What is the crime rate in South Melbourne?

South Melbourne recorded 3,561 offences in the year ending 31 March 2026. This is the latest complete locality-level figure published by Victoria's Crime Statistics Agency as of 1 September 2026.

The agency does not publish an official population-based crime rate for the South Melbourne suburb record. An indicative calculation produces 30,837 recorded offences per 100,000 residents. It combines the 3,561 offences with the 11,548 people counted in South Melbourne at the 2021 Census.

That calculated rate needs a clear warning. The offences occurred from 1 April 2025 to 31 March 2026, while the population comes from 2021. The Crime Statistics Agency and Australian Bureau of Statistics also use different methods to define locations. Their South Melbourne boundaries are not confirmed as exact matches.

The safest answer is therefore precise: South Melbourne had 3,561 recorded offences during the latest reporting year. Its indicative rate was 30,837 per 100,000 residents, but that rate is an editorial calculation rather than an official police rate. Neither figure proves that a resident has a particular chance of becoming a victim.

Which figure gives the clearest picture of recorded crime?

The most reliable starting point is the official offence count. The Crime Statistics Agency release for the year ending March 2026 was published on 18 June 2026. Its workbook lists 3,561 recorded offences under the agency's Suburb/Town Name of South Melbourne.

The reporting period matters as much as the total. It covers one full year, from 1 April 2025 through 31 March 2026. A figure labelled 2024, a calendar-year count or a result from an earlier annual release does not measure the same period.

The workbook contains a historical series beginning in 2017. That makes it useful for checking changes over time, provided each result uses the same measure and geography. Taking one figure from a suburb table and another from a council or police district can create a false trend.

The word offence also matters. The total is not the number of people arrested, charged or found guilty. It is not the number of residents who contacted police. It counts alleged violations recorded by Victoria Police and assigned to the South Melbourne locality.

A single event can involve more than one recorded offence. One person can appear in several events. An offence can affect someone who lives outside the suburb. People may also report events that occurred at a workplace, shop, venue, car park or other local site even though they live elsewhere.

This is why the count should always appear with four details:

• The measure is recorded offences.
• The total is 3,561.
• The period is 1 April 2025 to 31 March 2026.
• The geography is the Crime Statistics Agency Suburb/Town Name of South Melbourne.

If another source gives a different figure, check those details before treating either number as wrong.

How is the indicative population rate calculated?

Readers often want a rate because raw totals are hard to compare across places with different populations. The Crime Statistics Agency workbook does not provide an official population-based rate for its South Melbourne suburb record, so the indicative figure must be calculated from separate sources.

The Australian Bureau of Statistics 2021 Census QuickStats counted 11,548 people in the South Melbourne Suburb and Locality, code SAL22310. Combining that population with the official recorded-offence count gives this calculation:

3,561 divided by 11,548, multiplied by 100,000, equals 30,836.5.

Rounded to a whole number, the result is 30,837 recorded offences per 100,000 residents. The same calculation can be expressed as approximately 308.4 recorded offences per 1,000 residents.

Both versions describe the same indicative rate. They do not mean that 308 out of every 1,000 residents experienced crime. Recorded offences and residents are different statistical units. The numerator can include repeat offences, several offences connected with one event and offences involving people who do not live in South Melbourne.

The denominator also predates the offence period by several years. The Census recorded the population in 2021, while the offence count covers part of 2025 and part of 2026. Any population change between those periods is absent from the calculation.

There is a second limit. The offence total uses a police location assignment, while the population uses an ABS statistical boundary. Both carry the name South Melbourne, but the supplied methods do not establish that they cover exactly the same area.

For these reasons, the rate must always be labelled indicative. Calling it an official South Melbourne police rate would give it authority that the Crime Statistics Agency has not provided.

The figure still has value. It converts the count into a familiar scale and shows how a population denominator changes the way the total looks. Its best use is as a cautious reference point, not a precise forecast of personal risk.

Which offences account for most of the total?

Property and deception offences formed the largest broad class. South Melbourne recorded 2,489 property and deception offences in the year ending 31 March 2026, according to the Crime Statistics Agency recorded-offences workbook.

During the same period, the suburb recorded 434 crimes-against-the-person offences. Both numbers are offence counts, not victim counts.

This split changes how the headline total should be read. A single overall figure combines forms of offending that can affect people in very different ways. Property and deception is a broad class containing several distinct offence types. Crimes against the person is also a broad class. The class totals do not reveal the frequency of each offence type within them.

