This tool provides a simulation of the election procedure for filling casual vacancies in the ACT Legislative Assembly under their Hare-Clark electoral system. This is done by first estimating the original distribution of preference flows in that electorate. Then, the ballots that contributed to the election of the retiring MLA are determined. Finally, a countback can be run using an arbitrary set of candidates to nominate to contest the vacancy. For more information on how countbacks are performed, visit the relevant page on the Elections ACT website.
While this tool is fairly accurate, it doesn't reproduce the counts of the original distribution perfectly. Counting Hare Clark is hard and there are small procedural differences that vary year by year. The margin of error here is small (<0.5%) but it's still only an estimation. Don't rely on the results. If you want perfect counts, you'll need to use something more advanced like AndrewConway/ConcreteSTV. However, I am open to feedback and suggestions about how to improve my code to run more accurate preference distributions.
The tool is written in Kotlin and contains classes and methods for performing preference distributions and countbacks. This logic is built into a web worker that does the expensive computation in a browser. The counting logic was initially ported from liamblake/hcvote but has been heavily modified.
I've also included a simple web client bootstrapped using facebook/create-react-app that provides a UI for performing these counts.
You can find a hosted version of the tool here.
To build, run ./gradlew build in the root directory. The build targets JavaScript and produces a dist/ folder in the
project root.
To serve on your local machine, you can run:
npm install -g serve
serve -s dist