What the index measures
A single composite signal on a zero-to-one hundred scale: the week’s volume of community-relevant signals across three of the four categories below (Physical Security, Online Threats, and Threat Intelligence — Legal & Enforcement is tracked but not part of the composite), measured against each category’s trailing four-week baseline and averaged. Captures ambient threat temperature — events not tied to any single monitored institution.
An earlier version of this index also carried a named-institution “institutional pressure” signal — an average threat score across a panel of reference synagogues, JCCs, and federations. That signal was removed from the index’s design on 2026-07-14 (“we should not be relying on clients to make this a valid analysis”); the reference panel was never seeded into the production database, and this public figure has never scored named institutions.
Signal categories
Four signal categories
- Physical Security — Antisemitic incidents, bias crimes, protests with a direct Jewish target, and other in-person events. Highest-severity category, and the one that carries the most weight in the index.
- Online Threats — Social media volume anomalies, online hate, and coordinated digital activity with identifiable targeting of Jewish communities or institutions.
- Legal & Enforcement — Regulatory actions, sanctions, and enforcement proceedings that intersect with the threat environment for Jewish institutions.
- Threat Intelligence — Extremist organization activity, government advisories, international antisemitism reporting, and coordinated campaign signals with a domestic relevance dimension.
Interpretation bands
The index runs on a zero-to-one-hundred scale, read through five interpretation bands of rising severity.
Interpretation bands · 0–100 scale
- Low0–20
- Moderate20–40
- Elevated40–60
- High60–80
- Critical80–100
Currently pre-calibration (index launched June 2026). Ninety days of continuous operation — through approximately August 2026 — are needed to back-test the bands against documented incident peaks. Until then, the numeric value and the direction of movement are meaningful, and the band labels read as directional rather than fixed.
What we monitor
More than forty data sources: law enforcement and government records; community incident reporting from monitoring organizations; federal regulatory and enforcement databases; open-source extremist monitoring; news and media volume tracking; social intelligence systems. Source list is not published (disclosure would map detection coverage for adversaries). Institutions in contracted engagements receive full source disclosure under NDA.
Limitations
- Cannot capture private communications, encrypted coordination, or non-public institutional targeting until they surface in monitored data.
- Reflects aggregate community conditions, not a risk assessment for any individual institution — the public index does not track named institutions (see above).