Property Investment Data AnalysisThe Formula Behind 85.7% Suburb Boom Accuracy

Most suburb-scoring approaches try to rank everything. We do the opposite. Four conditions must all be met before a suburb enters the formula at all. If any one fails, the suburb isn’t scored — no matter how appealing it looks on paper.

The $800K cap isn’t arbitrary. It’s where the data stopped finding booms. Above that threshold, suburbs are typically at or above their capital city median price — and the backtest found that affordability headroom is the strongest cross-suburb growth signal. Expensive suburbs have already consumed the gap that drives outperformance.

The growth and tightness filters work together to screen for coherent market behaviour. A suburb showing 8% annual growth but 60-day DOM and 3% vacancy may be producing distorted figures — perhaps from a handful of high-value sales or a single development skewing the median. A suburb that clears all four filters is showing consistent demand pressure across multiple measures simultaneously.

Once a suburb passes the hard filters, it’s scored across five components. Each carries a weight determined through backtesting — not intuition or convention. The weights reflect how strongly each factor actually correlated with boom outcomes in the data.

The weights tell you what the data found most important. Momentum and Growth Strength together account for 55% of the score — the formula is primarily measuring whether a suburb is in active, accelerating growth. Tightness carries 20%, confirming that buyer demand is genuinely outpacing available supply. Sustainability and Headroom provide context: is the growth supported by real rental income, and is there room for further price appreciation?

The total score maps to four detection tiers:

The detection formula was validated on 78 Australian suburbs — 28 that boomed, 50 controls that didn’t. Accuracy: 85.7%. False positives: 0%. The separation gap between genuine booms and false signals was 20.2 points. That gap matters as much as the accuracy figure — a formula that scores real booms at 82 and non-booms at 60 is far more useful for decision-making than one that puts them at 55 and 52.

Detection tells you whether a suburb is booming. The next question is whether the formula’s tier labels actually predict which booming suburbs outperform the market. For that, we ran a separate walk-forward backtest across 12,360 postcode-months from 2012 to 2026 — no lookahead, no data leakage, each score made using only data available at the time.

Walk-forward backtest, 12,360 postcode-months, 2012–2026. No lookahead. Excess return = suburb 12-month growth minus market median growth. Full methodology →

The tier discrimination is perfectly monotonic. Every step down in tier produces a worse outcome, in exactly the expected direction. The spread between Strong Buy and Pass is 13.9 percentage points — that’s the real cost of suburb selection decisions, measured across thousands of actual postcode-months of Australian market data.

Pass-tier suburbs beat the market only 28% of the time. If the market is your benchmark, being in the wrong suburb isn’t neutral — it’s an active drag on returns. Strong Buy suburbs beat the market 71% of the time and added 7.5 percentage points of excess return. Same country, same asset class, 13.9 points of spread. The only variable is suburb selection.

Here is the insight most property data analysis gets wrong — and the one that required the most care to validate correctly.

When the property market runs hot nationally, almost every suburb rises. Growth figures look strong across the board. Any metric that correlates with national boom periods will also appear to correlate with strong suburb returns — because both are moving with the tide, not because the metric actually picks the right suburb.

This shows up dramatically in boom era analysis. The median suburb boom in our dataset was 1.3% above market before 2015. After 2020, the median was 16.2% above market. If you built a model that simply identified “boom years” and bet on the market during those years, it would look extraordinarily accurate — because boom years pull all suburbs up. The model would be ranking time periods, not suburbs.

We tested this directly. We built a ranking model and measured its performance across the full dataset — pooling all periods together. The result looked strong. We then measured the same model’s ability to pick the right suburb within any given month, cancelling the market tide. The pooled figure was a statistical illusion. Within any given period, the model offered no useful guidance about which suburb would outperform.

Mean reversion is particularly important to understand. The suburbs that outperformed strongly in one period tend to revert toward the market median in the next. Buying based on recent outperformance — chasing the story — consistently underperforms in the data. 5-year momentum failed as a forward-looking signal for exactly this reason.

One signal consistently predicts which suburb outperforms within any given period, after cancelling national market movement.

Every single boom in the 78-suburb backtest was led by a suburb priced well below its capital city median. Not some of them — all of them. Above the city median, booms in the dataset don’t just become rarer; they become essentially absent.

Combine this with boom timing detection — is the suburb currently in a detected boom, and how much of its affordability gap has already been consumed? — and you have a two-axis framework: what to buy (affordability) and when to buy it(timing). Detection catches booms 6–12 months after they start, and still captures 60–85% of total gains. The goal isn’t to call the top tick — it’s to enter early enough to capture most of the run.

One of the most important data analysis lessons has nothing to do with which metric you choose — it’s about whether the data behind that metric is actually reliable.

Days on market is a genuine signal. But it’s only meaningful when a suburb has enough transaction volume to produce a reliable median. In a suburb that sells fewer than 30 homes per year, a single unusually fast sale pulls the DOM figure to 10 days. A single stubborn listing pushes it to 150. The published median is an echo of a handful of transactions, not a market.

This matters especially for regional suburbs, which often have the strongest affordability headroom but the thinnest transaction volumes. A regional suburb priced at $350K with 5% growth and 20-day DOM might look like a Strong Buy signal — but if it sold 8 homes last year, the DOM figure is being driven by those 8 sales. One slow vendor and the whole picture changes.

The correct approach is to check annual sales volume before trusting tightness metrics. YIP publishes annual sales counts at suburb level alongside DOM. Cross-reference them. A suburb with 20-day DOM and 9 annual sales is a very different proposition from one with 20-day DOM and 85 annual sales.

You don’t need a formula to apply these principles. Every signal that matters is available for free.

The hard part isn’t knowing the framework. It’s applying it consistently across 8,417 Australian suburbs every fortnight, flagging new boom signals within weeks of their start, and filtering results by budget band so you’re only seeing suburbs you can actually buy in. That’s what BoomAU automates.

The full backtest methodology — all 78 detection suburbs, the walk-forward tier results, and the complete formula weighting rationale — is published on our proof page. No email required. Check the maths yourself.

Of the 8,417 suburbs in our national sweep, 393 currently pass all four hard filters and receive a scored tier label fortnightly. That’s not a small list — it’s a targeted one.

The budget bands reflect a structural property of the backtest: every boom in the dataset was led by suburbs priced below the city median. A $400K budget isn’t a constraint on your options — it’s a concentration in exactly the affordability band where the backtested growth signal is strongest. Budget-limited buyers aren’t at a disadvantage here. They’re searching in the right territory by default.