Mean Reversion and Property Prices in Australia — Why Past Winners Tend to Underperform

Mean reversion isn’t a theory. It’s a measured outcome. When you look at excess returns — how much each suburb grew above the market median in a given period — suburbs that outperformed significantly in one period tend to revert toward the median in the next. Not always. Not every suburb. But consistently enough that 5-year momentum failed to beat a coin flip in backtesting.

The mechanism is straightforward. When a suburb runs hard — say 25% annual growth when the city is averaging 8% — buyers start pricing in that trajectory. The suburb closes the gap between its median price and the city median. Once that gap is closed, there’s no longer a price differential to compress. The suburb stops outperforming. Sometimes it underperforms as buyers who stretched to get in hold an expensive asset with a narrowing buyer pool.

This is not the same as prices falling. In an overall rising market, a mean-reverting suburb might still produce 5% growth — it just lags behind a market median of 8%. After years of being called a hotspot, it quietly underdelivers. The investor who bought based on its track record earns below-market returns without ever seeing a nominal loss.

Before diving into why momentum fails, there’s a related problem with relying on price history: boom size is not consistent across time. It depends on the era.

A suburb that “proved itself” with a boom in 2013 delivered a median excess return of 1.3% above the market. The exact same detection signal fired in 2021 caught a 16.2% event. The suburb’s historical track record tells you almost nothing about what the next boom will produce. What matters is the macroeconomic era.

This explains why so many investors buy “cyclical” suburbs expecting a repeat of past booms. They’re not wrong that the suburb has boomed before. They’re wrong that history gives them useful information about magnitude or timing.

A suburb that boomed in 2018 and then again in 2022 — two events with a 12× difference in median boom size — was not performing consistently. The era was doing the work. The suburb happened to be in the right part of the affordability spectrum both times.

We put 5-year momentum directly into the scoring formula and backtested it across 78 suburbs and 12,360 postcode-months. The result: 55% accuracy. Statistically indistinguishable from a coin flip.

The failure has two layers.

First, the tide effect. When national credit conditions loosen or population inflows accelerate, almost every suburb grows. A suburb with strong 5-year momentum has that momentum largely because it rode the same rising tide as everything else. The momentum signal mostly reflects era, not suburb quality. Cancel the tide by measuring each suburb’s excess return above the period median, and momentum’s predictive power collapses.

Second, affordability consumption. The suburbs with the strongest 5-year momentum have, by definition, closed the most affordability gap against the city median. They’ve already done the hard work. New buyers arriving after five strong years are entering at the tail of the opportunity, not the start of it.

The harder version of this result: growth phase does not predict relative outperformance. Whether a suburb is accelerating, plateauing, or in an established upswing doesn’t tell you which suburb will beat the market median over the next 12 months. Within any given period, momentum-ranked suburbs performed no better than random.

Infrastructure spending, population growth, and building approvals were tested separately. All three failed for the same underlying reason: they measure forces that operate over decades and across corridors, not forces that determine which suburb beats its neighbours over an investment-relevant horizon.

Once momentum failed, the question became: if you can’t predict which suburbs will boom, can you at least catch booms once they’ve started?

Detection sounds like settling. If you’re not in before the boom starts, haven’t you missed the point? This framing misunderstands how booms actually work. They’re not brief events. Genuine suburb booms in Australia are multi-year runs. Getting in 6–12 months after a boom starts doesn’t mean you’re too late.

A detection signal that fires 6–12 months into a boom still positions you to capture 60–85% of total gains. With 85.7% accuracy and 0% false positives across 78 suburbs — 28 that boomed and 50 that didn’t. Compare that to 55% prediction accuracy. The trade-off is clear: you give up the very first months of a boom in exchange for dramatically higher confidence that you’re actually in one.

Detection works where prediction fails because it measures what’s actually happening, not what a model hopes will happen. Buyers are already moving. Stock is genuinely disappearing. Days on market is actually contracting. These are measurements of present-tense market behaviour, not extrapolations of historical trends.

The detection formula has five components. Each one scores something that is measurable in real time from free public sources:

Cancelling the tide is the hardest part of suburb analysis. Most signals that look promising across suburbs turn out to be measuring era. We tested every metric available — growth rate, acceleration, repeat-boom history, days on market, vacancy, yield, rental trend — against tide-cancelled excess returns. Which features predicted a suburb outperforming its peers in the same period?

One signal survived.

This is also the direct mechanism behind mean reversion. A suburb that has already run hard and closed its affordability gap has, by definition, consumed its headroom. There’s no longer a price differential pulling buyers across from more expensive suburbs. The primary demand driver is gone.

Conversely, a suburb priced well below the city median still has that gap to close. Buyers priced out of more expensive options look here next. Demand concentrates. Prices move to close the discount. Every single boom in the 78-suburb backtest was led by a suburb priced well below its city median. Not some. All of them.

The affordability signal also connects mean reversion to a workable strategy. Instead of chasing history — which the backtesting shows doesn’t work — you look for the gap that hasn’t closed yet. Below-median suburbs are the places mean reversion hasn’t happened to yet. They’re the next cohort to revert upward toward the city median.

Combining the detection formula with affordability headroom produces a suburb tier. The walk-forward backtest across 12,360 postcode-months showed how each tier performed against the market median.

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

Perfectly monotonic. Strong Buy outperforms. Buy slightly outperforms. Watch slightly underperforms. Pass significantly underperforms. The 13.9 percentage point spread between the top and bottom tier — same asset class, same Australian market — is the direct financial cost of buying a mean-reverting suburb with exhausted headroom instead of one with room to run.

Notice the Pass tier specifically: −6.4pp excess return, with only 28% of observations beating the market. Seven in ten Pass-tier suburbs underperformed. That’s not a run of bad luck. That’s mean reversion working exactly as predicted: exhausted affordability, past outperformers, no remaining gap to close.

The mean reversion insight leads directly to a two-step check you can run with free data before looking at any specific property.

The full formula scores five components across 393 suburbs and updates fortnightly. But the two steps above capture the core of it: drop the suburbs with exhausted headroom, find the ones with active booms and room to run. Everything else — momentum, history, infrastructure narratives — failed backtesting.

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