The score is a mechanical count of risk-relevant public filings — no opinions, no manual adjustments: the same filings always produce the same score. The exact composition and weights are proprietary. What we publish instead is the part that matters: historical validation results, measured against official outcomes.
A weighted count of monitored financing and capital-structure filings over a multi-year lookback. Every filing that appears in a report links to the original DART source — so you never have to take the score on faith; the evidence is on the page. Critical filings (embezzlement or breach-of-trust filings, audit-opinion issues, disclosure violation filings and KRX administrative-issue filings) are shown separately and do not enter the score. Alongside the score, every company carries one of five risk ratings — LOWEST to HIGHEST RISK; the exact construction is proprietary, and the validation below measures the ratings directly against outcomes.
We rated every listed company at seven different start dates — one per year, 2016 through 2022 — using only the filings available on each date, then checked who suffered a risk-type delisting (audit opinions, insolvency, requirement failures — official KRX reasons) in the years that followed. A decade of outcomes, no look-ahead: the rating never sees the future it is judged on.
The table below is the most recent full cohort (rated Feb 2022, outcomes tracked ~4.5 years). The older six cohorts are the robustness check underneath.
The result in one sentence: of the 76 companies carrying our highest rating, 32 were later delisted for cause — 42.1%, or 32 of 76. Across all 2,389 companies in the cohort, the historical delisting-for-cause rate was 4.2%, or about 1 in 24. Companies in the HIGHEST RISK group were later delisted for cause at about 10× the market-wide rate.
| RATING | COMPANIES | LATER DELISTED FOR CAUSE | RATE | ABOUT 1 IN |
|---|---|---|---|---|
| LOWEST RISK | 1,714 | 19 | 1.1% | 1 in 90.9 |
| LOW RISK | 416 | 14 | 3.4% | 1 in 29.4 |
| MODERATE RISK | 89 | 12 | 13.5% | 1 in 7.4 |
| HIGH RISK | 94 | 23 | 24.5% | 1 in 4.1 |
| HIGHEST RISK | 76 | 32 | 42.1% | 1 in 2.4 |
These are the same five ratings stamped on every report. Even the lowest rating is not an endorsement — it means the record is clean, not that the future is.
Robustness. We also re-ran the whole exercise at seven different start years (2016–2022, each with a full five-year lookback): the highest-score group failed 3.9–8.3× more often than average in every single cohort. The result does not depend on when you start the clock.
One exclusion, and why. SPACs (blank-check shells) are left out of both the population and the outcomes. A SPAC that fails to find a merger target is wound up at its three-year deadline and holders get the escrow back with interest — the exchange records it as a delisting, but it is a scheduled liquidation, not a company failing. Leaving them in made our own ratings look worse at discriminating, not better: it inflated the failure rate of the lowest band with events we never claim to predict. The headline 42.1% is unchanged either way — there were no SPACs in the highest band.
Every event comes from DART — Korea's official regulatory filing system — collected daily. Filings span 2011–2026 — hundreds of thousands of events from listed companies. Delisting outcomes are official KRX records. And every claim on a report links back to the original filing, so you can always check us.
KoreaStockCheckBACKGROUND CHECKS · KRX FILINGS