How to Analyse a Property Market in 2026

What Makes the Property Market Grow?

Jeremy's answer is unequivocal: supply and demand, the same fundamental economic principle that's held for centuries. He pushes back on commentators who add "sentiment" as a separate third factor, arguing sentiment is simply bundled within demand rather than a distinct force of its own. Asked how he'd respond to someone questioning his credentials given his engineering background rather than a formal economics or data science qualification, Jeremy points to the 15+ years he's spent building the DSR algorithm as his real-world grounding, motivated originally by having been misled by unreliable advice early in his own investing journey and deciding he needed to verify things independently before committing significant money.

Has Anything Changed in Property Market Analysis?

Jeremy notes more data sets and greater data availability now exist, alongside meaningful tax changes over the past couple of years, and that individual markets have naturally ebbed and flowed in that time. The underlying analytical approach itself, though, remains largely the same. He flags one clear shift: nearly every business in the space now markets itself as "data-driven," regardless of whether genuine rigorous analysis backs that claim. He shares an example of watching a podcast guest, who was heading a buyers agency and claiming to be a "data person," conduct what he describes as a hasty, low-quality five-minute analysis live on air, illustrating that claiming to use data isn't the same as using it well.

What's in DSR3?

Jeremy explains DSR (Demand to Supply Ratio) is a score out of 100 reflecting a suburb's overall investment potential, weighted heavily toward capital growth while also factoring in yield, combining a broad range of underlying metrics. DSR3, the latest version, improves on its two predecessors through more data, better analysis, and improved handling of anomalous or outlier metrics.

How Would Jeremy Analyse a Market Today?

Jeremy starts with the DSR itself, then shortlists via SA3s (Statistical Area Level 3, an ABS-defined grouping comprising dozens of suburbs) or local government areas. Checking at this broader level first means that if an area is genuinely hot, an even stronger individual suburb can usually be found within it; conversely, an isolated, one-off hot suburb with no supporting neighbours is treated as a reason to question the data rather than act on it. He also flags that the DSR3 is optimised for a roughly three-year outlook, while many investors have a much longer (5–20 year) time horizon. This creates real risk with a market like Perth, which has already had a strong, multi-year growth run: history shows a genuine surge can last anywhere from around three to six years, and a strong DSR score doesn't guarantee that run continues, since it simply reflects the current cycle's next few years rather than a longer-term outlook. Jeremy notes Perth previously went through an extended flat period after its earlier resources-driven boom, prompting the question of why a current boom wouldn't eventually face a similar multi-year catch-up period.

Damien highlights diversification across markets as one way to balance this risk, capturing a hot market's momentum while also gaining exposure elsewhere. Jeremy adds that for a genuinely long-term investor, a currently stagnant market like Melbourne might be the more appropriate pick, whereas Perth remains a good option specifically for an investor with a shorter, "get in, get out" strategy.

Can Investors Rely Too Heavily on One Metric?

Jeremy is firm: no single metric has ever proven free of exceptions across every period he's tested, which is exactly why combining many metrics into one overall score produces a more reliable result than relying on any single one, even the DSR itself isn't faultless.

Using population growth as a specific example (a claim he says he's "amazed" is still used in marketing), Jeremy reiterates a point covered in the show's Expert Busting Series: strong population growth reflects new housing supply becoming available, not a queue of demand waiting to move in, meaning it should be read as a forecast of supply, not demand.

On yield specifically, Damien notes there's nothing inherently wrong with prioritising cash flow during the accumulation phase, provided it isn't stretching household finances. The risk, both agree, is becoming obsessed with yield as the primary goal: Jeremy notes the effort required to find a market delivering even a modest 1% positive cash flow could, if redirected toward capital growth research instead, plausibly deliver 10% growth in the same year. Filtering specifically for high yield (6%, 7%, 8%+) mechanically narrows the pool of available markets and increases risk, though both agree a higher-yield, higher-risk property can still have a place later in a portfolio (a third or fourth purchase, for instance) for an investor willing to accept that trade-off.

How Much Weight Should Investors Give Recent Data?

Jeremy explains this depends entirely on the specific metric: a single month's raw value for some metrics is close to meaningless on its own, but the change in that same metric over 6 months, 5 years, or 10 years can carry real correlation to capital growth. Rather than investors needing to work out the right timeframe for each metric individually, all of these trend considerations are already built into the DSR itself.

When Should Investors Ignore a High DSR3 Score?

Jeremy identifies one specific case: an isolated, geographically standalone suburb whose score jumps by more than roughly half a dozen points in a single month, after previously sitting flat, which he treats as a likely data anomaly (perhaps a local marketing campaign, or a change in how transactions were being recorded) rather than a genuine signal. If that same suburb sits within a broader cluster of other suburbs also showing strength, his confidence in the data increases considerably.

How Do You Know If the Data Is Reliable?

Jeremy explains the statistical reliability score (out of 100) reflects factors like data volume, metric volatility, and whether different metrics or data sources contradict one another. His practical rule of thumb: rule out suburbs below a reliability score of around 60, relaxing that constraint only if budget constraints make it difficult to find any matching suburb otherwise. Damien adds that he personally pays close attention to dwelling count specifically, since a low count (say, under 500–1,000 dwellings) means very few monthly transactions, undermining the reliability of derived figures like median value or days on market calculated from just a handful of sales.

How Do You Avoid Buying the Wrong Property Within a Strong Suburb?

Jeremy clarifies the DSR operates at the suburb level (or aggregated further to postcode, LGA, and so on), but doesn't cover individual property selection. His primary asset-level consideration is land-to-asset ratio, how much of the purchase price is attributable to appreciating land versus the depreciating building, citing some greenfield estates with ratios as low as 30% as an example of what to avoid, generally looking for at least 50%.

