What score gets you a Seoul apartment?
Every cut-off of the last 12 months
Housing-subscription scores · 28 Seoul private-APT notices · 214 units, exhaustive · public-data analysis
Right after "maybe I should apply" comes the same question every time — "so what score do I need?" We collected every published minimum winning score from Cheongyak Home and answered it with numbers. Over the last twelve months, the median winning cut-off for Seoul private-apartment general supply was 64 points. Even at 60 points, barely one in five units (22%) had its door open; at 50 points, four in a hundred (4%). But once the data was in, the real finding was elsewhere. 64 is not a coincidence. We'll get to that in a moment.
01What we measured
We collected every published minimum winning score from Cheongyak Home (Korea Real Estate Board). From September 2025 through August 2026: 28 Seoul private-apartment notices — 27 of them with published cut-offs, split into 214 units by housing type — general supply, local (Seoul-resident) quota.
The arithmetic is simple: was my score at or above that unit's minimum winning score? We call that share "units with an open door" — coverage (%) = (units where my score ≥ minimum winning score) ÷ 214 × 100.
One line to draw before anything else: this is a comparison against past cut-offs, not a probability of winning. The same score splits differently depending on ties and supply, and next year's distribution will not equal last year's. Still, knowing where your score sits in the distribution — versus not knowing — is a large difference.
02The notches in the cut-offs are family sizes
The distribution piles up at 64–69. Why there? The answer was in the scoring formula, not the data. The score is a sum of three items: years without owning a home (max 32), dependents (5 points for yourself, 5 more per person), and years on the subscription account (max 17). Suppose you max out the first and the third. From that moment, the only variable left is how many of you there are.
| Household | Perfect score (32 + dependents + 17) | Units with an open door |
|---|---|---|
| Solo | 54 | 10% |
| Couple | 59 | 21% |
| Family of 3 | 64 | 51% |
| Family of 4 | 69 | 85% |
| Family of 5 | 74 | 98% |
Figure 1 · Coverage by score — all 214 units, Seoul private APT general supply, Sep 2025–Aug 2026. Source: Cheongyak Home
The median cut-off, 64, is exactly the perfect score of a three-person household. A family of three that maxes out — fifteen years without owning, fifteen years on the account — gets a perfect score and finds exactly half of Seoul's doors open. A couple's perfect 59 opens one in five (21%); a solo applicant's perfect 54, one in ten (10%). Even a family of four's 69 left 15% closed, and one complex in Seocho (Acro de Seocho) cleared only at 84 — a perfect score.
So this distribution does not say "Seoul is full of high scorers." The cut-off distribution is not the shape of demand — it is the shadow of the ceilings the system builds over each household size. Time fills in the other two items; dependents it does not. Which is why the real answer to "what score do I need" is less a number than "how many of you are there?"
Figure 2 · Distribution of winning cut-offs — the same 214 units, 5-point bins. Dashed line: median 64
We also counted where the 15% that even a perfect 69 could not open actually is. Of the 32 units with cut-offs at 70 or above, 18 are in Seocho-gu. More than half of "the Seoul where even maxed-out families lose" is really the story of a single district.
03Where the dammed-up demand goes — leftovers at 276:1
This is the coldest number in the piece. The standard advice for anyone under the ceiling is "try the post-sale leftovers" — homes whose contracts fell through, resold by straight draw with no points at all. So we counted every leftover sale in the same Seoul, over the same twelve months: 999 homes supplied, 276,141 applications — 276:1 overall, a median of 42:1 per unit, and at the top, one home in Yeongdeungpo drew 130,938 applicants.
This number and the table above are one body. When the points system builds a ceiling over each household size, the demand under it does not disappear — it pours into the doors that ignore points. The leftover round became a lottery not by accident but because it is the points system's overflow drain. "Try the leftovers" is not wrong; at 276:1 it is just not a strategy. It is buying a ticket.
04The doors that actually open are types, not luck
The lowest cut-offs: 29 points (De Pine Artia 110㎡A), 32 points (The Richmond Mia 49㎡). What they share is very large or very small — housing types demand passed by. Those doors did not open because low scorers won; they were open because nobody was queuing. "Complexes that opened at low scores" were a matter of type selection.
05So what do I actually do
Your score is not something you can change. The one thing you choose is which line to stand in. Read your household's row — every number is the same measurement as above.
| You are | Even at a perfect score | What you do |
|---|---|---|
| Living alone | 54 — one door in ten (10%) opened | Forget the points line. Start from the lottery-allocated units that ignore points, and special supply if you qualify |
| A couple | 59 — one in five (21%) | Same as above — on points, queuing simply does not win |
| A family of 3 | 64 — half (51%) | Pick the sizes nobody wants. The doors that opened at 29 and 32 points were all very large or very small homes |
| A family of 4 | 69 — most (85%) | Most doors open. Except Seocho, which is another world — 18 of the 32 doors above 70 points were there |
And the leftovers: applying is free, but never put them in the plan. 276 to 1 is not a strategy. It is a lottery ticket.
06Why these numbers were scattered
Every number in this piece is public. The reason you have never seen this picture is that the cut-offs are scattered across notice pages, mixed with unpublished "-" entries, and split by housing type and residency — no human hand turns them into one picture. You have to stitch together all 214 units, plus 114 leftover units, before these tables exist. We are the team that prices non-apartment property — multi-family homes, shops, small buildings — from actual transactions, and we ran this measurement so that next to "neighbourhoods you can reach with ₩300M" the same screen can say how far open the alternative door of a new-build subscription actually was. Enter your score, get your coverage and the doors your range should aim at — we are building it into hicce. In the meantime, to run the 214 units above against your own score, try the coverage tool — same data as this piece.
Method — Cheongyak Home open APIs (sale notices, competition ratios, winner scores; data.go.kr) · Seoul private APT general supply · local quota · notices 2025-09-01 to 2026-08-31, 27 with published cut-offs · 214 units ("-" unpublished excluded) · special supply and lottery allocations excluded. Scoring formula (32 max for years without owning · 5+5/person for dependents · 17 max for account years) per Cheongyak Home's published criteria. Leftovers: every leftover-round competition ratio for the same Seoul window (114 units). Coverage is a retrospective comparison, not a probability of winning.
Same method, different industry. The bottleneck differs every time — finding that difference is what this series does.
One industry, one bottleneck, public data only — straight to your inbox. Nothing else, ever.
This is as far as public data can see. Repurchase, incrementality, attribution — the numbers that change decisions live inside your own data, and making them countable is what Lambency does.
caffrey.w.lee@gmail.com