Data & methodology
Last updated: 3 August 2026
Balworth is a picture of published statistics. This page tells you which statistics, when they were collected, how we processed them, and where they fall down. If something here changes your mind about a colour on the map, that is this page working.
Balworth covers England today. Every region we add brings its own official sources, and each will be documented here in the same way.
The one-paragraph version
Almost everything on Balworth is published for a small statistical area of roughly 1,500 people, not for a street or a building. We paint the area's value onto the buildings inside it because that makes the map readable — but the number belongs to the area. Colours rank each area against every other area in England, in tenths. Two neighbouring houses can be different colours only because a statistical boundary runs between them.
Where each layer comes from
Deprivation — Indices of Deprivation 2025 and 2019
Published by the Ministry of Housing, Communities & Local Government. Geography LSOA (about 1,500 people). Vintage IoD 2025, built from data collected in the years before publication; IoD 2019 kept for comparison. Licence Open Government Licence v3.0.
The IoD combines seven domains — income, employment, education and skills, health, crime, barriers to housing and services, and living environment — into one overall index. Balworth shows the overall index and each of the seven domains as national deciles, where 1 is the most deprived tenth of England and 10 the least.
What it is not. The IoD is a relative ranking of areas against each other. It is not a measure of anyone's circumstances. Every deprived area contains people who are not deprived, and every affluent area contains people who are struggling. It also has no absolute scale: if England as a whole got poorer, the ranking would look identical.
Crime — data.police.uk street-level records
Published by the Home Office and police forces via data.police.uk. Geography we aggregate individual records to MSOA 2021 (about 7,500 people) and show the result across the LSOAs inside it. Vintage the last 12 complete months, currently to May 2026, refreshed monthly. Licence Open Government Licence v3.0.
We show offences per 1,000 residents per year, split into violence, burglary, theft, anti-social behaviour, vehicle crime and drugs, plus a total and our severity index.
Four things to know before reading a crime colour:
- The locations are deliberately approximate. police.uk snaps every crime to an anonymised map point near where it happened, to protect victims' privacy. Street-level precision is fuzzy by design, and a crime can land on the wrong side of a boundary.
- Recorded crime is not experienced crime. Only offences reported to and recorded by the police appear. Reporting rates vary a lot between areas and between offence types. An area with more recorded crime may simply be an area where more people report it.
- Rates are per resident. A city centre, a retail park, a transport hub or a seaside town has far more people in it by day than sleep there at night, so its rate per resident looks worse than the risk to any one person in it. This is the single biggest distortion on the crime layers.
- Twelve months is a short window. One unusual month moves a small area a long way.
Crime severity — our own index
This one is ours, and you should treat it accordingly. "All crime" counts a stolen bicycle and a robbery as one offence each, which is not how anyone experiences a neighbourhood. So we weight each offence by roughly how seriously the justice system treats it, then express the total per 1,000 residents.
The weights are relative units, loosely proportional to sentencing severity:
| Offence category (police.uk) | Weight |
|---|---|
| Robbery | 900 |
| Violence and sexual offences | 600 |
| Possession of weapons | 400 |
| Theft from the person | 300 |
| Burglary | 250 |
| Other crime | 150 |
| Drug offences | 120 |
| Vehicle crime | 90 |
| Criminal damage and arson | 90 |
| Other theft | 75 |
| Bicycle theft | 60 |
| Public order | 45 |
| Shoplifting | 25 |
| Anti-social behaviour | 10 |
So one robbery counts about the same as 36 shoplifting incidents. The resulting number has no real-world unit — it is only useful for comparing one area with another.
It is not the ONS Crime Severity Score. Our index was inspired by that work and shares its logic of weighting by sentencing severity, but the weights are our own, applied to police.uk's coarse public categories rather than to detailed offence codes. It is not produced, endorsed or checked by the ONS, the Home Office or any police force. Where our index and the raw offence counts disagree, look at both — they answer different questions.
House prices
Published by the Office for National Statistics, from HM Land Registry records. Geography LSOA. Vintage the most recent ONS release available when we last rebuilt this layer. Licence Open Government Licence v3.0.
This is the median price actually paid for homes that sold in the neighbourhood during the period — not an asking price, not a valuation, and not an estimate for your specific house. In areas where few homes sold, a handful of unusual sales moves the median a long way. A neighbourhood of large Victorian houses and small flats produces one middling number that describes neither.
Affordability
Median house price divided by a modelled estimate of median household income before housing costs, at MSOA level. Income at small-area level is a modelled estimate, not a count — it carries real uncertainty, and combining two figures from different geographies and periods adds more. Read this layer as a rough signal, not a measurement.
Who lives here — Census 2021
Published by the Office for National Statistics. Geography LSOA 2021. Collected 21 March 2021 — a single day, during a pandemic, when a lot of people were not in their usual home. Licence Open Government Licence v3.0.
We show shares of residents by country of birth, religion, age, highest qualification and household tenure. These are counts of a population at one moment. Areas change: new housing, new students, new arrivals. A figure from 2021 can already be well out of date in a fast-changing neighbourhood.
What these layers are for, and what they are not for. They describe who the census recorded living somewhere. They tell you nothing about any individual, nothing about whether a place is welcoming, and nothing about whether it would suit you. Please do not use them to choose or avoid an area because of the religion, ethnicity, nationality or country of birth of the people who live there — and never use Balworth to make decisions about anyone's access to housing, lending, insurance, employment or any other service. That is against our Terms, and in some cases it is against the law.
