Methodology

HDB Insights is built to answer a simple but often misunderstood question:

What do comparable HDB flats actually transact for in the real resale market?

To do this reliably, we analyse transaction data in a way that reflects how buyers, sellers, banks, and valuers compare flats in practice, rather than relying on broad town averages that mix very different flat profiles together.

This page explains the principles behind our analysis and why our approach differs from most property portals.

Our Core Principle: Compare Like With Like

HDB resale pricing is often misunderstood because many summaries treat flats within the same town and flat type as if they are broadly comparable. In reality, two HDB flats that look similar at a glance can still transact at very different prices.

That is because buyers do not compare flats using town medians alone. They compare HDB flats based on key attributes that materially affect value, including remaining lease, floor level, floor area, and proximity to key points of interest such as MRT stations and popular primary schools. When these attributes are mixed together, headline numbers can shift simply because a different mix of flats transacted.

Our core principle is to compare like with like. We group flats into comparable profiles first, then analyse prices and trends within those profiles. This makes the insights more realistic, easier to apply to your own flat, and more consistent over time.

Data Source And Scope

We base our analysis on officially reported HDB resale transactions, with over 30 years of historical data to capture long term patterns across different market cycles.

At the same time, we do not treat the full history as one single trend line. For each town and flat type, we analyse transactions across a relevant multi year window so the insights reflect current market behaviour rather than short term noise.

We use median prices rather than averages because medians are less affected by unusually high or low transactions. This makes the results more stable and easier to interpret when you are comparing flats that are meant to be genuinely similar.

How We Group Flats Before Analysing Prices

Before calculating any medians or trends, each flat is first grouped using three data-driven dimensions.

This grouping step is critical. Without it, town-level medians can shift simply because more newer or larger flats transacted in a given period.

1. Remaining Lease As Behavioural Bands

Remaining lease is one of the biggest drivers of HDB resale prices, but it does not affect prices in a smooth, year by year way. Buyer behaviour tends to change more noticeably around certain lease ranges due to financing rules, CPF usage limits, and expectations about long term value.

Because of this, resale prices often adjust in steps rather than declining evenly as flats age. This behaviour is commonly described through concepts such as Bala’s Curve and age adjusted pricing, which explain why similar flats can transact at different prices once they fall into different lease ranges.

Our methodology reflects this market behaviour by grouping flats into practical behavioural lease bands before analysing prices. Flats are compared within comparable lease ranges, rather than treating remaining lease as a single continuous number, which helps reduce distortion when interpreting price trends.

2. Floor Level As Relative Position Within The Block

Floor level is another factor that strongly influences HDB resale prices, but it is often misunderstood when analysed using absolute storey numbers alone. A flat on the 10th floor can feel very different depending on whether the block is 12 storeys tall or 25 storeys tall.

HDB Buyers typically compare flats based on their relative height within the same block, rather than the exact storey number printed on the listing. A unit near the top of a low rise block is often perceived as a high floor flat, even if its storey number appears modest compared to taller blocks nearby.

Our methodology reflects this behaviour by classifying floor levels based on a flat’s relative vertical position within its own block. This allows flats in blocks of different heights to be compared more fairly, and helps prevent price analysis from being distorted by differences in block design rather than actual buyer preference.

3. Floor Area Using Stable Size Clusters

Floor area plays a significant role in how HDB resale flats are priced, but simple size ranges can be misleading. Flat sizes are not evenly distributed. Instead, they tend to cluster around common HDB design layouts from different building periods.

Because of this, comparing flats using broad or arbitrary size ranges can distort price analysis. A small difference in floor area can represent a meaningful layout change, while a larger difference may have little impact if both flats fall within the same design cluster.

Our methodology groups flats into stable size clusters based on observed transaction patterns, rather than redefining size ranges for each town. These clusters are applied consistently across all towns and over time, allowing price comparisons to reflect how buyers actually compare layouts, and keeping trends comparable even as the mix of flats sold changes.

