The Same "CCR" Label Covered a $4.8 Million Gain and a $2.4 Million Loss. The Region Label Predicted Almost Nothing.
Ardmore Park in Tanglin and Marina Collection in Sentosa Cove share the same three-letter label on every market report: CCR. Sellers at Ardmore Park averaged a $4.8 million gain across 12 resales, holding for roughly 15 years. Sellers at Marina Collection averaged a $2.4 million loss across 7 resales, holding for nearly 17 years. Neither the region label, nor freehold tenure, nor patience separated them. This article tests each of those explanations one by one — and finds that all three fail.
Marina Collection, Sentosa Cove: 7 resales. 0 profitable. Average loss: $2.4 million. Typical hold: nearly 17 years. ⚠️ Fewer than 10 transactions — treat as directional.
Two Projects. Same Label. Opposite Outcomes.
Here is the full picture before any explanation. CCR winners on the left, CCR losers on the right, OCR winners below.
CCR — Sellers who gained
| Project | Sub-area | Tenure | Typical unit (sqft) | Win rate | Avg outcome | Avg hold | Transactions |
|---|---|---|---|---|---|---|---|
| Ardmore Park (D10) | Orchard/Tanglin | Freehold | 2,885 | 100% | +$4,841,612 | 15.7 yrs | 12 |
| Grange Residences (D10) | Orchard/Tanglin | Freehold | 2,852 | 100% | +$4,248,044 | 17.8 yrs | 10 |
CCR — Sellers who lost
| Project | Sub-area | Tenure | Typical unit (sqft) | Win rate | Avg outcome | Avg hold | Transactions |
|---|---|---|---|---|---|---|---|
| Marina Collection (D4) | Sentosa Cove | 99-yr | 3,272 | 0% | –$2,434,336 | 15.2 yrs | 7 ⚠️ |
| Seascape (D4) | Sentosa Cove | 99-yr | 2,669 | 0% | –$936,491 | 6.1 yrs | 11 |
| Marina Bay Suites (D1) | Downtown Core | 99-yr | 1,625 | 10% | –$648,914 | 13.6 yrs | 20 |
| The Coast At Sentosa Cove (D4) | Sentosa Cove | 99-yr | 2,024 | 16% | –$464,712 | 13.8 yrs | 19 |
| OUE Twin Peaks (D9) | River Valley | 99-yr | 1,055 | 7% | –$331,651 | 7.9 yrs | 43 |
| Marina One Residences (D1) | Downtown Core | 99-yr | 1,044 | 4% | –$292,539 | 7.0 yrs | 56 |
| Marina Bay Residences (D1) | Downtown Core | 99-yr | 1,076 | 44% | –$221,233 | 11.6 yrs | 36 |
OCR — Sellers who gained
| Project | Sub-area | Tenure | Typical unit (sqft) | Win rate | Avg outcome | Avg hold | Avg resale PSF | Transactions |
|---|---|---|---|---|---|---|---|---|
| Treasure At Tampines (D18) | Tampines | 99-yr | 915 | 99.7% | +$345,653 | 4.7 yrs | $1,753 | 341 |
| Riverfront Residences (D19) | Hougang | 99-yr | 721 | 100% | +$327,823 | 5.9 yrs | $1,721 | 163 |
| Sol Acres (D23) | Choa Chu Kang | 99-yr | 926 | 100% | +$569,735 | 7.7 yrs | $1,491 | 158 |
| Inz Residence (D23) | Choa Chu Kang | 99-yr | 1,039 | 100% | +$659,237 | 7.6 yrs | $1,428 | 118 |
| Kingsford Waterbay (D19) | Hougang | 99-yr | 689 | 94.4% | +$161,246 | 7.1 yrs | $1,453 | 143 |
The CCR losers are not outliers. Marina One Residences alone accounts for 56 transactions — 2 profitable out of 56. OUE Twin Peaks adds another 43, with a 7% win rate. The loss pattern holds across multiple projects, districts, and hold lengths. All of this happened while the broader private market appreciated roughly 6.8% over the same period, measured by the URA non-landed price index.
The obvious first question: is freehold tenure what separated the winners from the losers?
