Five offences are rising across Bathurst and Orange. Not one of them is rising in both

Read the caution first, because it decides how much weight the rest can carry. These are recorded incidents. BOCSAR says plainly that recorded crime statistics for some categories DO NOT accurately reflect the actual level of crime in the community, because the count depends on public willingness to report crime and on shifts in policing policy. It adds that irrespective of how trends in recorded crime are presented, their interpretation is a difficult task. Keep that sentence in view; below, it turns out to be the whole story.

The two lists

Each of those five is Stable in the other city on BOCSAR's test. Amphetamine possession and breach of bail are not rising in Orange; stealing from motor vehicles, retail theft and trespass are not rising in Bathurst. Bathurst's stealing-from-motor-vehicle count has in fact gone the other way, from 195 two years ago to 110.

The line between the two lists is who reports the offence

Look at what is in each column rather than the percentages. Bathurst's two are offences police find. Orange's three are offences someone reports to police. Nobody rings the station to report their own drug possession or their own bail breach: those numbers move when police look, and BOCSAR's own caution names exactly this category. It says recording of offences which are detected by, rather than reported to police, are strongly affected by policing practices, and it does not leave you to guess which offences it means: its worked examples are drug offences, drink driving offences, offensive behaviour and receiving stolen goods. Of those it concludes, in its own words, Recorded rates for such offences do not accurately reflect actual rates.

Our reading, and it is labelled as ours, though the Bureau has done most of the work. One of Bathurst's two rising offences is a drug offence, which is the first example on BOCSAR's own list of categories whose recorded rates do not accurately reflect actual rates. A 49 per cent rise in amphetamine possession and a 40 per cent rise in bail breaches in one city, with nothing else in that city moving, is at least as consistent with a change in police activity as with a change in behaviour. Orange's three are harder to explain that way, because a stolen car stereo or a shoplifting incident enters the count when a member of the public or a retailer reports it. We are not saying Bathurst's rise is only enforcement, and we cannot rule out that it is partly real. We are saying the two lists are not the same kind of number, and a story that added them together and called it "crime up in the Central West" would be describing something that does not exist in the data.

Levels and movement are different questions

Every one of the five sits well above the state rate, between 1.7 and 2.3 times it, which is worth knowing on its own. But the level and the movement do not track together, and the clearest example is bail.

Orange has the higher breach-of-bail rate of the two cities, at 1,342.7 per 100,000 against Bathurst's 1,237.2, and both are close to double the state's 689. Yet Orange's is Stable and Bathurst's is up 40.1 per cent. The city with more of it is not the city where it is growing. Statewide, breach of bail is also stable, so Bathurst is moving against both its neighbour and the state.

Stealing from motor vehicles runs the same way in reverse. The NSW figure is Stable and falling slightly in raw count, from 28,710 two years ago to 25,054. Orange is up 58.8 per cent over the same period.

What the shires say, which is nothing

Blayney, Cabonne and Oberon return no significant trend on any offence. That is not the same as saying nothing happened there. It is BOCSAR declining to publish a trend it cannot stand behind, which in small populations is the usual and correct outcome, and it is why this piece is about the two cities.

What we are not saying

We have not asked police, either council or the local commands for comment, and this is an arithmetic reading of a published dataset rather than an account of what is happening on any street. Nothing here identifies a person, describes an incident, or touches a matter before a court. We offer no view on whether police should be looking harder or less hard at anything, and the dataset supports no such view.

The question we would ask next, and cannot answer from this workbook, is whether the number of bail orders in force in Bathurst has risen alongside the breaches. A rising breach count with a flat number of people on bail means something quite different from a rising breach count with more people on bail, and only the second is visible in court data.

If your own or someone else's drug use is the part of this that matters to you, the Australian Government's National Alcohol and Other Drug Hotline describes itself as providing free and confidential information and support and can be reached 24 hours a day, 7 days a week on 1800 250 015.