One university, two defensible rules for missing evidence, two different bands.
“UNC Chapel Hill measures 28.1 with missing evidence scored 2 and 41.3 scored 3; once the evidence-gap convention is settled, will it sit in the Developing band or the Functional band (40 to 60)?”
Paris is ranked 30th among global cities with a published score of 62.5 out of 100, in the Established band (60 to 80). That number was never checked against evidence.
Why it matters
Paris's published 62.5 was never measured. A first evidence-based reading is 40.6 out of 100, at low confidence. The city's yearly homeless count earned credit. The October curfew was ordered by the national police, not the city.
The next scheduled Florida execution and the status of the paused Thompson case.
How to read this briefing— Bands, scores, and terms
Schema guide
The 5 performance bands
critical0–20
developing20–40
functional40–60
established60–80
exemplary80–100
Two scales
Each of the 8 dimensions is scored 1.0–5.0; these combine into a 0–100 composite score, mapped to the 5 bands above.
Key terms
Band crossing
A score change large enough to move an entity from one performance band into an adjacent one — the most structurally significant finding in a given cycle.
Boundary watch
An entity whose current score is within 3 points of a band threshold, flagged for priority reassessment in the next cycle.
Carry-forward
A dimensional credit retained from a prior assessment when new evidence is insufficient to revise a specific dimension; disclosed explicitly.
First baseline
An entity's inaugural composite score — no published score exists to compare against, so delta is not shown.
Floor designation
The most serious finding: all 8 dimensions resolve at the lowest behavioral anchor (1.0/5.0) across multiple cycles, yielding a composite of 0.
Forward trigger
A scheduled future reassessment event (e.g., a policy implementation date or legislative deadline) that may materially change an entity's score.
UNC settled 18 federal civil-rights probes without paying or admitting fault. A first reading is 28.1 out of 100. If missing evidence were scored 3 instead of 2, it would read 41.3, so the band depends on an unsettled rule.
The University of North Carolina at Chapel Hill (UNC) is ranked 64th among 100 universities with a published score of 43.8 out of 100, in the Functional band (40 to 60). That number was never checked against evidence.
Where this sits
43.816.2 pts to Established
Read the full signal
A first full assessment measures 28.1, in the Developing band (20 to 40), with low confidence. The published score is unchanged, and the proposed change has not been applied.
This is a first measurement, not a measured decline in conduct. Most first measurements so far have come in below their starting value.
On 19 October 2026 UNC signed a voluntary agreement with the federal government that closed 18 Title VI civil-rights investigations. UNC paid nothing and admitted no wrongdoing. It must share complaints and report to the Justice Department every quarter. UNC had already closed its diversity offices and cut 61 positions by August 2026.
UNC is credited for the Carolina Covenant, which lets its lowest-income students graduate debt-free.
Twenty-one of the 40 sub-measures have no evidence. The assessment scores them at 2 out of 5. At 3, UNC would read 41.3, which stays in the Functional band. The band crossing exists only under the first rule.
Within 90 days, share all complaints of alleged Title VI violations.
Portugal's published 60.6 sits 0.6 points above the Established line at 60. New evidence on police handling of housing protests puts the measured score at 59.4, just below it. Confidence is low and nothing is applied.
Portugal is ranked 39th among countries with a published score of 60.6 out of 100, in the Established band (60 to 80). A new assessment measures 59.4, in the Functional band (40 to 60), a difference of 1.2 points, at low confidence.
Where this sits
60.619.4 pts to Exemplary
Read the full signal
The published score is unchanged, and the proposed change has not been applied.
Before dawn on 20 October 2026, national police cleared housing-protest camps in Lisbon and Porto and detained 55 people, according to Reuters. The police called the clearances calm and proportionate. Amnesty International Portugal asked on 15 October 2026 for an independent inquiry into an earlier Porto operation. Protesters said they were hit on the head.
On 30 September 2026 parliament backed a bill that speeds up evictions and ends rent controls on new leases.
Three sub-measures are now scored 3 instead of 4 out of 5: Validation, Priority for the Vulnerable, and Consistency Under Pressure. None of the adverse sources comes from a court, government watchdog or international body, so the crossing is filed at low confidence.
Lisbon's own pending 35.0 reading matched again. Its police seized 228 structures at the Rossio camp, and Mayor Carlos Moedas called the action proportionate.
calm, proportionate and appropriate, with no police charge
France's riots from 17 October 2026 brought thousands of arrests and a minister's resignation offer. The measured score is 52.5 against a published 62.5, but no court or watchdog source backs a band crossing, so no proposal was filed.
France is ranked 28th among countries with a published score of 62.5 out of 100, in the Established band (60 to 80). That number is a shared starting value. A new assessment measures 52.5 for the second time, in the Functional band (40 to 60), at low confidence. The published score is unchanged.
Where this sits
62.517.5 pts to Exemplary
Read the full signal
No proposal was filed.
The riots began on 17 October 2026. About 1,200 people were arrested that first evening, and about 2,400 later. Curfews followed, and a teenager lost a hand to a stun grenade. A police watchdog is examining about two dozen complaints of police violence. Interior Minister Nunez offered his resignation and President Emmanuel Macron rejected it.
The crossing is not filed for two reasons. No court, government watchdog or international body source was fetched; the sources are wire reports. And the starting value of 62.5 is shared with other unmeasured entities, so the gap reflects calibration as much as conduct.
The Paris curfew of 19 and 20 October 2026 was ordered by the national police prefecture and is scored under France. No sub-measure changed.
