Methodology

Every suburb gets a single 0–100 AusScore. It starts at a neutral 50, everyday amenities add points, and recorded crime subtracts them:

Score = 50 + amenities (up to +35) − crime (up to ~−39)

It is deliberately safety-first: amenities are capped at +35 while crime can pull down by about −39, so a very safe suburb with modest amenity (≈50) still outranks an amenity-rich suburb carrying the state's worst crime (≈46). Great shops and transport count — but they can never fully cancel out a serious safety problem. Suburbs under 200 residents aren't scored (see below).

Each suburb page itemises every point: the “how this score is built” panel lists each factor, its grade and its exact contribution, and links to the government dataset behind it. Nothing is a black box.

Seven amenities, each graded A–F and worth a share of the +35. The guiding question is whether a suburb's everyday needs are met locally — not whether it has the most of something — so a self-sufficient country town isn't out-scored by sheer big-city quantity. A factor a suburb simply lacks scores 0 — absence of an amenity is never treated as a penalty.

Named shops (supermarkets, hardware, etc.) aren't in any government register, so they never affect the score; where shown, they are clearly labelled as OpenStreetMap context.

We grade amenities on whether a suburb's everyday needs are met locally, not on having the most of something:

• Education is graded on the local school ladder: a town that educates kids prep→year 12 locally earns an A, whether it has two schools or twenty; one rung present is a B; otherwise the grade reflects how far the nearest school is. A small town with its own primary and secondary college rates as highly as a big-city suburb — as it should.
• Parks, recreation & sport and family (childcare) are graded on how much is available locally (inside the suburb or a short trip away), on absolute thresholds — not ranked against Melbourne — with a drive-distance fallback when nothing is local.
• Public transport is ranked within the suburb's region— country towns are judged against other country towns, not metropolitan Melbourne — so limited rural services aren't treated as a failure against a standard that doesn't apply.

Daily essentials, healthcare and emergency services are graded on how close the nearest one is, using distance bands calibrated against the real statewide spread so each grade reflects a meaningful share of suburbs. A facility located inside the suburb counts as in-town, so a large rural town isn't penalised when its own hospital, police or shops sit far from the suburb's geographic centre.

The same Crime Statistics Agency measures used in the grades below, weighted by how directly each affects daily safety, and computed from suburb-level data. Person crime hits hardest; bail/judicial is discounted as it overlaps the others.

Youth offending is published by the CSA only at council level — there is no suburb breakdown — so it is notpart of a suburb's score (stamping one council figure on every suburb would penalise quiet towns for offending happening elsewhere in the LGA). We show it as council-level context on the council and suburb pages instead.

AusBurbs does not change, estimate or invent any data. We take the official government figures exactly as published and apply standard analytical tools — dividing by population to calculate a per-1,000 rate, and ranking against the median — to present them clearly. Every figure on this site links back to its original government source, and you can download the same free datasets yourself and reproduce every number. Transparency and reproducibility are core to how we work.

A crime grade compares a suburb's rate to the median Victorian suburb — the typical suburb, not the average (the average is dragged up by a few CBD hotspots). So a grade tells you, plainly, how far above or below the typical rate a suburb sits, and we always show the raw per-1,000 figure beside it:

Raw incident totals unfairly punish big, busy, or commercial suburbs. We convert crime counts into a rate per 1,000 residents using Australian Bureau of Statistics 2021 Census populations, so a quiet street of 400 people and a town of 40,000 can be compared fairly.

Suburbs with fewer than 200 residents are not graded — a handful of incidents would produce a wildly unstable rate. Rather than a dead end, these show the surrounding council's population-weighted figure instead, clearly labelled as a regional fallback.

From the Crime Statistics Agency, graded as three separate measures rather than one vague score:

• Property & Deception — theft, burglary, property damage, fraud (offence division B).
• Person-Related — offences against the person (division A).
• Bail & Judicial — breaches of bail and orders, justice-procedure offences (division E).

Figures are year-ending-December totals; we hold ten years so you can see the trend, not just a snapshot. The grade itself is based on a trailing 3-year averagerate, so a single unusually quiet or busy year can't swing it — the grade reflects a sustained pattern.

Using the public GTFS timetable feed, we count every train, tram, bus and V/Line stop within a 1 km walk of a suburb, weighting rail and coach above bus. Suburbs with no stop within 1 km receive an F; the rest are graded on their relative density.

Each suburb page also names its closest train station(metro and V/Line) — whether it's inside the suburb or how far away — and the lines that run from it with where they go. Rail-replacement coaches are excluded, as they're temporary substitutes rather than scheduled lines.

From the Department of Education's designated-neighbourhood zone maps, we compute which primary and secondary catchments cover each suburb, and how much of the suburb each one covers.

Each suburb page shows both rent and buy cost: the current median weekly rent, and the latest annual median sale price (the most recent full calendar year) for houses and units — each with a multi-year trend. This is context only — deliberately not part of the AusScore: “expensive” isn't good or bad in itself, and folding price into the score would muddle a liveability measure with a value judgement.

Both are medians— the middle value across all rentals or sales in the period, so one unusually high or low figure can't skew them. So rent and buy read consistently, each also shows an average annual change— the compound growth rate of the median from the earliest to the latest year we hold, expressed as a percent per year. It's a simple, like-for-like read on how prices have moved, not a forecast.

Rent comes from the Department of Families, Fairness & Housing Rental Report (Residential Tenancies Bond Authority bond lodgements), reused under CC BY 4.0, and is reported per bedroom. Sale prices come from the Victorian Property Sales Report (Valuer-General Victoria), reused under CC BY 3.0 AU, and are reported by property type (house / unit) — there is no bedroom breakdown in the published sales data. Where a locality isn't reported separately, we label the nearest reported area in the same postcode.

Each suburb opens with a short overview in two clearly-separated parts. The factual “about” sentence — what the place is and where it sits — is taken from Wikipedia and reused under its CC BY-SA licence, with a link back to the article on every page that uses it. We keep only neutral, descriptive wording and omit anything about crime, disasters, politics or religion. The second part — what the AusScore means and which pillars lift or lower it — is generated automatically from the same graded figures shown elsewhere on the page; no opinion is added, and no AI writing service is involved. Suburbs without a Wikipedia article simply show the data-driven summary on its own.

Every pillar on every suburb page carries a “Source & methodology” link to the exact government dataset it was built from — title, publisher, date, licence and a direct download.