The companion page to Winners asked which Census variables predict where each party won. This page asks where each party gained, lost, or held. Of the 2,186 wards with both a declared 2026 winner and a known prior party, 1,191 changed hands — a 54% turnover. Labour shed 726 of its 1,152 prior seats; the Conservatives lost 277 of 517; Reform took 551 from a standing start; the Greens went from 66 wards to 321 (269 of those by flip). The structural signal: flipped-to-Reform tracks +0.48 with apprenticeships and −0.44 with degree-holding; flipped-to-Green inverts the polarity with +0.27 with under-29s, +0.21 with private renters, +0.20 with density — softer than the GM-only equivalents (+0.67 / +0.65 / +0.59) because Green gains outside Manchester / inner London sit in less extreme structural cores. The before/after cards in §3 show how each party's surviving coalition shifted demographically (Labour's 426 remaining wards skew slightly more graduate / SOC 1-3; the Greens' 52 held wards are denser and more renter-heavy than their 2022 footprint). At GB scale every main-party flip sample is comfortably above the noise floor — Conservative 120 gains, LibDem 91, Labour 75, Independent 63, Other 22 — so every correlation card is readable.
Read each column as: "if a ward flipped to this party, how does each Census variable correlate with that flip?" Positive r in green, negative in red, opacity scaled to |r|. Cells with — are below the n=3 minimum for Pearson.
Each card shows how a party's structural footprint shifted between the prior and 2026 elections. Reform has no 2022 baseline (Reform UK won zero wards across the GB councils that elected in 2022), so its card shows 2026 means only. Other spans a long tail of small local-party labels with both 2022 and 2026 entries (residents' associations, Workers Party, single-ward minor parties), so its 2022 ↔ 2026 comparison mixes heterogeneous footprints and the Δ should be read cautiously. Δ for percentages is in percentage points; for density it's residents/km²; for age it's years. Bold positive Δ in green, bold negative in red — colour intensity scales with magnitude relative to typical inter-party spread.
"Prior" is the top-of-poll party the last time the seat was contested — usually 2022 (thirds and halves cycles), with 2021 or 2024 for the all-out councils on those cycles. flip = ward changed party hands; hold = same party as before; no prior = boundary change, no clean predecessor. ≈ next to a ward name marks an approximate prior mapping.