At the 2022 Victorian state election, 726,959 enrolled Victorians (16.7%) were not counted in the result for their Lower House seat (see About the data). Most of them did not vote, and the rest cast a ballot that could not be counted. In Broadmeadows it was 28.2%, more than one in four; in Eltham it was 10.4%. Victoria votes again on 28 November 2026.

Chart 1 of 5: How big is the gap?

The gap has widened at every election since 2002

Share of enrolled voters whose Lower House vote was not counted, 1999–2022

Source: Victorian Electoral Commission (n.d.), State election statistics: Lower House voter participation and informal voting rates, 1999–2022.

Chart description: a stacked area chart covering Victorian state elections from 1999 to 2022. The share of enrolled voters who were not counted has risen at every election since 2002, from 10.0% to 17.7%. The jump since 2014 comes mostly from people who did not vote (pink); informal votes (green) form a smaller band on top.

Chart 2 of 5: Where is it happening?

The gap is widest in Melbourne’s outer west and north

Source: Victorian Electoral Commission (2022), 2022 State election results: enrolment, formal and informal votes by district; Victorian Electoral Commission (n.d.), State regions; Australian Bureau of Statistics (2021), ASGS Edition 3: State Electoral Divisions 2022 boundaries.

Use the layer switch to compare the two causes on the same colour scale, and the globe button to zoom out to all of Victoria. Hover over a district for its figures.

Chart description: a map of Victoria’s 88 Lower House districts, shaded by the share of enrolled voters who were not counted in 2022 (Narracan, with no 2022 general-election result, is grey). The darkest districts are in Melbourne’s outer west and north, led by Broadmeadows (28.2%).

Chart 3 of 5: A crowded ballot

More candidates, more spoiled ballots, especially in 2022

Informal vote rate vs number of Lower House candidates, by district
Total candidates: 2014: 545 · 2018: 507 · 2022: 731. Correlation: 2014: r = 0.46 · 2018: r = 0.34 · 2022: r = 0.66

Source: Victorian Electoral Commission (2014, 2018, 2022), 2014, 2018 and 2022 State election results: candidates, formal and informal votes by district; metropolitan/regional grouping from Victorian Electoral Commission (n.d.), State regions (2022) and Australian Bureau of Statistics (2022) Census region names (2014, 2018).

Press Play or drag the slider to move between elections. Each circle is a district, sized by enrolment. Click a legend entry to show or hide metropolitan or regional districts.

Chart description: an animated scatter plot of informal vote rate against the number of candidates for each district at the 2014, 2018 and 2022 elections. In 2022 the number of candidates rose sharply and the relationship strengthened (r = 0.66).

Chart 4 of 5: Two different gaps

In Melbourne, spoiled ballots follow language. In the regions, they follow the length of the ballot.

Informal vote rate by district at the 2018 election, compared with 2021 Census characteristics

Choose what to compare against from the menu.

Source: Victorian Electoral Commission (2018), 2018 State election results: candidates and informal votes by district; Australian Bureau of Statistics (2022), 2021 Census General Community Profile DataPack: State Electoral Divisions, Victoria, tables G01 (language used at home) and G13 (proficiency in spoken English).

Use the menu to change what each district is compared against. The r values update with them. Hover over a circle for the district’s full profile.

Chart description: two scatter plots of 2018 informal vote rates by district, metropolitan Melbourne on the left and regional Victoria on the right. Against language spoken at home, the metropolitan relationship is strong (r = 0.60). Against the number of candidates it is weak in Melbourne (r = 0.16) but strong in regional Victoria (r = 0.71).

Chart 5 of 5: Watch the ballot in 2026

Where the ballot grew between 2018 and 2022, informal voting grew with it

Change in candidates vs change in informal vote rate, 2018 → 2022 (78 districts, r = 0.74)

Source: Victorian Electoral Commission (2018, 2022), 2018 and 2022 State election results: candidates, formal and informal votes by district; Victorian Electoral Commission (n.d.), State regions.

Search for your district, or click a circle, to see how its ballot and its informal vote changed. Double-click to clear.

Chart description: a scatter plot of the 78 districts that existed at both the 2018 and 2022 elections. Districts that gained candidates mostly saw informal voting rise, and those that lost candidates saw it fall (r = 0.74).

What it means for 28 November 2026

Watch the long ballots in the outer suburbs

In 2022 the five districts with the highest informal vote (Melton, Broadmeadows, Point Cook, Werribee, Greenvale) had an average of 12.2 candidates on the ballot, against 8.4 across the state. All five sit on Melbourne’s outer western and northern edge, where Chart 4 shows informal voting also rising with the share of residents who speak another language at home. When nominations close for 28 November 2026, the length of the ballot in those seats will be the first sign of how many votes are at risk. Spoiled ballots are still the smaller part of the story: 515,168 enrolled Victorians did not vote at all in 2022, against 211,791 informal votes, and non-voting is the part that has grown since 2014.