Proximity Analysis

Proximity to amenities such as MRT stations and popular primary schools can influence HDB resale prices, but it is often overstated when analysed on its own. Location matters, but it interacts closely with other factors such as flat type, remaining lease, floor level, and size.

Our methodology uses block level location data to measure proximity in a consistent way. Distances are calculated objectively and flats are grouped by proximity before any price comparisons are made, rather than relying on rough area boundaries or assumptions.

Most importantly, proximity is never analysed in isolation. Comparisons are only made within the same flat type and within similar lease, floor, and size profiles. This helps ensure that any observed price differences reflect genuine proximity effects, rather than differences caused by the mix of flats being compared.

Why Town Medians Can Be Misleading

Town level medians are useful as a broad reference, but they often move for reasons that have little to do with true price growth. The number can shift simply because the mix of flats sold in that period is different, even when prices for comparable flats are stable.

For example, a town median may rise because more newer flats transacted that month. It may also rise because more higher floor units were sold, or because larger layouts made up a bigger share of transactions. The median can fall for the same reason in reverse, even if demand has not meaningfully changed.

Our approach reduces these distortions by grouping flats into comparable profiles first, then analysing prices within those profiles. This makes trends easier to interpret and helps buyers and sellers understand whether price movements reflect real changes, or just a change in what kind of flats happened to transact.

Sample Size And Reliability

Not every segment carries the same level of statistical confidence. Some groups naturally have many more transactions than others, which makes their medians and trends more stable and easier to interpret.

When transaction volumes are low, we flag this clearly and avoid drawing strong conclusions from limited data. In those cases, the figures should be treated as a rough reference, and it may be more helpful to look at a wider time window or a broader comparable group.

Our goal is to provide insight that is useful and defensible, not to over interpret patterns that may simply reflect normal noise in small samples.

What This Methodology Is And Is Not

What it is:

This methodology is a structured, data led approach to understanding HDB resale prices in a way that is actually useful for real decisions. It is designed to reflect how buyers compare HDB flats in practice, by looking at genuinely similar profiles rather than relying on broad town averages. The same definitions and grouping logic are applied consistently across towns and over time, so price trends can be compared meaningfully without being distorted by changes in flat mix.

What it is not:

This is not a price prediction model and it is not a valuation tool for any individual HDB flat. It also should not be treated as a substitute for professional advice, especially when making decisions on a specific unit. Every HDB flat has unique characteristics that data alone cannot capture, such as renovation condition, interior layout, noise levels, and stack level appeal. The insights here are meant to guide judgement and provide context, not to replace it.

Why This Matters For Buyers And Sellers

For most people, the hardest part about HDB resale pricing is not finding numbers. It is knowing which numbers are actually relevant to a flat like yours. A town median can be a useful headline, but it often hides the fact that transactions are a mix of very different flat profiles. When the mix changes, the median can move even if prices for truly comparable flats are not moving in the same way.

For HDB Buyers:

It matters because it helps you avoid paying the wrong price for the wrong reason. If you rely on a broad median, you might overestimate what a lower lease or lower floor flat is worth, or you might underestimate what a higher lease or higher floor flat typically transacts for. By comparing within comparable profiles, you can make a clearer offer decision, understand what you are paying a premium for, and set expectations more realistically before you commit.

For HDB Sellers:

It matters because it helps you price with confidence and explain your expectations using relevant comparables. Instead of anchoring to a single town number, you can see how HDB flats like yours have actually been transacting, and whether recent movements are coming from genuine price shifts or simply from changes in the mix of flats sold that month. This reduces the risk of pricing too high and extending your time on the market, or pricing too low relative to comparable transactions, while giving you a clearer basis for discussions with buyers and agents.

Our Commitment To Transparency

We believe methodology should be explainable. You should be able to understand how an insight is produced and what it means before you rely on it for a decision.

At the same time, meaningful analysis depends on more than the high level concept. It also requires disciplined data handling, consistent definitions applied the same way every time, and ongoing maintenance as new transactions are added.

This page outlines the principles behind our work so you can understand how insights are produced and how to interpret them responsibly in the context of your own flat.

FAQs About HDB Insights Methodology