NZ General Election 2026 Explorer

NZ General Election 2026 Explorer

Explores New Zealand's 2026 general election with seat allocation, overhang and forecast models, in a Gradio dashboard and a static page. Suits analysts, students and curious readers.

Made by edithatogo on Hugging Face. These notes are Mindket's, written from its Space page and README; the app's own files and license are on Hugging Face.

About

What it is: This Space is one author's analysis of New Zealand's 2026 general election. It models how votes turn into seats, including the extra overhang seats, and adds causal studies, a swing model for special votes and a forecast for 2029. It comes as a Gradio dashboard and as a static page. The author has retired an earlier Streamlit version.

Who it suits: It suits people who follow New Zealand politics and want to test seat scenarios, and students of electoral systems and statistics who like worked examples.

Industry | Who uses it | How they use it | Example
Politics and media | Analysts and journalists | Test how vote shares turn into seats under the allocation rules | Change a parameter and watch the seat counts recalculate
Academia | Political science students | Study disproportionality measures and causal methods on one case | Read how the Gallagher index is reported
Data science | Data scientists | See classifiers and causal designs applied to election data | Look at the special votes swing model
Civic interest | Curious readers | Browse the modelled results and the visual extras | Open the static page and explore the figures

Skip it if you need official results or neutral polling, since every figure here comes from the author's own models.

Where to use it:

• A browser: open the Space's page on Hugging Face, with nothing to install
• A downloaded copy on Windows, macOS or Linux, where you open index.html in a web browser
• The Gradio dashboard run locally with Python

Use case | Who | What they get back
Seat scenarios | Analysts | Seat counts that recalculate when you change inputs
Causal reading | Students | Worked examples of causal methods applied to overhang seats
Browsing visuals | Curious readers | Many figures, a slide deck and interactive pages

How to set it up: The online copy needs nothing. For a copy on your own computer, you need the following.

• Python with pip on Windows, macOS or Linux
• The Hugging Face hub client, which the first step installs
• The packages listed in requirements.txt

1. Install the Hugging Face hub client.
2. Download the Space's files into a local folder, which holds 567 files.
3. Move into that folder.
4. For the static page, load index.html in your browser.
5. For the dashboard, install the packages and start app.py.

pip install -U huggingface_hub
hf download edithatogo/nz-election-explorer --repo-type space --local-dir nz-election-explorer
cd nz-election-explorer

pip install -r requirements.txt
python app.py

How to use it: Online, open the Space's page. Locally, start the dashboard with the command below, or open the static page, which runs inside the browser.

python app.py

Tool | What it does
Gradio dashboard in app.py | The live version, which recalculates results when parameters change and runs the forecasts
Static page in index.html | A client-side version with sound and shader visuals that needs no server quota

Example: You open the dashboard and change a vote share. The seat allocation and the overhang count recalculate, and you compare the new result with the original figures on the page.

For newcomers and experts: If you are new, overhang seats are extra seats a party wins when it takes more local seats than its share of the party vote would give. Open the static page first and browse the figures before you change anything.

If you know the field, the page applies Sainte-Laguë allocation under the Electoral Act and a Shapley split of the overhang seats. It adds a regression discontinuity on incumbency, an instrumental variable design, XGBoost for special votes and a Hibbs-Stokes model for 2029. The source does not list the assumptions behind the headline numbers.

You open the dashboard or the static page
You set vote and model parameters
The seat and overhang models recalculate
Causal and forecast views update

Good to know: The numbers below are what the page reports from its own models, not official counts.

Title: Disproportionality measures the page reports
Gallagher Index: 4.82%
Loosemore-Hanby: 6.14%
Rae Index: 1.24%
Maximum Deviation: 5.34%

• License: apache-2, as listed on Hugging Face
• Cost to run: no API key or paid service is named, and the page says the static copy uses no server quota
• Data it reaches: its own files, plus a public dataset that the README says this Space belongs to
• Not for: the source lists no exclusions, so treat the figures as modelled results and not as official counts

On Hugging Face, 13 November 2026

• Likes: 1
• Built with: static
• License: apache-2.0

Links

• Space: https://huggingface.co/spaces/edithatogo/nz-election-explorer
• Files: https://huggingface.co/spaces/edithatogo/nz-election-explorer/tree/main

Try it

Open the Space in your browser, or copy it to your own Hugging Face account from its page. Nothing in this kit runs by itself.