Arundel is safer than the local-authority average, which is the honest answer. The recorded crime rate here works out at roughly 85 incidents per thousand residents over the last twelve months. The Arun local-authority rate is about 90. That puts Arundel around six per cent below the average for the district it sits in. Small market town, big castle, quiet evenings, and the numbers agree with the impression.
The volume behind that rate is about 370 recorded incidents over twelve months, which is what a town of roughly 3,500 residents plus a heavy day-visitor trade produces. Both the rate and the comparison are police-published figures. Do not try to check them by dividing the incident count by the town's resident population, because the published rate uses a different statistical-area population that includes the visitor footfall and the geography the town's incidents actually occur across. The two numbers use different denominators and cannot be reconciled by hand. What matters is what they say together: the recorded rate is real, the comparison to Arun is real, and Arundel sits at the safer end of the district it belongs to.
That six-per-cent gap is not dramatic. It is not the kind of gap that tells you Arundel is a fortress or that its neighbours are a war zone. It tells you Arundel is a bit quieter than the wider Arun picture, which pools it with Bognor Regis, Littlehampton and the rural parishes. Bognor and Littlehampton carry the coastal-resort profile that comes with a busier evening trade, a bigger night-time economy, and the crime pattern that goes with them. Arundel does not. Its evening trade is a couple of pubs, a couple of restaurants, and people going home reasonably early. The tourists are day-trippers, not night-trippers. The residents are older on average. The town shuts up when it wants to.
What the brief does not carry is street-level crime detail, so this answer will not give you any. Anyone telling you that a particular street or a particular postcode is the trouble spot in Arundel is telling you something they think rather than something the data confirms, and small-town rumours about which corner is rough are usually a decade out of date. If you want the granular picture before you buy, the police-published crime maps are free, they update monthly, and they will show you the actual incident points for the address you care about. Look at those before you look at anything a stranger tells you at a viewing.
Context matters here more than the headline. A crime rate of 85 per thousand is not zero, and no town's rate ever is. What it means in daily life for a market town in West Sussex is what you would expect: some low-level anti-social behaviour, some vehicle crime, some shop-related incidents on the busier days, and a very small volume of anything more serious. It is not a town where safety is a dominant concern for buyers, and it is not a town where safety should be the reason not to buy, but it is also not a town where nothing ever happens.
The verdict on safety is 4 out of 5, which is what you get for a small market town sitting around six per cent below its local-authority average on a modest crime rate, with a quiet evening pattern and a resident base that skews older. Not perfect, better than most of the district, and unlikely to be the reason a serious buyer walks away.