--- title: "The Scale Illusion: Population Size and the APN Bedrock Score Across Australia’s 50 Most Populous LGAs" url: https://australianproperty.network/analysis/demographic-analysis/demographic-trend-analysis-demographic-analysis/the-scale-illusion-population-size-and-the-apn-bedrock-score-across-australias-50-most-populous-lgas/ date: 2026-07-13 modified: 2026-07-13 author: "APN LGA Intelligence" description: "Brisbane (1.375M people) and Knox (163,820) score identically on APN Bedrock composite — 57.87 each. Across the 50 most populous LGAs, population size explains just 0.6 per cent of the variance in Bedrock composite (r = 0.077); density fares little better (r = -0.165). This analysis rules out scale as an explanation before other factors are considered." categories: - "Demographic Trend Analysis" tags: - "24110" - "APN Bedrock" - "Brisbane" - "City of Knox" - "Fairfield" - "LGA Intelligence" - "population density" - "Sutherland" - "Sydney" - "Toowoomba" word_count: 929 --- # The Scale Illusion: Population Size and the APN Bedrock Score Across Australia’s 50 Most Populous LGAs ### The Scale Illusion: Population Size and the APN Bedrock Score Across Australia’s 50 Most Populous LGAs Brisbane holds 1,375,301 people. Knox holds 163,820 — an 8.4-times difference, the largest population gap between any two LGAs in this sample. Their APN Bedrock composite scores are 57.87 and 57.87. Identical, to two decimal places. Across the 50 most populous LGAs, population size explains essentially none of the variation in Bedrock composite: r = 0.077, r² = 0.006. This analysis sets that non-relationship out in full, and rules out population and density as explanations before other factors are considered. #### The Non-Relationship Regressing Bedrock composite on population (Estimated Resident Population, 2025) across the 50 most populous LGAs produces a trend line that is, for practical purposes, flat: across the full range of the sample — from Knox’s 163,820 to Brisbane’s 1,375,301 — the fitted line moves by roughly 5 points. Population accounts for 0.6 per cent of the variance in Bedrock composite. The scatter below shows the shape of that non-relationship directly: LGAs of every size sit at every point on the vertical axis. Population vs APN Bedrock Composite, All 50 LGAs 20 35 50 65 80 200k 400k 800k 1.4M Population (ERP 2025, log scale) APN Bedrock composite score Brisbane / Knox / Sutherland / Fairfield Each point is one of the 50 most populous LGAs (population axis log-scaled). Gold points are Brisbane, Knox, Sutherland and Fairfield, referenced in the text. Dashed line is the linear trend (r = 0.077, r² = 0.006). #### Same Size, Opposite Outcomes The clearer test of the non-relationship is not the extremes but the middle. Sutherland (241,327 people) and Fairfield (213,677 people) sit within 13 per cent of each other in population — about as close to “the same size” as any pair in this sample gets. Sutherland posts the highest Bedrock composite score of the 50 LGAs: 73.27. Fairfield posts the lowest: 26.74. A 46.5-point gap, the widest spread anywhere in the data, between two LGAs that are effectively the same size. Whatever separates them, it is not population. #### Density Doesn’t Rescue the Story Population density is a marginally cleaner signal than raw population count, but not by enough to matter. Density correlates with Bedrock composite at r = −0.165 — weak, and in the opposite direction to what a simple “density supports value” thesis would predict. Sydney, the densest LGA in the sample at 9,064 persons per km², posts a middling Bedrock composite of 46.34. Toowoomba, the sparsest at 14.4 persons per km² — roughly 600 times less dense — scores higher, at 53.13. Neither density nor population count offers a usable explanation for Bedrock composite on its own. Population Density vs APN Bedrock Composite, All 50 LGAs 20 35 50 65 80 10 100 1,000 10,000 Population density, persons per km² (log scale) APN Bedrock composite score Sydney / Toowoomba / Sutherland / Fairfield Each point is one of the 50 most populous LGAs (density axis log-scaled). Gold points are Sydney, Toowoomba, Sutherland and Fairfield. Dashed line is the linear trend (r = −0.165). #### Population Rank vs Bedrock Rank Ranking all 50 LGAs by population and separately by Bedrock composite makes the scrambling visible at a glance. Brisbane is 1st by population and 8th by Bedrock — a 7-place drop. Knox is 50th by population and tied for 8th by Bedrock — a 42-place rise. Sutherland climbs 21 places, from 22nd by population to 1st by Bedrock; Fairfield falls 22 places, from 28th to 50th. None of the five highest Bedrock scores in the sample belong to a top-10 LGA by population. | Pop. Rank | LGA | State | Population (ERP 2025) | Bedrock Composite | Bedrock Rank | Rank Shift | | --------- | --- | ----- | --------------------- | ----------------- | ------------ | ---------- | | 1 | Brisbane | Queensland | 1,375,301 | 57.87 | 8 | -7 | | 2 | Gold Coast | Queensland | 691,230 | 51.13 | 19 | -17 | | 3 | Moreton Bay | Queensland | 532,445 | 54.22 | 15 | -12 | | 4 | Blacktown | New South Wales | 449,385 | 45.42 | 36 | -32 | | 5 | Casey | Victoria | 414,929 | 55.96 | 10 | -5 | | 6 | Logan | Queensland | 403,515 | 47.38 | 31 | -25 | | 7 | Canterbury-Bankstown | New South Wales | 389,687 | 35.72 | 47 | -40 | | 8 | Sunshine Coast | Queensland | 381,957 | 50.97 | 20 | -12 | | 9 | Central Coast (NSW) | New South Wales | 357,816 | 47.41 | 30 | -21 | | 10 | Wyndham | Victoria | 347,830 | 47.93 | 29 | -19 | | 11 | Greater Geelong | Victoria | 295,052 | 50.65 | 21 | -10 | | 12 | Parramatta | New South Wales | 279,014 | 46.53 | 34 | -22 | | 13 | Hume | Victoria | 278,885 | 39.66 | 45 | -32 | | 14 | Northern Beaches | New South Wales | 272,656 | 70.39 | 2 | +12 | | 15 | Ipswich | Queensland | 268,272 | 43.22 | 41 | -26 | | 16 | Liverpool | New South Wales | 261,231 | 40.00 | 44 | -28 | | 17 | Whittlesea | Victoria | 259,759 | 49.02 | 26 | -9 | | 18 | Cumberland | New South Wales | 256,906 | 27.40 | 49 | -31 | | 19 | Stirling | Western Australia | 254,821 | 55.45 | 12 | +7 | | 20 | Wanneroo | Western Australia | 246,147 | 53.99 | 16 | +4 | | 21 | Sydney | New South Wales | 241,797 | 46.34 | 35 | -14 | | 22 | Sutherland | New South Wales | 241,327 | 73.27 | 1 | +21 | | 23 | Penrith | New South Wales | 231,701 | 55.27 | 14 | +9 | | 24 | Melton | Victoria | 231,567 | 48.08 | 28 | -4 | | 25 | Lake Macquarie | New South Wales | 224,540 | 49.61 | 25 | 0 | | 26 | Wollongong | New South Wales | 224,327 | 48.17 | 27 | -1 | | 27 | The Hills | New South Wales | 222,675 | 70.28 | 3 | +24 | | 28 | Fairfield | New South Wales | 213,677 | 26.74 | 50 | -22 | | 29 | Monash | Victoria | 211,833 | 47.05 | 32 | -3 | | 30 | Townsville | Queensland | 206,260 | 46.69 | 33 | -3 | | 31 | Brimbank | Victoria | 198,181 | 38.84 | 46 | -15 | | 32 | Melbourne | Victoria | 194,481 | 40.89 | 43 | -11 | | 33 | Inner West | New South Wales | 193,125 | 55.58 | 11 | +22 | | 34 | Campbelltown (NSW) | New South Wales | 191,285 | 44.04 | 39 | -5 | | 35 | Merri-bek | Victoria | 189,108 | 50.03 | 22 | +13 | | 36 | Bayside (NSW) | New South Wales | 187,770 | 40.91 | 42 | -6 | | 37 | Swan | Western Australia | 187,090 | 52.79 | 18 | +19 | | 38 | Toowoomba | Queensland | 186,276 | 53.13 | 17 | +21 | | 39 | Whitehorse | Victoria | 185,256 | 49.70 | 24 | +15 | | 40 | Onkaparinga | South Australia | 184,827 | 49.92 | 23 | +17 | | 41 | Cairns | Queensland | 179,334 | 43.84 | 40 | +1 | | 42 | Newcastle | New South Wales | 178,935 | 45.21 | 37 | +5 | | 43 | Boroondara | Victoria | 178,601 | 61.18 | 6 | +37 | | 44 | Joondalup | Western Australia | 176,595 | 69.11 | 4 | +40 | | 45 | Redland | Queensland | 172,831 | 44.97 | 38 | +7 | | 46 | Mornington Peninsula | Victoria | 172,217 | 62.67 | 5 | +41 | | 47 | Greater Dandenong | Victoria | 168,684 | 33.82 | 48 | -1 | | 48 | Kingston (Vic.) | Victoria | 168,061 | 58.68 | 7 | +41 | | 49 | Georges River | New South Wales | 163,919 | 55.28 | 13 | +36 | | 50 | Knox | Victoria | 163,820 | 57.87 | 8 | +42 | #### What Actually Predicts Bedrock Ruling population and density out matters because it clears the way for the variables that do carry explanatory weight. A companion analysis of this same sample found IRSAD decile — socio-economic advantage — correlates with Bedrock composite at r = 0.615, and that housing typology (apartment share of dwelling stock) explains a further, independent share of the variance once advantage is controlled for (r = −0.496 against the IRSAD-adjusted residual). Scale was never a serious candidate; this analysis closes that door formally so it does not need reopening. #### Methodology Note Population figures are Estimated Resident Population (ERP) as at 30 June 2025. Population density is persons per square kilometre, calculated from ABS land area figures. APN Bedrock composite is drawn from the APN LGA Intelligence Routine as at 10 July 2026. Correlation coefficients and regression lines (population: Bedrock ≈ 48.703 + 0.00000397 × population; density: Bedrock ≈ 51.225 − 0.000903 × density) are calculated across the 50 most populous LGAs in Australia, compiled through the APN LGA Intelligence Routine — a queue ordered by population, not a random or representative sample. Findings here describe the sample analysed only and should not be read as representative of the full national distribution of Australia’s 547 Local Government Areas.