Storm Adam killed at least four people in eastern Spain on 19 and 20 October 2026. Fresh readings of Spain (51.2) and Valencia (41.9) match proposals already on file. The published scores are 60.0 and 60.9 and nothing is applied.
Storm Adam brought up to about 700 litres of rain per square metre in 24 hours to eastern Spain on 19 and 20 October 2026. At least four people died, and about 800,000 residents were told to stay indoors under a regional order. National forces made more than 400 rescues.
Where this sits
60.0At Established threshold
Read the full signal
Spain's published score is 60.0 out of 100. A new reading measures 51.2 for the third time, against 50.6 in the proposal filed on 26 August 2026. No sub-measure moved, and no duplicate proposal was written.
Valencia's published score is 60.9 out of 100. A new reading measures 41.9, identical to the proposal filed on 1 October 2026. The city of Valencia was outside the stay-indoors order for 145 municipalities. The deaths were in Oliva, Pego and Sagunto, so no municipal conduct was scored.
Both readings rest on earlier evidence. The storm adds context, not a new finding.
Florida executed James Robertson on 13 October 2026, two weeks after a failed execution in Tennessee. The practice is already in Florida's score. The reading is 37.5 out of 100 against a published 38.1.
Florida's published score is 38.1 out of 100, in the Developing band (20 to 40). The new reading is 37.5, within the 5-point threshold. The published score is confirmed and unchanged. Florida executed James Robertson, 63, by lethal injection on 13 October 2026.
Where this sits
38.11.9 pts to Functional
Read the full signal
. It was the state's 17th execution of the year and its first since Tennessee's failed execution of Christa Pike. Governor Ron DeSantis paused the execution of Thompson, 74, who has a dementia claim, for now, according to EFE.
The repeated practice confirms the existing reading. Florida sits 1.9 points below the Functional line at 40.
Global Cities — Paris Joins the Cities Measured Far Below Their Starting Values
Paris measures 40.6 against a published 62.5; Lisbon and Valencia match proposals already on file.
Read the full signal
Paris: published 62.5, measured 40.6, at low confidence. The annual homeless count and shelter expansion are credited; the national police curfew is scored under France.
Lisbon: published 60.9, measured 35.0 again, matching the proposal of 14 October 2026. Its police seized 228 structures at the Rossio camp.
Valencia: published 60.9, measured 41.9 again, matching the proposal of 1 October 2026. The city was outside the stay-indoors order.
10 signals shown
Risk signals
Developments that may affect future scores. Watch items from the Oct 22 briefing.
Risk
Published scores that were never measured may sit well above what a first reading of the evidence supports.
Risk
The value given to missing evidence can decide a band. UNC reads 28.1 at a gap value of 2 and 41.3 at 3.
Risk
Europe-wide cost-of-living and housing unrest may bring further riots, curfews and camp clearances.
Risk
American Express was fined $350 million by the US bank regulator on 8 October 2026 and has no detailed assessment yet.
Risk
Tennessee's failed execution and Florida's pace keep US execution practice under scrutiny.
Measured differences
Entities whose new measurement differs from the published score (not applied), followed by confirmed positions.
15 assessed
Measured differences— 13 not applied to the published score
A pause or reform of Florida's execution pace, such as the paused Thompson execution becoming permanent, would be new evidence on its execution practice.
boundary-watch
Evidence ledger
Primary sources reviewed in this briefing cycle. 22 sources linked.
parisTier 5 · Gov/Court2026-07-24
3 858 personnes rencontrées étaient sans solution d'hébergement durant la nuit du 22 janvier 2026.
Lisbon's police cleared the Rossio housing-protest camp on 20 October 2026, seizing 228 structures; Mayor Carlos Moedas called the action proportionate.
Florida's 17th execution of 2026 raised pressure against the death penalty; a separate execution, of Thompson, a 74-year-old with a dementia claim, was paused for now.
New York's attorney general announced a settlement requiring Dollar General to pay a $1.875 million penalty as food and goods, after about 14% of secret-shopper items were overcharged.
Amnesty International reported five protest-related executions between 30 September and 5 October 2026, after trials it said relied on torture-tainted evidence.
Police raided the Istanbul headquarters of opposition-aligned Halk TV on 14 October 2026 and a government trustee took over the channel.
Floor designations
·8 entities at composite 0 with documented evidence pattern
Composite scores resolving at zero — methodology disclosure
These entities consistently score the worst result across all 8 dimensions of compassionate conduct — the benchmark's most serious classification.
What “floor” means: every one of the 8 dimensions (Recognition, Response, Reduction, and 5 others) resolves at the lowest behavioral anchor (1.0/5.0) across multiple assessment cycles, yielding a composite score of 0. Full methodology.
University of North Carolina at Chapel Hill — The benchmark's choice of one gap value for missing evidence, and UNC's first quarterly report to the Justice Departm…
Portugal — A court, watchdog or international-body record on the October 2026 camp clearances in Lisbon and Porto.
France — The police watchdog's findings on complaints of police violence during the October 2026 riots.
We reassess nightly.
Special Briefings
Thematic deep-dives: cross-index analysis, structural patterns, and interpretive findings.
Copy-ready citation string for journalism, research, or academic use.
Compassion Benchmark. "Daily Briefing — Oct 22." compassionbenchmark.com/updates/2026-10-22. Accessed [Month Year]. Independent — entities never pay for inclusion, score changes, or suppression of findings.
Daily briefings surface the headline finding. Full benchmark reports include all 40 subdimension scores, complete evidence trails, certified assessments, and sector-level analysis packages — the record researchers and journalists cite.
Independence note: entities never pay for inclusion, score changes, or suppression of findings. Commercial services support access, interpretation, and institutional use only.