Bristol · Release comparison · 2 day interval
Bristol Observation Change Note: 18 to 20 August 2026
A fixed-grid comparison of two rental observation releases, separating Bristol allocation change, national record movement and collection completeness from claims about the underlying housing market.
Abstract
The evidence system accumulated 1,570,931 additional dated market states while the national current map changed by only +104 records. Bristol’s weighted allocation moved from 354 to 338, but collection completion improved at the same time. This is therefore a release-monitoring result, not evidence that Bristol’s underlying rental supply contracted by 4.5%.
current Bristol allocation
Weighted current record representations, not unique homes
allocated record change
A 4.5% release-to-release decrease
net national change
Mapped current records increased by 0.1%
new dated market states
A 14.2% increase in retained repeated observations
Principal findings
What the comparison supports
The observation history grew much faster than the current map.
Dated market states increased by 1,570,931 while net mapped current records increased by only 104. The collection system accumulated substantial new time evidence without a comparable rise in the current cross-section.
The Bristol allocation fell while collection completion improved.
The weighted Bristol local-authority allocation moved from 354 to 338 records, a decrease of 16. Over the same interval, weighted search completion increased from 25% to 100%, so the count change cannot be read as simple evidence of shrinking supply.
A nearly flat national total concealed large spatial movement.
Of 3,847 fixed cells, 2,199 changed. Gross increases totalled 10,083 records and gross decreases totalled 9,979, leaving a net change of 104. The stable headline total therefore masks substantial redistribution in the released observations.
Figure 1
National stability, local movement
The same fixed grid is used in both releases. The index makes the different scale of change visible without confusing Bristol’s weighted allocation with the national record total.
Reading note. The axis is deliberately narrow and labelled. It compares proportional release change only; it does not compare the size of the two markets.
Table 1
How the Bristol allocation changed
Five fixed cells intersect the Bristol boundary. Each cell is sampled spatially and apportioned across the local authorities it intersects. The published area total is the sum of the allocated column, not the raw cell column.
| Observation cell | 18 Aug cell | 20 Aug cell | 18 Aug allocated | 20 Aug allocated | Allocated change |
|---|---|---|---|---|---|
| Bristol | 527 | 521 | 242 | 239 | -3 |
| Severn Beach | 200 | 180 | 88 | 80 | -8 |
| Kingswood | 132 | 106 | 19 | 15 | -4 |
| Mangotsfield | 90 | 75 | 4 | 3 | -1 |
| Portishead | 10 | 6 | 1 | 1 | 0 |
| Bristol total | Weighted boundary allocation | 354 | 338 | -16 | |
Collection completion
25% to 100%The equivalent intersecting area stayed at 1.2 cells while completion rose by 75 percentage points. Improved completion and lower allocated count can coexist, which is why record change and quality must be read separately.
Figure 2
The net total hides the movement
Only 177 cells rose, while 2,022 fell and 1,648 were unchanged. Large increases in a small number of cells offset many smaller declines.
Reading note. Gross movement is the sum of cell-level count changes. It is not a count of properties entering or leaving the market.
Largest cell increases
- London+2,469
- Wembley+2,115
- Edinburgh+456
- Brighton+291
- Hayle+255
Largest cell decreases
- Enfield Town-84
- Hounslow-56
- Brent-45
- Bexley-39
- Croydon-35
Method
A reproducible release comparison
- Match the 3,847 fixed public grid cells by their published geometry and compare current record representations cell by cell.
- Keep the current cross-section separate from the cumulative dated market-state count.
- Allocate intersecting cells to Bristol using a 7 by 7 spatial sample and proportional integer apportionment across local-authority boundaries.
- Calculate record change and collection completion independently.
- Publish the comparison with its interval, data extract and interpretation limits.
Each 0.15 by 0.09 degree observation cell is sampled at 49 evenly spaced points. The share falling within each local-authority geometry sets the allocation weight. Whole records are assigned by the largest-remainder method so the national total is preserved.
Interpretation boundary
What this release cannot prove
- 01
The two-day interval is suitable for release monitoring, not market trend estimation.
- 02
Observed records are not deduplicated homes and can change as location and identity evidence improves.
- 03
Local-authority totals are weighted allocations from fixed cells, not exact property counts.
- 04
A record appearing or disappearing from a release does not prove that a home entered or left the market.
Data and citation
Inspect the evidence
Upgraded Intelligence (2026), Bristol Observation Change Note: 18 to 20 August 2026, version 1.0, UI-CMP-BRI-2026-01.
Questions, corrections or requests for an accessible extract can be sent to hello@theupgradeauthority.co.uk.