On election day in November 2022, 214,410 Victorians queued, collected a ballot paper, and voted, and their votes were thrown away. That is enough people to fill the MCG 2.1 times. They were not evenly spread across the state.
An informal vote is a ballot that can’t be counted: left blank, numbered incorrectly, or marked in a way the rules don’t accept. In Victoria’s Legislative Assembly, a vote only counts if every box is numbered in order. Miss a number, repeat one, or tick a box instead, and the ballot is set aside. Across the state, 5.54% of lower-house ballots were informal in 2022. That figure hides the real story: who those voters are, and where they live.
1 · The hook
Hover over a seat for its numbers. Tick a party to filter; double-click the legend to isolate one party; drag to zoom, double-click to reset.
Scatter plot of 88 Victorian districts: winning margin on the horizontal axis and informal ballots on the vertical axis, both on log scales. A dashed diagonal marks where informal ballots equal the margin; 13 seats lie above it.
Source: Victorian Electoral Commission (2022), 2022 State election results, Legislative Assembly, all 88 districts (scraped from district results pages). Margin = difference between the final two candidates after preferences.
Every dot above the dashed line is a seat where the pile of uncounted ballots was bigger than the gap between winner and runner-up: Northcote, Bass, Pakenham, Mornington, Benambra, Mildura, Hastings, Croydon, Pascoe Vale, Polwarth, Preston, Ripon and Melton. This doesn’t mean those results were wrong. Informal votes would have split between candidates if they had counted. But it shows how much voice was on the table, and in a close 2026 contest these are the seats where it could matter.
2 · It’s getting worse
Hover over each election. Click a legend entry to hide or show that house.
Line chart of informal voting at Victorian state elections from 1999 to 2022. The Lower House rate rose from 3.02% to 5.53% while the Upper House rate stayed lower.
Source: Victorian Electoral Commission (2024), State election statistics: informal voting rates (1999–2022) and numbers of candidates (both houses).
In 1999, 3.02% of lower-house ballots were informal. By 2022 it was 5.53%, peaking at 5.83% in 2018. Over the same years the number of candidates grew from 406 to a record 1,194. The comparison with the Upper House is the telling part: there, a single “1” above the line is a valid vote, and informality stays far lower. The same voters, on the same day, do better on the ballot that asks less of them. This points to the design of the lower-house ballot, not to apathy alone. The VEC’s own review found that most informal ballots in 2022 (about 56%) still showed a preference, suggesting the voter tried and made a mistake, while about 41% appeared deliberate (Victorian Electoral Commission, 2023).
3 · Where
Switch between the two layers (top right) to compare the patterns. Hover a district for its numbers. Use the globe button to see all of Victoria and the target button to return to Melbourne.
Choropleth map of Victoria's 88 electoral districts shaded by informal vote rate, with an alternative layer shaded by the share of residents who speak a language other than English at home. The highest informal rates are in Melbourne's western and northern suburbs, led by Melton.
Sources: Victorian Electoral Commission (2022) for informal votes; Australian Bureau of Statistics (2022a) 2021 Census, table G01, rebuilt on the 2022 district boundaries using the ABS allocation files (ABS, 2022b) and mapped with ABS 2022 State Electoral Division boundaries (ABS, 2022c).
Melton topped the state: 11.1% of ballots, more than one in every 9, were informal. In Hawthorn it was 2.6%. Overall, 5.5% of ballots were informal in Greater Melbourne compared with 5.6% in regional Victoria. Flip the map to the language layer and compare. The six most linguistically diverse districts are St Albans, Laverton, Dandenong, Thomastown, Mulgrave and Kororoit, and their average informal rate was 8.2%, well above the state figure.
4 · Who
These filters control this chart and the next one. Search for your own district, narrow by area, or drag the slider to compare short and long ballot papers. Click a bubble to highlight that seat in both charts; double-click to clear.
Bubble scatter plot of 88 districts: share speaking a language other than English at home on the horizontal axis, informal vote rate on the vertical axis, bubble size showing the number of candidates and colour showing Greater Melbourne or regional Victoria. The correlation is 0.39.
Sources: Victorian Electoral Commission (2022); Australian Bureau of Statistics (2022a, 2022b), 2021 Census table G01 (“language used at home”), aggregated from SA1s to 2022 districts. Correlation is Pearson’s r across the 88 districts.
The relationship is weak positive (r = 0.39). In the quarter of districts with the highest share of residents speaking a language other than English, the average informal rate was 6.9%. In the quarter with the lowest share, it was 5.4%. The length of the ballot paper shows a moderate positive relationship of its own (r = 0.66). Together, language and number of candidates explain about 53% of the difference between districts.
A note of caution. These are district averages, not individual voters. The chart can’t tell us that any particular person who speaks another language voted informally. That would be the ecological fallacy. What it shows is that the communities where English is less often spoken at home are the communities whose votes most often go uncounted. Instructions printed only in English, a long ballot, and a rule that every box must be numbered weigh hardest there.
