VALUE FINDS · ISSUE 006

Which condos in Singapore made owners the most money? Two very different lists. Here are both.

By rate, Penrose in D14 made owners the most: a median 7.0% a year across 222 exits since 2021. By dollars, Rivergate in D09 did: a median gain of $1.68m over about 13 years. We ranked 439 projects on 50,427 matched exits to find out, and the two winners could hardly be more different.

Farhan Adenan · 29 Sep 2026 · 9 min read

POV matched-pair model: matched pair = same unit bought then sold. Exits from January 2021 to 1 Sep 2026, 439 projects with at least 40 matched exits each. Gross of stamp duty, interest and agent fees.

Penrose in D14 made owners the most money per year held. Across 222 exits since January 2021, its median seller earned 7.0% a year, every single exit was sold above its purchase price, and the median gain was $399,500 after a hold of about four years.

Rivergate in D09 made owners the most money in dollars. Across 71 exits, its median seller walked away $1.68m ahead, after about 13 years, at a rate of roughly 4% a year.

First, the yardstick. We ranked every condominium and apartment project in Singapore with at least 40 matched exits since January 2021: 439 projects and 50,427 exits. The median project returned 3.07% a year. In that median project, 98.5% of exits were sold at a gain. The 90th-percentile project managed 4.74% a year. Anything above 5% a year is therefore top-decile, and every name on the first list below clears it.

The spread between the best and worst record is more than nine percentage points a year.

Eight of the twelve are 99-year leasehold. Seven of those eight share one profile: Penrose, Parc Clematis, Pasir Ris 8, Whistler Grand, Clavon, Parc Esta and Treasure at Tampines. Their median holds run from 4.0 to 4.6 years, at least 99.7% of their exits were profitable, and their median gains sit between $257,000 and $399,500. None is in D09, D10 or D11. They are D05, D14 and D18 projects.

What this list records is a cycle, not a formula. A median hold of about four years ending in 2021 or later is, for most of these buyers, a purchase made from new and a sale made around completion, into a rising private market. The rate is high because the hold was short and the market underneath it was rising.

Treasure at Tampines earns a line of its own. It has 728 exits, the largest exit pool on either list, and still posts a 5.5% median with every exit profitable.

The four freeholds on the list got there the slow way. Haig Court (D15), The Calrose (D26), Bullion Park (D26) and The Nexus (D21) have median holds of 13.3 to 15.7 years, at 5.4% to 5.8% a year. The one long-held 99-year name is The Gardens at Bishan (D20): 5.4% a year over 12.8 years. A rate like that, compounded for more than a decade on a larger entry price, is how you get to List 2.

Every one of the twelve is freehold or 999-year. Every median hold falls between 11 and 16 years. Four of the twelve are in D21 (Maple Woods, The Blossomvale, The Nexus, Pandan Valley), three are in D15 (The Sea View, Haig Court, The Esta), two are in D10, and D03, D09 and D11 have one each.

And the rates are ordinary. Rivergate's median is roughly 4% a year; Sommerville Park's roughly 3%. The Sea View, Maple Woods, Pandan Valley and Thomson 800 sit around 5%. None of these would top the first list. They are large entry prices compounding for a long time.

Put the two list-toppers side by side. Penrose ran at 7.0% a year and handed its median seller $399,500. Rivergate ran at roughly 4% a year and handed its median seller $1.68m, more than four times the cheque at a slower rate. The difference is quantum times time: a much larger starting price, held about three times as long.

Only two projects make both top-12 lists. Haig Court (D15, freehold) posts 5.8% a year and a $1.46m median gain over a 15.7-year hold. The Nexus (D21, freehold) posts 5.4% a year and $1.18m over 14.8 years. Widen each list to 20 and The Calrose (D26) and The Sea View (D15) join them, with the same profile again: freehold, held 12 to 14 years, rates above 5%.

So “which condo made the most money” is two questions, and each has a different kind of building for an answer.

• If you mean rate, the answer is a 99-year project in D05, D14 or D18, bought and sold about four years later, inside a rising cycle.
• If you mean dollars, the answer is a large freehold unit in D09 to D11, D15 or D21, held for more than a decade.

The buildings that win both are the ones that combined a long hold with an above-average rate. That is holding power, not timing. Haig Court's median seller did not pick a moment to sell. They stayed for more than fifteen years, and the rate compounded on a large base the whole way.

This is also why a headline “best performing condo” figure is close to useless without the hold beside it. A 7.0% rate over four years and a 4% rate over thirteen are not the same achievement, and they are not available to the same buyer.

The bottom of the table is just as concentrated as the top. OUE Twin Peaks (D09) sits last at −2.28% a year, with only 3.9% of its 76 exits sold at a gain. Marina One Residences (D01) is at −1.62% a year, 15.8% profitable. Marina Bay Suites (D01) is at −1.27% a year, with a median loss of $646,000 after 13.1 years.

Then come The Coast at Sentosa Cove (D04, −0.62%), V on Shenton (D01, −0.61%), One Shenton (D01, −0.50%), Marina Bay Residences (D01, −0.30%), The Oceanfront @ Sentosa Cove (D04, −0.17%), The Clift (D01, −0.16%) and The Laurels (D09, −0.16%).

Nine of the ten are 99-year projects in D01, D04 and D09. The exception is The Laurels, a freehold in D09, and freehold alone did not protect it. Long holds did not rescue these buildings either: Marina Bay Suites, The Coast at Sentosa Cove and The Oceanfront all have median holds of 13 years or more. Patience compounds a sound entry price. It does not repair a stretched one. We covered the district-level version of this in who actually loses money.

If you own in a project on either list, the record says your building has rewarded the people who held it. It does not say the next four years repeat the last four. The fast-money list is largely the scoreboard of one cycle, and a scoreboard is not a forecast.

If you are buying, the lesson is not “buy a new launch and sell at completion”. That trade worked for a particular cohort in a particular market, and the figures here are before the stamp duties and fees that trade attracts. The durable lesson sits in List 2 and in the overlap: a quantum you can carry comfortably, a hold you can sustain through a flat patch, and a building whose exit record shows the next buyer keeps turning up.

Either way, check your own project against the yardstick. 3.07% a year is the median project; 4.74% is the 90th percentile; below zero is where the ten names in section 05 live. A project median also hides the spread between its best and worst stacks, which is why we pull the unit-level record first. If you are hunting for value rather than past winners, the condos trading below their own district is the companion read.

MARKET & POLICIES · INSIGHT 002

We checked 67,000 property exits. Here's who actually loses money.

67,354 five-year exits, decomposed by district. D01 lost money most of the time; D28 almost never did.

VALUE FINDS · ISSUE 005

318 condos trade 20%+ below their own district. The 50% club is where it gets interesting.

318 projects trade 20%+ below their own district median. The deep-end table — value or trap, decided by exit records.

Dataset — POV matched-pair model (matched pair = same unit bought then sold) on 623,578 private non-landed sale records, 1995 to 1 Sep 2026, yielding 224,023 pairs held at least one year. This ranking uses the 50,427 exits since January 2021 across 439 projects with at least 40 pairs each. Gross of stamp duty, interest and fees.