Damien lists the more typical factors a buyers agent considers (street quality, zoning, easements, flood and bushfire risk, building condition and age, comparable sales, proximity to public housing or main roads), all of which affect the fair price of an individual property. Jeremy's key distinction: these features determine a property's price, not its capital growth, since demand is a function of both features and price together, not features in isolation. In other words, these factors matter for making a sensible offer, but the suburb itself does most of the work in driving growth over time. Any initial "headaches" from an older property needing work tend to become comparatively insignificant once genuine suburb-level capital growth plays out.

A Real Data Example: LGAs From 2016 to 2026

Jeremy pulls up a real back-tested example using DSR3 data. Looking at the top five local government areas by DSR score as of January 2016, Goulburn (NSW, ~28 suburbs), Kiama (NSW, ~22 suburbs), Hobart (~16 suburbs), Central Coast (~152 suburbs), and Gold Coast (~79 suburbs), all had already shown solid growth in the prior three years (Goulburn 33%, Kiama 45%, Hobart a comparatively modest 8%, Central Coast and Gold Coast both around 52%), and had underperformed the national 10-year average leading into that point, described by Jeremy as still "playing catch-up." Despite already having strong recent growth, the DSR3 score still pointed to further strength ahead.

Looking at actual growth from roughly June 2016 to 2019, Hobart led the group with around 35% growth, with Goulburn and Central Coast both adding well over $100,000 in typical value over the same period, confirming the DSR3's signal held up.

At the other end, the bottom five LGAs as of January 2016 were dominated by Western Australian markets (Victoria Park, Wanneroo, Gosnells, and Burwood, alongside Whitehorse in Victoria), each still working through a post-boom hangover from an earlier strong growth run (Whitehorse's growth, for instance, reflected an earlier Victorian boom that had already passed its peak by 2016). Looking ahead to the following three years (June 2016 to 2019), most of these markets were flat or negative, while Gosnells specifically went backwards, while the top five LGAs delivered around 22% combined growth over the same window, again validating the DSR3's earlier signal.

Extending the same comparison out to 2026 (roughly a 10-year view from 2016), Jeremy highlights Gosnells' remarkable turnaround: 152% cumulative growth, moving from around $550,000 to a Gold Coast comparison point of roughly $1.6 million (a $900,000 increase) over the same extended period, illustrating that a market lagging over a shorter 3-year window can still deliver outstanding results over a longer horizon. Whitehorse, by contrast, lagged over this same extended period, a reversal of its earlier position 10 years prior. Jeremy's takeaway: the DSR is explicitly geared toward the next roughly three years, and while buying and selling isn't as simple or low-cost as trading shares, a market with a persistently low DSR score will, sooner or later, tend to show these kinds of below-average results if held regardless.

How Has AI Changed Property Analysis?

Jeremy draws a clear distinction between large language models (ChatGPT, Claude, Gemini, Copilot) and the type of AI actually used in the DSR (data science techniques like regression analysis). He argues LLMs offer no genuine advantage for property market analysis, since they can only summarise or recite existing online content, and given how much low-quality, unverified commentary exists in the property space (referencing the show's own 40+ Expert Busting episodes debunking various claims), an LLM is just as likely to repeat flawed information as sound analysis. The genuinely useful application of AI, in his view, is the data science layer applied to historical market data specifically, which has meaningfully improved the ability to identify and forecast high-growth markets.

The Biggest Mistakes Investors Make With Data

Jeremy identifies several recurring mistakes:

• Starting from a personal belief rather than the data. Some users arrive with a pre-formed idea of what drives growth and search only to confirm it, rather than trusting a properly weighted composite score built on 15+ years of analysis.
• Over-filtering. Setting multiple individual thresholds (for example, vacancy under 2%, days on market under 60, online search interest above 50) can eliminate genuinely excellent markets, since every market has some contradictory metrics. Jeremy repeats his now-familiar example: a 200-house suburb with 40 rentals, where a single vacant property already equals 2.5%, potentially excluding an otherwise outstanding market purely due to small-sample noise. His guidance: use filters only to rule out markets that are genuinely inapplicable regardless of quality (for example, priced outside an investor's budget), not to try to find the best markets.
• Manually chasing trend metrics already built into the DSR. Wanting to see days on market falling, vacancy falling, or online search interest and auction clearance rates rising is redundant, since these trends already feed into the score, and manually choosing an arbitrary lookback period (commonly three years) for this kind of check-in isn't grounded in the same rigorous, metric-specific analysis the DSR applies.
• Applying arbitrary binary thresholds. For example, automatically excluding any suburb with more than 40% growth over three years, treating a nuanced, continuous metric as a simple pass/fail cutoff rather than a properly scored input.
• Combining outside data with the DSR score. Since the exact composition of the DSR isn't published, layering another independent metric on top risks double-counting something already factored in. Jeremy also notes that correctly combining any two variables requires knowing each one's actual statistical correlation to capital growth, without that, an investor is simply guessing, similar to how buyers agents used to operate 15 years ago.

Jeremy's overall framing: investors are increasingly at a point where the real choice is between trusting a well-built algorithm or risking a worse outcome through manual guesswork. Damien adds that this earlier three-year window matters most for shorter-term investors specifically, since for a genuinely long-term buyer, precise suburb selection becomes progressively less critical the longer the intended holding period.

Key Takeaway

Asked what a listener should take away if they remember only one thing, Jeremy points to price point, personal strategy, time horizon, and risk profile as the key starting considerations. Asked whether his own overall approach has changed much despite the rise of AI, Jeremy says no, the core philosophy remains the same, it's simply a matter of using progressively more and better data over time.