First Hypothesis: Freehold Tenure Explains the CCR Winners
Ardmore Park and Grange Residences are both freehold. The loss leaders — Marina Collection, Seascape, Marina Bay Suites, The Coast, Marina One Residences, OUE Twin Peaks — are all 99-year leasehold. On the surface, this looks like a clean freehold-versus-leasehold story.
It is not.
CCR freehold projects that lost money on most or all resales
| Project | Sub-area | Tenure | Typical unit (sqft) | Win rate | Avg outcome | Avg hold | Transactions |
|---|---|---|---|---|---|---|---|
| Scotts Square (D9) | Orchard | Freehold | 947 | 18% | –$575,271 | 14.4 yrs | 17 |
| Helios Residences (D9) | Newton | Freehold | 1,491 | 0% | –$1,173,443 | 12.6 yrs | 10 |
| Cliveden At Grange (D10) | River Valley | Freehold | 2,498 | 0% | –$1,920,380 | 13.9 yrs | 8 ⚠️ |
⚠️ Cliveden: 8 transactions — directional only.
Scotts Square is freehold. It sits in D9, Orchard, within the same broad Orchard belt that Ardmore Park and Grange Residences occupy. It has 17 transactions — enough to state the finding plainly. Three of those 17 resales were profitable. The average seller lost $575,271 over a 14-year hold.
Helios Residences is also freehold, also D9, and lost on every single one of its 10 resales. Cliveden at Grange is freehold, D10 — the same district as Ardmore Park — and lost on all 8 resales, though at only 8 transactions that is directional, not definitive.
Notice something else: Scotts Square, Helios Residences, Cliveden at Grange, and the worst Sentosa Cove performers — Marina Collection, Seascape — all completed in 2011. That timing matters. The CCR market peaked in 2011–2012. Buyers who completed purchases in that window were exposed to that cycle peak. A plausible explanation for why freehold did not protect them is not that freehold is a flawed concept — it is that entry timing appears to have overwhelmed the tenure advantage. That inference is drawn from completion year, not individually verified per transaction, so it should be treated as the most probable account rather than a proved cause.
Freehold tenure is not the variable that separated winners from losers. Entry timing and location within CCR appear to matter far more.
Second Hypothesis: Holding Longer Would Have Rescued the Losers
The second defence of the CCR loss leaders is a patient-capital argument: maybe sellers simply did not hold long enough. Give it more time, and the losses would have unwound.
The hold-duration data does not support this.
| Project | Typical hold | Win rate | Transactions |
|---|---|---|---|
| Marina Collection (D4) | 16.7 yrs | 0% | 7 ⚠️ |
| Marina Bay Suites (D1) | 15.1 yrs | 10% | 20 |
| Ardmore Park (D10) | 14.7 yrs | 100% | 12 |
Marina Collection sellers held for a typical 16.7 years. Marina Bay Suites sellers held for a typical 15.1 years. Ardmore Park sellers held for a typical 14.7 years. All three are within two years of each other on holding period. The outcomes are opposite.
Marina Bay Suites adds a further layer. A typical unit of 1,625 sqft was reselling for around $3.15 million. At a typical rent of $11,000 per month for a 3-bedroom unit, that implies roughly 4.2% gross yield — not a weak number in isolation. Sellers held for an average of 13.6 years with tenants in the building throughout. Still, only 2 of 20 resales were profitable.
Seascape shows a different pattern but arrives at the same destination. While the average hold was 6.1 years, the typical hold was only 3.7 years — sellers were cutting losses early rather than riding it out. Despite that, not one of 11 resales recovered. Seascape sellers who left quickly and Marina Collection sellers who stayed nearly 17 years both lost on every transaction.
Duration is not the differentiating variable.
Third Hypothesis: High PSF Caused the CCR Losses
Before arriving at the structural explanation, one more competing account needs testing: the idea that any project transacting above a certain price per sqft will eventually see losses, regardless of sub-market.
Parc Esta eliminates this cleanly. Parc Esta sits on Sims Avenue in Geylang (D14, RCR, 99-year leasehold), about an 8-minute walk from Aljunied MRT. Its typical unit size is 743 sqft, and its average resale PSF is $2,284 across 236 transactions — higher than Seascape's $1,925 PSF on a typical 2,669 sqft unit and higher than The Coast at Sentosa Cove's $1,556 PSF on a typical 2,024 sqft unit. Its win rate is 100%.
