Does King County's 96% 'stay housed' rate leave out deaths?

1. Pass Title matches the evidence

   The title asks whether the 96% stay-housed rate leaves out deaths, and the evidence (HUD measure 7b.2 excludes exits recorded as Deceased; the authority's slides do not mention deaths) supports that the rate leaves out residents who die. The title's numbers (96%) appear in the evidence, and the title makes no claim beyond what the evidence shows.

2. Pass No obvious contradictions

   The finding True agrees with the answer and conclusion. Key point 1 (established) that the rule removes residents who die is consistent with the conclusion. The conclusion says the 94% and 91% figures are rough estimates, matching the answer's roughly and the estimate's limitation. The 98% and 93% figures are described as not shown to leave deaths out, consistent across the conclusion, key point 5 and the other-rates evidence. The arithmetic in the article checks out (7,842/8,366 is about 94%; 7,842/8,628 is about 91%).

3. Pass Plausible justification

   The conclusion says the criticism holds for the federal 96% rate because HUD's rule removes anyone whose exit is recorded as a death and the authority reports the 96% without saying so. That reason follows directly from the evidence for a True rating, and the conclusion also limits the rating to the 96% rate.

4. Pass Sources are relevant

   Sources are HUD specifications, the regional authority's reports and dashboard, Health Through Housing and operator annual reports, the county overdose reviews, the Medical Examiner dashboard and the HUD bed count. Each is on topic for the point it supports. The FactHub records are internal analyses of those sources.

5. Pass Common questions answered

   There are seven objections, each with a response that addresses its question: size (no, by about 2 to 5 points), fairness (fair, but it says nothing on deaths), the 98% figures (yes, but neither explains deaths), other rates (not shown), the rule's bias (not in every way), Health Through Housing's lower death count (yes, with caveats), and who wrote the rule (a federal rule).

6. Pass Headline and verdict agree

   Sentence read by a reader: 'Does King County's 96% stay housed rate leave out deaths? Yes.' That tells the reader the rate does leave out deaths, which matches the claim ('leaves out residents who die'), the finding True and the answer ('A federal rule drops residents who die from the rate').

Coverage: Read the article payload only: title, claim, finding, answer, conclusion, key points, evidence and limitation text, objections and responses, method, and each source's name, title, description, role, reliability note and takeaway statements. I did not open source snapshots or the web, and I had no tools. Every takeaway carries status 'not yet checked' and none is marked established, so I did not rely on any takeaway's accuracy. Key point 1 is labeled established by the article, but its support is unchecked. I used source takeaways only for relevance and internal consistency. I checked the article's arithmetic: 299/8,141 is about 3.7%; 225 added gives 6.3%; 487 added gives 9.1%; 7,842/8,366 is about 93.7%; 7,842/8,628 is about 90.9%; 225/7,768 is about 2.9 per 100; 15/1,281 and 10/1,380 are about 1.2 and 0.7 per 100. I did not rate, because source accuracy belongs to the source auditors: the tk-kcrha-counts quote reads '7,874 8,141 267' while its statement says 7,842 of 8,141 and 299. The article mixes 2022-23 counts with 2024 overdose deaths and discloses this. The 487 all-cause figure covers a housing service site category that includes income-restricted and recovery housing; the article discloses this as a limitation. I did not evaluate the photo, the SEO title, the changelog or the related-node notes.