We Tested Our Street Forecasts Against 11 Years of Real Outcomes
353,295 streets across Australia. Our top picks outperformed the national market by 3.0 percentage points per year and added $141,000 on a $1M property over 4 years.
Luke Metcalfe · Microburbs Research · March 2026Suburb Forecasts Hide What Actually Matters
Two streets in the same suburb with the same postcode and the same school catchment can have completely different growth trajectories. The best and worst streets in a suburb typically perform 3–5 times differently over a 4-year period.
Suburb-level forecasts smooth over this variation. They tell you the average. But you are not buying the suburb. You are buying one specific address.
We built a street-level forecasting system trained on 17.4 million house sales from 2005 to 2025, covering 353,295 streets across 14,069 suburbs. We then backtested it against 11 years of real outcomes using rolling quarterly walk-forward validation — no future data was ever used in the training set.
The Backtest Results
We forecast relative performance for 353,295 streets across Australia, then measured actual outcomes over the following four years. The test period included COVID, the 2022 rate hikes, and the subsequent recovery.
Our Top Picks Outperformed Our Bottom Picks by $265,000
On a $1M property, our top 10% of street picks added $141,000 more equity than the market average over 4 years, and $265,000 more than our bottom picks (2012-2023 backtest). Our top picks outperformed the market by 3.0 percentage points per year.
The bottom 10% of picks underperformed the market by 2.9 percentage points per year. Streets we flagged as likely underperformers significantly lagged the broader market. Picking the right street added $141,000 to a $1M property over 4 years versus the market average.
Backtested Streets: What We Predicted vs What Happened
We forecast these streets to outperform. Here is what actually happened over the following four years.
Castlecrag, NSW
Same suburb. Same postcode. Three streets with very different outcomes.
| Street | Forecast | Actual Result |
|---|---|---|
| The Barricade | Strong outperformance | +5.9pp/year |
| Edinburgh Road | Around average | +0.2pp/year |
| Linden Way | Below average | -1.5pp/year |
The Barricade and Linden Way are 1.2km apart in the same suburb. On a $2.6M Castlecrag property, that 7.4pp/year difference represents over $190,000 in annual equity.
McMahons Point, NSW
McMahons Point, NSW
Backtest period: 4-year horizon. Suburb average outperformance vs national market: +1.3 pp/yr
| Street | We Forecast | Actual Result | Outcome |
|---|---|---|---|
| Munro St | Strong outperformance | +10.9 pp/yr | Beat market by 10.9% annually |
| East Crescent St | Above average | +7.3 pp/yr | Beat market by 7.3% annually |
Munro St was priced below the suburb average at forecast time. Four years later it had nearly tripled the market return. The lower entry price combined with higher growth meant significantly better returns per dollar invested.
Mosman, NSW
Mosman, NSW
Backtest period: 4-year horizon across 200+ streets
| Street | We Forecast | Actual Result | Outcome |
|---|---|---|---|
| Warringah Rd | Moderate outperformance | +1.9 pp/yr | Beat market by 1.9% annually |
| Military Rd | Average performance | -0.3 pp/yr | Tracked the market |
Even in a well-established suburb like Mosman, the streets we flagged as outperformers delivered measurably better returns. The difference compounds significantly over a typical holding period.
What Makes Some Streets Grow Faster
Four patterns showed up consistently across 20 years of sales data.
Priced Below the Suburb Average
Streets cheaper than their suburb average tend to catch up over time. This was the strongest predictor in our backtest.
Signs of Upgrading
When recent sales showed larger houses and renovations, prices followed. New builds signalled confidence.
Suburb Momentum
Streets in suburbs with strong recent growth tended to carry that momentum forward.
Deep Sales History
More past transactions meant more reliable forecasts. We tell you which streets have strong data and which do not.
Streets priced below the suburb median catch up. Mean reversion is the single strongest signal. Streets trading at a discount to their suburb tend to close that gap over 2–4 years. The effect is most reliable for streets that are 10–25% below the median.
Housing stock upgrades signal demand. Streets where bedrooms, bathrooms, or floor area have increased via renovation or extension outperform streets with static stock. Renovation activity is a leading indicator.
Transaction depth determines confidence. Streets with 100+ historical sales produce forecasts with 3–10x lower error than streets with under 20 transactions. Low data = low confidence = wider range.
Not All Forecasts Are Equal
Some streets have dozens of recent sales. Others have had two transactions in 10 years. We show you the difference.
Every forecast comes with a confidence rating. You can filter to see only streets where we have high confidence, or explore all of them.
How This Helps You
If you are buying a home, this shows you which streets in your target suburb have historically outperformed. Not every street in a good suburb is a good buy.
If you are an investor, picking the right street added $141,000 to a $1M property over 4 years versus the market average (2012-2023 backtest). That compounds to a large difference over a typical holding period.
If you are a buyers' agent, this gives you street-level data your competitors do not have. It turns a suburb recommendation into a street recommendation, backed by tested results.
See Your Street's Forecast
Street-level forecasts are available for 353,295 streets across Australia. High-confidence streets show predicted outperformance vs their suburb median.
Check a StreetRead the Research