Council Cycle Cohort Trends

How councils that vote in the same years move together, cycle over cycle.

Assign each council to one cohort: Councils are grouped by which years they hold elections, so a cohort's members always vote together.

Aggregate council by council, then across the group: Party vote shares are computed per council per cycle, then combined into a cohort figure.

Keep comparisons clean: Only scheduled whole-council and by-thirds rounds count; by-elections are excluded so they can't distort the trend.

Withhold thin cycles: Where fewer than 30% of a cohort's councils have results for a cycle, no figure is shown rather than a misleading one.

Flag the caveats on the card: Reorganisation bridging, low confidence, small groups and mixed voting systems are surfaced where they apply.

All insights

Insight 2

Council Cycle Cohort Trends

How councils that vote in the same years move together, cycle over cycle.

Live for 16 council groups, 1995–2025

Councils that always go to the polls in the same years form a true like-for-like group. Tracking those groups from 1995 to today turns scattered local results into the only honest long-run read on local political movement.

What the report does

Group councils by their election calendar — county cycle, district whole-council, district by-thirds, unitaries, London boroughs — and compare each group's party vote share from one of its elections to the next. Because every council in a group votes in the same years, each cycle is a genuine like-for-like comparison.

How it's built

1
Assign each council to one cohort
Councils are grouped by which years they hold elections, so a cohort's members always vote together.
2
Aggregate council by council, then across the group
Party vote shares are computed per council per cycle, then combined into a cohort figure.
3
Keep comparisons clean
Only scheduled whole-council and by-thirds rounds count; by-elections are excluded so they can't distort the trend.
4
Withhold thin cycles
Where fewer than 30% of a cohort's councils have results for a cycle, no figure is shown rather than a misleading one.
5
Flag the caveats on the card
Reorganisation bridging, low confidence, small groups and mixed voting systems are surfaced where they apply.

How to read it (and how not to)

• This is a local-election trend, not a general election prediction — people vote differently in local and national contests.
• Northern Ireland is excluded: council results there are not yet in the dataset at the level this report needs.
• 233 authorities that did not cluster cleanly — mostly councils abolished before 2013 and small or non-periodic cases — are excluded and named, not silently dropped.
• Cohort membership is treated as fixed over time; hints that a council has changed cycle are not yet reliable enough to use.
• Scottish and Welsh councils use different voting systems; their vote shares are comparable but not identical in meaning to English ones.

What it reads

• Ballot-box results for every scheduled local election round since 1995
• A cycle-cohort map assigning each council to the group that shares its election calendar
• Predecessor-council bridging where boundaries were reorganised

What you get back

• Party vote share and swing for each cohort, cycle by cycle from 1995
• Strongest-party, most-marginal-council and direction-of-travel summaries
• Council-by-council drill-down with the live results behind every figure
• Trend, composition and council-swing charts with plain-English reading guides

Readiness

Cycle-cohort assignments
382 councils grouped into 16 cohorts.
Aggregated cohort results
271 cycle rows computed, covering 1995–2025.

Council Cycle Cohort Trends

How councils that vote in the same years move together, cycle over cycle.

Assign each council to one cohort: Councils are grouped by which years they hold elections, so a cohort's members always vote together.

Aggregate council by council, then across the group: Party vote shares are computed per council per cycle, then combined into a cohort figure.

Keep comparisons clean: Only scheduled whole-council and by-thirds rounds count; by-elections are excluded so they can't distort the trend.

Withhold thin cycles: Where fewer than 30% of a cohort's councils have results for a cycle, no figure is shown rather than a misleading one.

Flag the caveats on the card: Reorganisation bridging, low confidence, small groups and mixed voting systems are surfaced where they apply.

What the report does

How it's built

Assign each council to one cohort

Councils are grouped by which years they hold elections, so a cohort's members always vote together.

Aggregate council by council, then across the group

Party vote shares are computed per council per cycle, then combined into a cohort figure.

Keep comparisons clean

Only scheduled whole-council and by-thirds rounds count; by-elections are excluded so they can't distort the trend.

Withhold thin cycles

Where fewer than 30% of a cohort's councils have results for a cycle, no figure is shown rather than a misleading one.

Flag the caveats on the card

Reorganisation bridging, low confidence, small groups and mixed voting systems are surfaced where they apply.

How to read it (and how not to)

• This is a local-election trend, not a general election prediction — people vote differently in local and national contests.

• Northern Ireland is excluded: council results there are not yet in the dataset at the level this report needs.

• 233 authorities that did not cluster cleanly — mostly councils abolished before 2013 and small or non-periodic cases — are excluded and named, not silently dropped.

• Cohort membership is treated as fixed over time; hints that a council has changed cycle are not yet reliable enough to use.

• Scottish and Welsh councils use different voting systems; their vote shares are comparable but not identical in meaning to English ones.

What you get back

• Party vote share and swing for each cohort, cycle by cycle from 1995

• Strongest-party, most-marginal-council and direction-of-travel summaries

• Council-by-council drill-down with the live results behind every figure

• Trend, composition and council-swing charts with plain-English reading guides

Council Cycle Cohort Trends

Council Cycle Cohort Trends

What the report does

How it's built

How to read it (and how not to)

What it reads

What you get back

Readiness

Council Cycle Cohort Trends

Council Cycle Cohort Trends

What the report does

How it's built

How to read it (and how not to)

What it reads

What you get back

Readiness

English county cycle (2013/2017/2021/2025) — 2025