Price Growth vs Population Growth — Why Population Data Failed Our Suburb Backtest

The case for population growth as an investment signal isn’t unreasonable on paper. If a suburb sits near a new employment hub, a rail extension, or a growing university precinct, the argument goes that demand for housing will increase ahead of supply. Prices follow demand.

At the national and state scale, there is a genuine long-run relationship between population growth and dwelling values. Post-2020 interstate migration did correlate with price growth in Queensland and Western Australia at a broad level. The thesis has enough surface truth to keep appearing in research reports and property podcasts.

The problem is that property investment doesn’t happen at the national level. It happens at the suburb level. And at that resolution, population growth falls apart as a signal.

When we built our first suburb-scoring formula, we included demographic demand pressure as an explicit component. It sat alongside days on market, vacancy rate, vendor discount, infrastructure spend, and new supply in a seven-factor model.

This formula aimed to predictbooms before they started. The theory was that leading macro indicators — population inflows, infrastructure commitments, tightening supply — could identify a boom 12–18 months out and let investors position early.

A coin flip with six extra inputs. The formula’s predictions were barely more reliable than chance. Population growth, infrastructure spend, and building approvals — all three of the macro thesis inputs — contributed near-zero signal at the suburb level.

Vendor discount data barely exists at suburb level from free sources. Infrastructure figures are state-level at best. And population growth, the most frequently cited signal, was too coarse and too slow to tell us anything useful about which specific suburb would outperform in the next 12–24 months.

The failure comes down to three mismatches between how population data exists and how suburb-level price growth actually works.

The breakthrough in our formula development wasn’t finding a better macro predictor. It was abandoning prediction entirely and shifting to detection.

Instead of asking “will this suburb boom?” the question became “is this suburb booming right now?” It sounds like a retreat — surely the whole point is to be early? But booms are multi-year events. Catching one 6–12 months after it starts still captures 60–85% of total gains, with far greater confidence than any prediction formula we built.

The current formula — v2.3 — has five components. Every one of them measures what’s happening at street level in a suburb right now:

Every one of these signals can be checked on realestate.com.au, Domain, SQM Research, or a CoreLogic-backed source. Population growth projections are not in the formula. Infrastructure pipelines are not in the formula. Building approvals are not in the formula.

85.7% accuracy versus 55%. Zero false positives. A 20.2-point separation gap between suburbs that genuinely boomed and those that didn’t — meaning the formula doesn’t just get the right answer. It gets it with conviction.

Detection tells you whethera suburb is booming. But which booming suburb outperforms the others? That’s a harder question — and most of the metrics investors use to answer it don’t survive scrutiny.

We tested growth rate, momentum, acceleration, and past outperformance as ranking signals. When we stripped out the broader market tide — measuring each suburb’s growth against the median growth across all comparable suburbs in the same period — almost everything failed to discriminate. Past outperformers tended to underperform going forward. Momentum within a growth phase told you the tide was rising, not which suburb would rise furthest.

One signal survived.

The mechanics make sense. When a growth cycle builds, buyer demand concentrates where the price gap between a suburb and the city median is widest. Affordability headroom isn’t just a value filter — it’s the mechanism that drives demand outward as inner-ring prices climb.

Population growth tells you something broad about a council area. Affordability headroom tells you something specific and checkable about a single suburb — something with a direct connection to where buyers will move next as the city median rises.

The detection formula doesn’t score every suburb. It first applies four hard filters that a suburb must clear before it gets scored at all. These thresholds came directly from the backtest data — every boom in the dataset passed all four.

The price cap is the most important of the four. It reflects the affordability finding directly: above $800K median, the headroom signal degrades sharply. The formula is built for suburbs priced below the capital city median, which is where every boom in the dataset started.

Boom scores run 0–100 after the hard filters are cleared: 80+ is Boom, 65–79 is Early Boom, 50–64 is Warming, below 50 is no boom detected. The 20.2-point separation gap means the boundary between real booms and false signals is wide enough to be actionable.

The formula’s output — Strong Buy, Buy, Watch, Pass — was tested against actual outcomes across 12,360 postcode-months in a walk-forward backtest. No lookahead. Excess return is each suburb’s 12-month growth minus the market median growth in the same period.

Walk-forward backtest, 12,360 postcode-months. No lookahead. Excess return = suburb 12-month growth minus market median growth in the same period. Full methodology →

Perfectly monotonic. Strong Buy outperforms Buy outperforms Watch outperforms Pass. The 13.9-point spread between the top and bottom tier — +7.5pp versus −6.4pp — is the gap between a suburb that beats the market by nearly 8 percentage points and one that trails it by more than 6.

Compare that to the coin flip that population growth delivered. The difference isn’t a marginal improvement on the same approach. It’s a completely different kind of signal — one built from what buyers are actually doing in the market right now, not from what demographers projected years ago.

You don’t need our formula. The two signals that actually work are public knowledge and freely available.

Population growth projections won’t be on that checklist. Infrastructure pipelines won’t be on it either. The two signals that survived 78 suburbs of backtesting are both measurable, both freely accessible, and both available for any suburb in Australia today.

The difficult part is doing it across hundreds of suburbs — filtering to your budget band, updating fortnightly as market conditions shift, and catching booms in the 6–12 months after they start, when 60–85% of the gains are still ahead. That’s what BoomAU automates.

Full backtest methodology, the 78-suburb validation, and the walk-forward tier discrimination results are published on our proof page. No login, no email required. Check the data yourself.