Air quality
Published by Defra, via uk-air.defra.gov.uk. What annual mean PM2.5 (fine particles) from Defra's modelled 1 km background grid. Licence Open Government Licence v3.0.
This is a model, not a measurement at your front door. It describes background concentrations over a 1 km square. It does not capture the difference between the front of a house on a main road and the garden behind it, and it does not include indoor air, a neighbour's wood-burner, or a single busy junction.
Road traffic
Published by the Department for Transport. What the highest annual average daily traffic count on any major road within 2 km. Licence Open Government Licence v3.0.
DfT counts cover major roads only, and many are estimates between survey years. "Within 2 km" is a deliberately blunt rule: a motorway two kilometres away registers here even if you cannot hear it, and a busy residential rat-run does not register at all. Treat this as "how close is the big road", not "how noisy is this street".
Places on the map
From OpenStreetMap, via our own extract and processing. How many about 77,000 points across England. Licence Open Database Licence (ODbL) — © OpenStreetMap contributors.
OpenStreetMap is maintained by volunteers. Places may be missing, closed, moved or wrongly labelled, and coverage is better in cities than in the countryside. The circle drawn around a place when you click it is a rough straight-line distance ring, not a walking route and not a walking time. A real walk is always longer than the ring, sometimes much longer where there is a river, railway or dual carriageway in the way.
Context overlays
Flood risk zones, green belt, national parks, national landscapes, conservation areas and ancient woodland come from planning.data.gov.uk, drawing on the Environment Agency, Natural England and Historic England, under the Open Government Licence v3.0. These are planning designations with legal definitions of their own — check the source before relying on one. Flood risk zones in particular describe modelled probability, not a prediction, and do not account for flood defences in the way an insurer will.
Boundaries and the base map
Boundaries and the building-clipped vector tiles come from the ONS Open Geography Portal and contain OS data © Crown copyright and database right. The base map is derived from OpenStreetMap.
How we put it together
Everything lands on one canvas
The tiles we paint are keyed to LSOA 2011 boundaries. Our sources are not all published on that geography: Census 2021 and IoD 2025 come on LSOA 2021, and crime and the living layers are aggregated to MSOA 2021. We map them across using the ONS best-fit lookups.
This is honest but imperfect. Where boundaries were redrawn between 2011 and 2021 — usually because a lot of new housing was built — a value gets attached to an area that is not quite the area it was calculated for. It matters most in the places that have changed most. We would rather tell you than pretend the joins are seamless.
What the colours mean
Colours are a ranking, not a scale. For each layer we sort all of England's areas by value and split them into ten equal-sized groups. Your area's colour tells you which tenth it falls in — not how far it is from the next area along.
Two consequences worth holding on to:
- The middle is exaggerated. Most areas cluster near the average, so tiny real differences get stretched into visible colour differences. Two very differently coloured areas in the middle of the range may be almost identical in the underlying number.
- The extremes are compressed. The worst tenth contains everything from "somewhat worse than average" to "the worst area in England", all in one colour.
Always read the number in the area card, not only the colour. Green ranks better on the measure you have chosen; rose ranks worse. On the People layers there is no better or worse — the colour only shows more or fewer.
How often it updates
Crime refreshes monthly, as police.uk publishes. Everything else updates when its publisher issues a new release — which for the census means once a decade, and for the deprivation indices roughly every five or six years.
What this map cannot tell you
The useful list, in the order people get caught out.
- What a particular house or street is like. Every value belongs to an area of roughly 1,500 people. Nothing on Balworth is calculated for an address.
- Anything about an individual. Assuming an area's statistic applies to a person who lives there is a well-known reasoning error, and it is wrong far more often than it is right.
- How safe you will feel. Recorded crime, resident-based rates and our own severity weighting together produce a comparison between areas — not a risk to you.
- What a place is like now. Census data is from 2021. Deprivation indices are built from data collected years before publication. House prices are completed sales in a past period.
- Whether you will like it. Light, noise, neighbours, the walk to the station in the rain, whether the bakery is any good. None of that is in any dataset.
- Anything about who should live somewhere. Balworth describes populations. It does not recommend, rank or endorse them, and it must not be used to decide anyone's access to housing, money, work or services.
So what is it good for?
Shortlisting, and asking better questions. Use Balworth to narrow twenty places to five, and to notice things worth checking — a price that looks odd for the area, a quiet pocket two streets off a main road, a school run that turns out to cross a dual carriageway. Then go and stand there on a wet Tuesday evening. No map replaces that.
Licences and attribution
Contains public sector information licensed under the Open Government Licence v3.0: Indices of Deprivation (MHCLG); Census 2021, house prices, boundaries and vector tiles (ONS); recorded crime (Home Office, via data.police.uk); air quality (Defra); traffic counts (DfT); planning designations (planning.data.gov.uk, drawing on the Environment Agency, Natural England and Historic England).
Contains OS data © Crown copyright and database right 2026.
Places and the base map are derived from OpenStreetMap — © OpenStreetMap contributors, available under the Open Database Licence (ODbL).
None of these publishers endorse Balworth or have checked what we do with their data.
Found something wrong?
Tell us. If a number looks impossible, a place is in the wrong spot, or this page is out of date, write to help@balworth.com and we will check it. Corrections make the map better for everyone.