About the data

  • What “not counted” means here. It is enrolled voters minus formal votes for the Lower House district: people who did not vote plus people whose ballot was informal. It is measured against the electoral roll at the close of rolls.
  • Two slightly different totals. Chart 1 uses the VEC’s official statewide participation and informality rates, which give 17.7% for 2022. The opening figure and Charts 2–5 are built from the district results pages and give 16.7%. The VEC calculates its statewide participation rate slightly differently, and the district figures exclude Narracan.
  • Narracan, 2022. The Narracan election failed after a candidate died and was re-run as a stand-alone supplementary election in January 2023. It is shown in grey on the map and is excluded from the 2022 district figures.
  • The roll has changed. Since 2010 the VEC has enrolled people directly from other government records. The roll is now more complete but includes people who may never have engaged, which partly explains the rise in non-voting.
  • District-level patterns are not individual behaviour. Charts 3–5 compare districts. They do not show that any particular person or community votes informally (the ecological fallacy).
  • Boundaries. The 2014 and 2018 elections share one set of boundaries and the 2022 election uses another. Census 2021 data is published for the earlier boundaries, so Chart 4 uses 2018 results, and Chart 5 includes only districts whose name was unchanged in 2022. Their boundaries may still have shifted.
  • Correlation is not cause. The r values describe how closely two measures move together across districts. Other factors, such as age, income and how far people live from a voting centre, are not controlled for.
  • Processing. All data is downloaded from the publishers and processed with 01_prepare_data.R. That script scrapes the VEC results pages, joins Census variables by district name, and simplifies the ABS boundaries for the web map.

References

Australian Bureau of Statistics. (2021). Australian Statistical Geography Standard (ASGS) Edition 3: State Electoral Divisions 2022, digital boundary files [Data set]. https://www.abs.gov.au/statistics/standards/australian-statistical-geography-standard-asgs-edition-3/jul2021-jun2026/access-and-downloads/digital-boundary-files

Australian Bureau of Statistics. (2022). 2021 Census General Community Profile DataPack: State Electoral Divisions, Victoria [Data set]. https://www.abs.gov.au/census/find-census-data/datapacks

OpenAI. (2026). ChatGPT [Large language model]. https://chatgpt.com

Victorian Electoral Commission. (2014). 2014 State election results [Data set]. https://www.vec.vic.gov.au/results/state-election-results/2014-state-election

Victorian Electoral Commission. (2018). 2018 State election results [Data set]. https://www.vec.vic.gov.au/results/state-election-results/2018-state-election

Victorian Electoral Commission. (2022). 2022 State election results [Data set]. https://www.vec.vic.gov.au/results/state-election-results/2022-state-election-results

Victorian Electoral Commission. (n.d.). State election statistics [Data set]. Retrieved October 27, 2026, from https://www.vec.vic.gov.au/results/electoral-statistics/state-election-statistics

Victorian Electoral Commission. (n.d.). State regions. Retrieved October 27, 2026, from https://www.vec.vic.gov.au/electoral-boundaries/state-regions

Software. R Core Team. (2026). R: A language and environment for statistical computing (Version 4.6.1) [Computer software]. R Foundation for Statistical Computing. https://www.R-project.org/ · Sievert, C. (2020). Interactive web-based data visualization with R, plotly, and shiny. Chapman and Hall/CRC. https://plotly-r.com · Cheng, J., Schloerke, B., Karambelkar, B., & Xie, Y. (2026). leaflet: Create interactive web maps with the JavaScript ‘Leaflet’ library (Version 2.2.3) [R package]. https://CRAN.R-project.org/package=leaflet · Pebesma, E. (2018). Simple features for R: Standardized support for spatial vector data. The R Journal, 10(1), 439–446. https://doi.org/10.32614/RJ-2018-009

Acknowledgements

Election data © Victorian Electoral Commission. Census and boundary data © Commonwealth of Australia (Australian Bureau of Statistics), CC BY 4.0. Base map: Esri World Light Gray Canvas (Esri, HERE, Garmin, © OpenStreetMap contributors).

Generative AI acknowledgement

I used ChatGPT (OpenAI, 2026) to check how particular R functions work while preparing 01_prepare_data.R and the charts. I did not use it to generate the story, the written content or the plots. Example prompts:

  • How does plot_ly() with stackgroup turn two percentage series into a stacked area chart, and how do I show the combined total in the hover text?
  • How do I give three leaflet polygon layers one shared colorBin scale and switch between them with addLayersControl()?
  • Why transform the ABS boundaries to VicGrid (EPSG:3111) with st_transform() before st_simplify(), then transform them back to WGS 84 for the web map?
  • How does a plotly updatemenus dropdown replace the x variable in both panels of a subplot(), including the fitted line from lm()?
  • How do I count candidate rows with rvest::html_table() when the archived VEC first-preference tables end with a non-numeric footer row?