2025-10-30

Aim And Objectives

Aim Tell a clear, evidence-based story about the 2025 Australian Federal Election using open AEC data—highlighting where contests were truly competitive, how preference flows converted primary votes into wins, and how a fragmented primary vote shaped outcomes.

Objectives - Use AEC 2025 House datasets (first preferences, TCP, TPP); clean and harmonise divisions and party groupings.

  • Build non-deceptive visuals (state competitiveness, preference leverage vs primary, national primary mix, TPP spread) with clear titles, subtitles, and sources.

  • Summarise insights (marginals, leverage, party mix) and acknowledge limits (e.g., TPP abstraction, “Other” aggregation).

  • Publish a concise, R (ioslides) web presentation.

State-by-state seat competitiveness (TCP margin)

How Preferences Convert Primary Votes into Seats

Primary Vote Landscape (2025)

The most marginal seats (watchlist)

Top 20 marginal seats (2025 TCP margin, %)
division state winner margin (%) winner_tcp (%)
Bradfield NSW Independent 0.0 50.0
Goldstein VIC Other 0.2 50.1
Longman QLD Coalition 0.2 50.1
Bean ACT ALP 0.7 50.3
Bullwinkel WA ALP 1.0 50.5
Kooyong VIC Independent 1.3 50.7
Fremantle WA ALP 1.4 50.7
Menzies VIC ALP 2.2 51.1
Petrie QLD ALP 2.3 51.2
Solomon NT ALP 2.6 51.3
Bendigo VIC ALP 2.8 51.4
Wills VIC ALP 2.9 51.4
Berowra NSW Other 3.3 51.6
Forde QLD ALP 3.5 51.8
La Trobe VIC Other 4.1 52.1
Forrest WA Other 4.5 52.2
Flinders VIC Other 4.6 52.3
Banks NSW ALP 4.8 52.4
Bowman QLD Coalition 4.9 52.4
Cowper NSW Other 5.1 52.5

ALP seat safety distribution (TPP)

Methods, what we measured & limits

  • Data: Official AEC 2025 House results for every seat

    • first-preferences, two-candidate preferred (TCP) and two-party preferred (TPP).

  • Preparation: Skipped the banner row, fixed headers, and grouped parties into ALP / Coalition / Greens / Independent / Other. Used simple, folder-relative file paths.

  • Measures:

    • Winner & margin in each seat from TCP (margin = winner TCP − runner TCP).
    • Safety band: marginal <6%, fairly safe 6–10%, safe ≥10%.
    • Winner’s primary = that party’s first-vote share in the seat.
    • National mix = first-vote shares summed by party across Australia.

  • Charts shown: (1) Margins by state, (2) Margin vs first-vote (plus boxplots by bands),(3) National first-vote split, (4) ALP TPP distribution.

  • Limits:

    • TPP can hide seats where the final two weren’t ALP vs Coalition.
    • “Other” lumps many small parties; Coalition combines LNP/Nats/CLP.
    • It’s one year only (2025); preference flows are inferred, and we didn’t weight for turnout.

Key Takeaways

  • Not a uniform landslide: Competitiveness varies by state—VIC/NSW have many marginals, QLD/WA skew safe; treat TCP <10% as true battlegrounds.

  • Preferences decided outcomes: With a fragmented primary (large “Other”, strong Greens), winners often turned <45% primary into solid TCP margins via preference flows.

  • National mix matters: Outcomes were preference-driven, not primary-vote alone—ALP is the largest major but relies on transfers in many seats.

  • ALP position: TPP ~55–65% in most divisions (tail >70% safe, <50% vulnerable) → broad advantage with pockets of real risk.

References

Australian Electoral Commission. (2025). House of Representatives — First preferences by candidate by vote type (Event 31496) [Data set]. https://results.aec.gov.au/31496/Website/Downloads/HouseFirstPrefsByCandidateByVoteTypeDownload-31496.csv

Australian Electoral Commission. (2025). House of Representatives — Two-candidate-preferred by candidate by vote type (Event 31496) [Data set]. https://results.aec.gov.au/31496/Website/Downloads/HouseTcpByCandidateByVoteTypeDownload-31496.csv

Australian Electoral Commission. (2025). House of Representatives — Two-party-preferred by division (Event 31496) [Data set]. https://results.aec.gov.au/31496/Website/Downloads/HouseTppByDivisionDownload-31496.csv