A project transacting at higher PSF than both Sentosa Cove loss leaders, on a 99-year lease, produced a perfect win rate over 236 transactions. The loss pattern is not a PSF threshold. It is sub-market-specific.
What the OCR Data Actually Shows
The five OCR projects in the table above share the same 99-year leasehold tenure as every Sentosa Cove and Downtown Core loss leader. The difference is not tenure. The difference is not holding period — OCR sellers held for 5–8 years and still won at 94–100% rates. The difference is not PSF.
The mechanism appears in the buyer pool.
OCR vs CCR resale transaction volume (URA, non-landed private, resale only)
| Quarter | OCR resale | CCR resale | Total resale transactions | OCR:CCR ratio |
|---|---|---|---|---|
| Q4 2024 | 1,940 | 1,131 | 3,071 | 1.72× |
| Q1 2025 | 1,826 | 1,064 | 2,890 | 1.72× |
| Q2 2025 | 1,921 | 1,086 | 3,007 | 1.77× |
| Q3 2025 | 2,006 | 1,165 | 3,171 | 1.72× |
| Q4 2025 | 1,834 | 1,021 | 2,855 | 1.80× |
| Q1 2026 | 1,715 | 913 | 2,628 | 1.88× |
The OCR resale market has been 1.7 to 1.9 times as liquid as CCR in every single quarter across six consecutive quarters, and the ratio has drifted wider — from 1.72 in Q4 2024 to 1.88 in Q1 2026.
This is not simply a market-size difference. It reflects the pool of buyers who can access each price point at resale. The implied entry prices for the five OCR projects range from roughly $810,000 (Sol Acres) to $1.26 million (Treasure at Tampines). When those sellers come to market at $1.3 million to $1.6 million, the pool of people who can buy is large — first-time private buyers, HDB sellers moving into private homes, investors, downsizers. Each transaction creates a competitive resale environment.
The implied entry prices for a large Sentosa Cove unit run to several million dollars. When a Marina Collection seller comes to market at $4–6 million for a 3,272 sqft unit, the available buyer pool is much narrower than at a $1.3 million to $1.6 million OCR resale price point. The data is consistent with a buyer-pool problem rather than a simple holding-period problem: waiting longer did not change Marina Collection's outcome, while lower-quantum OCR projects kept clearing at high win rates.
The rental data reinforces this. Treasure at Tampines generated 751 rental transactions in under two years. Riverfront Residences generated 539. Marina Bay Suites generated 155 despite its 4.2% gross yield. Marina Collection generated 92. Tenant demand at Sentosa Cove and the Downtown Core waterfront is active — but it is not deep enough to reliably produce a resale buyer.
One genuine anomaly in the Downtown Core cluster deserves a mention. Marina Bay Residences achieved a 44% win rate — 16 of 36 sellers came out ahead, noticeably better than Marina One Residences (4%) or OUE Twin Peaks (7%). It completed in 2010, one year before the rest of the cluster, which likely means some buyers entered before the 2011–2012 peak. The average profit across all 36 resales is still negative at –$221,233. It is the least bad Downtown Core project, not a successful one.
What the Data Actually Shows
The CCR label contains at least two completely different risk profiles. D10 freehold in the Orchard/Tanglin belt — where Ardmore Park and Grange Residences sit — produced 100% win rates and multi-million dollar gains over 15–18 year holds. The Sentosa Cove and Downtown Core waterfront cluster — same three-letter label — produced 0–16% win rates over holds of similar length, in a period when the broader private market was appreciating.
The variables that appear to predict the outcome are sub-market, entry price level, and the depth of the resale buyer pool at the price point where the seller needs to exit. Freehold tenure did not predict it — Scotts Square and Helios Residences lost while holding freehold. Holding duration did not predict it — 16.7-year typical holds at Marina Collection produced the same 0% win rate as shorter holds elsewhere. High PSF did not predict it — Parc Esta at $2,284 average resale PSF maintained a 100% win rate across 236 transactions.
The region label, on its own, predicted almost nothing. The specific geography within that label, the entry price, and whether a large enough pool of buyers could realistically access the resale price — those were the variables that separated a $4.8 million gain from a $2.4 million loss, across projects that an aggregate market report would file in exactly the same column.