Suburb Scoring Tools Australia — What Separates a Backtested Methodology from a Guess

The First Question: Is It Backtested?

A suburb score without a backtest is an opinion with extra steps. Backtesting means taking historical suburb data, running the formula on it, and checking whether it correctly identified suburbs that actually boomed — versus suburbs that went nowhere. The formula has to prove itself against real outcomes, not just look logical on paper.

Why does this matter so much? Because many metrics that sound convincing don’t survive contact with real data. Infrastructure spending sounds like a logical boom driver. Population growth sounds like it should move prices. Building approvals sound like a supply signal. We tested all of them. They delivered 55% accuracy — a coin flip. The metrics that passed backtesting are different from the ones experts talk about at property meetups.

What Good Accuracy Actually Looks Like

Accuracy alone doesn’t tell the full story. A formula that correctly labels every suburb “not booming” would hit very high accuracy just by being conservative — and would never surface a genuine opportunity. The number that matters alongside accuracy is the false positive rate: how often did the formula flag a suburb as booming when it wasn’t?

There’s also the question of sample size. Testing on 10 suburbs proves nothing. A formula optimised on a small sample will look brilliant on that sample and fall apart everywhere else. BoomAU’s current formula was validated across 78 suburbs — 28 that genuinely boomed and 50 controls that didn’t.

The 20.2-point separation gap is the figure we pay most attention to. It measures the average score difference between suburbs that genuinely boomed and those that generated false signals. A tight gap means the formula is barely distinguishing real booms from noise. A 20-point gap means it’s identifying booms with conviction, not just getting lucky on edge cases.

Zero false positives matters for a different reason: it means the Strong Buy label can actually be trusted. If a formula flags dozens of suburbs as booming and half of them aren’t, the label is worthless. The value of a rating system lives and dies on the signal quality at the top tier.

The 5 Components of the Formula

BoomAU’s detection formula has five scored components. Each was chosen because it measures something real about market dynamics — buyer urgency, seller behaviour, rental demand, growth trajectory, or affordability position. The weights reflect how much each component contributed to separation between real booms and false signals in the backtest.

Momentum carries the most weight because price growth acceleration is the clearest signal that a boom is actively underway — not just that prices rose last year. A suburb can have strong annual growth but decelerating momentum, which is a very different situation from one where growth is still picking up speed.

Tightness is the combination of days on market and vacancy rate. When properties sell fast and rental vacancies are low, demand is outstripping supply at both ends — buyers and tenants are competing. That combination has consistently marked active boom conditions in the backtest data.

Headroom is weighted lowest in the scored components (0.10), but it plays a bigger role than that number suggests. It’s also the basis of the hard price filter. Every boom in the 78-suburb backtest was led by a suburb priced well below its capital city median. That finding was consistent enough that we made it a structural constraint, not just a scored variable.

Hard Filters: What Gets Excluded Before Scoring

Before a suburb gets scored at all, it must pass four hard filters. These aren’t soft preferences — they’re gates. Failing any one of them means the suburb doesn’t receive a score, regardless of what its other numbers look like.

Of the 8,417 suburbs in the national dataset, 393 currently pass all four filters and receive a score. That’s a meaningful filter — it concentrates attention on the suburbs where boom conditions are actually present, rather than applying a score to every suburb in the country and leaving investors to sort through noise.

The budget bands within the 393: 35 suburbs are under $400K, 149 are under $600K, and 204 are under $800K. The filter naturally concentrates on the price ranges where the affordability signal is strongest.

What the Tiers Show You

A suburb score is only useful if the tiers it produces actually correspond to different outcomes. If Strong Buy suburbs don’t outperform Pass suburbs in any meaningful way, the entire scoring exercise is cosmetic. The walk-forward backtest measured this directly across 12,360 postcode-months.

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

The result is perfectly monotonic: every tier step down corresponds to a worse outcome. Strong Buy outperforms the market by 7.5 percentage points and beats it 71% of the time. Pass underperforms by 6.4 percentage points with only 28% of suburbs beating the market. That’s a 13.9-point spread between the top and bottom tier — in the same market, the same year.

The monotonic separation is the validation that matters most. A scoring tool that produces Strong Buy and Pass labels is only useful if those labels point to genuinely different investment outcomes. These do.

Where the Data Comes From

A suburb scoring tool is only as reliable as its data sources. The key questions to ask: Are the sources independently reputable? Is the coverage national? And how fresh is the data?

BoomAU draws from four sources, all publicly available. This is deliberate — it means the underlying numbers can be checked against sources you already use.

One caveat worth knowing: thin-market suburbs — those with fewer than roughly 30 annual sales — produce unreliable days on market figures. When a suburb only transacts a handful of times per year, a single fast sale pulls the median DOM down to 10 days, and a single slow listing pushes it to 150. Below about 15 annual sales, the DOM figure is not meaningful. BoomAU’s filters implicitly screen most of these out because thin markets rarely pass the growth threshold, but it’s worth understanding the limitation when reading any suburb-level data.

The Signal That Matters Most: Affordability Headroom

The most important finding from backtesting wasn’t about the formula itself — it was about which single variable best predicted whether a suburb would outperform its peers after cancelling the broader market tide.

The answer is affordability headroom: how a suburb’s median price compares to the capital city median. Every boom in the 78-suburb backtest was led by suburbs priced well below their city median. Above 1.5 times the city median, suburbs consistently underperformed. Below the city median, they consistently outperformed. The effect holds across states, across market conditions, and across time periods in the dataset.

This is also why boom size is era-dependent in the data. Pre-2015, the median boom in the backtest produced 1.3% excess gains. Post-2020, the median was 16.2%. Affordability headroom acts differently depending on the rate at which the city median is moving. A suburb’s relative position matters more than its absolute price.

The timing dimension comes from boom detection: BoomAU measures how much of the affordability gap has already closed. Suburbs that are early in a detected boom — with less than 30% of the headroom consumed — capture the most remaining upside. Detection typically catches booms 6 to 12 months after they start, which still leaves 60–85% of total gains on the table. You don’t have to be first. You have to be early enough.

How to Evaluate Any Suburb Scoring Tool

The framework is simple. Any suburb scoring tool worth using should be able to answer these questions directly:

BoomAU’s full backtest methodology, the 78-suburb validation results, and the walk-forward tier discrimination data are published on the proof page. No sign-up required. The maths is there to check.