It would be wrong to describe all 2,489 property and deception offences as theft or burglary. Those terms refer to narrower offence types. The supplied figures support only the broad-class total. They also do not support claims that South Melbourne holds a particular rank for burglary, assault or violent crime among Melbourne suburbs.

Someone assessing local risk should match the offence category to the decision being made. A person concerned about a parked vehicle needs different data from someone reviewing personal safety during a daily routine. A business assessing property security also has a different question from a household checking activity near a home.

Consider an explicitly hypothetical example. Two suburbs each report the same total number of recorded offences. One total is driven mainly by property and deception offences, while the other contains a much larger share of crimes against the person. Calling both places equally safe or unsafe based on the total alone would hide a material difference in the types of recorded activity.

South Melbourne's figures show why the offence mix deserves attention. Property and deception contributed far more recorded offences than crimes against the person during this period. That fact explains part of the total, but it does not reduce every property offence to the same event or severity.

Why is the total different from a resident's chance of being a victim?

The Crime Statistics Agency defines a recorded offence as each instance of an alleged violation of criminal law recorded by Victoria Police in its Law Enforcement Assistance Program database. Its methodology on recorded offences and criminal incidents makes clear that these measures describe different parts of police data.

A recorded offence sits at an early stage of the police process. It does not measure the number or type of charges later laid. An allegation can be recorded without producing a charge, and the count does not state the final legal outcome.

The 3,561 total is therefore not a count of unique:

• criminal incidents;
• offenders;
• victims;
• residents affected; or
• charges laid.

One incident can lead police to record several alleged violations. A person or place may be involved more than once during the year. Someone affected by an offence in South Melbourne may live elsewhere. The indicative population rate places all recorded offences over the resident population even though the two groups do not line up person for person.

This issue becomes easy to miss when the rate is expressed per 1,000 residents. The indicative figure of 308.4 offences per 1,000 residents sounds like a percentage, but it is not one. It cannot show that about 30.8 per cent of residents became victims. It compares a volume of police-recorded offences with an older Census population.

The total also cannot settle whether South Melbourne is safe, unsafe, rough or dangerous. Those labels require more specific evidence about offence types, location, timing and exposure. An aggregate count does not say whether activity was spread across the suburb or concentrated around particular sites. The supplied locality data also do not measure unreported crime.

Use the total as an indicator of recorded police activity within the named locality. Use the broad classes to understand what contributes to it. Do not turn either measure into a personal probability that the data cannot support.

How can South Melbourne be compared with another suburb fairly?

A fair comparison begins by defining South Melbourne in the same way on both sides of the calculation. The ABS denominator used here is the South Melbourne Suburb and Locality, code SAL22310. It is distinct from the South Melbourne Statistical Area Level 2, postcode 3205, the City of Port Phillip and any police district.

The Crime Statistics Agency publishes offence information at suburb and town level using police location variables intended to show where an offence occurred. Its geographic methodology also explains that an offence may still be counted in a relevant area when the incident location is uncertain.

A police location assignment is not necessarily identical to an ABS boundary. This does not make either source unusable. It means the indicative rate carries a boundary caveat that should remain beside the number.

Suburb comparisons can fail in several other ways. A list may compare a raw offence count in one place with a population rate in another. It may use a 2024 result for South Melbourne and a later period elsewhere. It may compare all recorded offences with one narrow category such as burglary. It may also use a current offence count with population data from a different year without saying so.

Rankings such as highest crime suburb or top ten crime areas need matched official data for every location. The supplied South Melbourne figures alone cannot support such a ranking. A high raw count may reflect a different population, local activity or reporting geography. A standardised rate can improve comparison, but only when its denominator and boundaries are sound.

Before deciding that another suburb has a higher or lower crime rate, check:

1. Reporting period: Make sure both offence totals cover the same dates.
2. Geographic unit: Compare suburb with suburb, not a suburb with a postcode, council area or police district.
3. Statistical measure: Keep recorded offences, incidents, victims and charges separate.
4. Population denominator: Check the population year and whether it matches the offence period closely enough.
5. Offence category: Compare total recorded offences with total recorded offences, or match the same narrower class.

Actionable takeaway: Use 3,561 as South Melbourne's latest official recorded-offence count, treat 30,837 per 100,000 as indicative only, and apply these five checks before making any safety claim or suburb comparison.

Related guides: What are the rough suburbs of Melbourne?, Is South Melbourne an expensive suburb?.