators.
@watson watch – I believe @jknight was referring to the election promises made before the 2022 election, when the then Albanese Opposition said that if it were elected average electricity bills would fall by $275 per annum. The fact that that didn’t happen is presumably what prompted the rebates you’re referring to.
@rogerroughead I’m what you say is right – in all the teal-held seats, not just Curtin, most teal voters are Labor or Greens supporters voting tactically; very few of them are estranged Liberal voters. The import of this, I would say, is that the Liberals should not suppose that nominating more centrist candidates will win teal voters back to them, most teal voters being people who would never contemplate voting Liberal, whatever kind of candidate they nominated. I seem to recall that the teal tide in 2022 swept away progressive Liberals and conservative Liberals with -liberals and conservatives with no discernable difference.
@Alan – If teal voters were just former Greens and Labor voters than the teals wouldn’t have won any seats as in most of the seats that the currently hold the Lib primary would have been over 50% not that long ago. They had to pick up former Liberal voters to be able to win
In terms of the local liberal candidates, I think it was more about the party leadership than the local member. What is the point of a “moderate” local Liberal MP if you hold moderate views but they just get ignored by the hard right libs and Nats?
Bazza without being knowledgeable to take a side on this issue, their argument is that what’s really happening is a long-term trend towards the left in these seats, and the teals only peel off a small amount of liberal support, but due to a perception that tactical voting was needed, they got up even if Labor would have been close otherwise.
@bazza – an alternative explanation is that in the teal seats, older Liberal voters have died off (or moved to the Gold Coast), and younger voters, who’ve never voted Liberal, taken their place, thereby causing a leftward shift in the seat over time which doesn’t rely on any former liberals changing their vote.
I agree that ‘moderate’ Liberals don’t seem to have attracted any personal following in seats they’ve contested. The fact that conservative Liberals, like Andrew Hastie, can do so, but moderates it seems can’t, would tend to suggest that it would be a feeble and ineffective strategy for the Liberals to consciously track to the centre in an attempt to make themselves more appealing in teal-held seats.
Allan
Nah. Primary swing against in MacPherson, Fadden and Moncreiff averaged 5.1%, 2PP average Swing against LNP of 3.67% across the slate: No new Lib voters here. If by Gold Coast you mean Gold Stein, sure.