5 · So what for 2026
Lollipop chart ranking all 88 districts by informal vote rate, from lowest at the bottom to highest at the top, with each line drawn from the state average and dots coloured by Greater Melbourne or regional Victoria.
Sources: Victorian Electoral Commission (2022); Australian Bureau of Statistics (2022a, 2022b). Districts are ranked by informal rate; each line runs from the state average to the district’s value. Use the filters above Chart 4 to search or narrow this list.
Victorai vote again on 28 November 2026. Informal votes aren’t counted, but they arenot random either They gather in the same suburbs, election after election, where ballot papers are longest and where English is least often the language of home. Campaigns tht send volunteers to explain “number every box” in Punjabi, Vietnamese, Arabic or Mandarin, multilingual how-to-vote material at polling places, and a closer look at whether full preferential voting should stay the rule are all practical responses. The voters in these seats have already turned up. The question for 2026 is whether their votes will count,
Data and method. Election results for all 88 Legislative Assembly districts were scraped from the Victorian Electoral Commission’s 2022 results pages (including the Narracan supplementary election of January 2023), and validated against the VEC’s published statewide informal rate. The ABS Census DataPack for State Electoral Divisions uses the 2021 boundaries, which do not match the redrawn districts used at the 2022 election (nine districts were abolished and nine created). To avoid a mismatched join, Census table G01 was downloaded at Statistical Area Level 1 and rebuilt on the 2022 boundaries: each SA1 was assigned to the 2022 district containing the majority of its residential Mesh Blocks, using the ABS allocation files. All processing and every chart were produced in R.
Design choices. Colours follow the Okabe–Ito colour-blind-safe palette, with party also shown by marker shape so colour is never the only cue; the map uses the perceptually uniform viridis scale. Chart 1 uses log axes so that seats of very different size can be compared, with the diagonal kept straight. Every chart has a short “how to use it” note and a hidden text description for screen-reader users.
I use Claude (Anthropic 2026) as a support tool during this assignment for technical guidance on using R and its visualisation packages, refining my own writing, and checking spelling and grammar. I ran and tested the code, checked the results against the published totals from the Victorian Electoral Commission (VEC), and reviewed the data obtained from the Australian Bureau of Statistics (ABS). I made the final decisions regarding the analysis, interpretation, and presentation of the results and take responsibility for my submitted work.
I acknowledge the VEC and the ABS for providing publicly available data, as well as the developers of the R packages used in this assignment.
Anthropic. (2026). Claude [Large language model]. https://claude.ai
Australian Bureau of Statistics. (2022a). 2021 Census of Population and Housing: General Community Profile DataPack, Statistical Area Level 1, Victoria [Data set]. https://www.abs.gov.au/census/find-census-data/datapacks
Australian Bureau of Statistics. (2022b). Australian Statistical Geography Standard (ASGS) Edition 3: Allocation files – Mesh Blocks 2021 and State Electoral Divisions 2022 [Data set]. https://www.abs.gov.au/statistics/standards/australian-statistical-geography-standard-asgs-edition-3/jul2021-jun2026/access-and-downloads/allocation-files
Australian Bureau of Statistics. (2022c). Australian Statistical Geography Standard (ASGS) Edition 3: Digital boundary files – State Electoral Divisions 2022 [Data set]. https://www.abs.gov.au/statistics/standards/australian-statistical-geography-standard-asgs-edition-3/jul2021-jun2026/access-and-downloads/digital-boundary-files
Cheng, J., Schloerke, B., Karambelkar, B., & Xie, Y. (2024). leaflet: Create interactive web maps with the JavaScript ‘Leaflet’ library [R package]. https://CRAN.R-project.org/package=leaflet
Cheng, J., & Sievert, C. (2023). crosstalk: Inter-widget interactivity for HTML widgets [R package]. https://CRAN.R-project.org/package=crosstalk
Okabe, M., & Ito, K. (2008). Color universal design (CUD): How to make figures and presentations that are friendly to colorblind people. https://jfly.uni-koeln.de/color/
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
R Core Team. (2026). R: A language and environment for statistical computing (Version 4.5.3) [Computer software]. R Foundation for Statistical Computing. https://www.R-project.org/
RMIT University Library. (2025). Artificial intelligence (AI): Acknowledgement and referencing. https://rmit.libguides.com/referencing_AI
Sievert, C. (2020). Interactive web-based data visualization with R, plotly, and shiny. Chapman and Hall/CRC. https://plotly-r.com
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. (2023). Victorian State election 2022 [Annual report 2022–23]. https://www.vec.vic.gov.au/about-us/publications/annual-reports/selections/victorian-state-election-2022
Victorian Electoral Commission. (2024). State election statistics [Data set]. https://www.vec.vic.gov.au/results/electoral-statistics/state-election-statistics