BLACKSWAN

1. Six drivers

Each driver reads one public source and counts recent activity.

2. Each becomes 0 to 100

A fixed, published rule turns each reading into a score from 0 to 100 (rules below). There is no model and no tuning by hand.

3. Equal weights

The headline is the plain average of the drivers measured right now. Every driver counts the same. 0 to 24 is Calm, 25 to 49 Elevated, 50 to 74 High, 75 to 100 Severe.

4. Updates

Sources are re-read about every 15 minutes. This panel refreshes every 5 minutes.

5. Missing data

If a source fails, its last value is kept for up to 6 hours and marked Stale with its time. After that it is marked Unavailable and left out of the average, so the headline then covers fewer drivers. The count under the gauge always says how many.

The rules

• Natural disasters (GDACS): orange and red alerts in the last 7 days. 25 points per red, 10 per orange, capped at 100.
• Space weather (NOAA): highest planetary Kp index in the last 24 hours. Quiet conditions score 6 points per Kp; storm levels G1 to G5 score 40 to 100.
• Pandemic and bio (WHO): Disease Outbreak News reports in the last 30 days, 10 points each, capped at 100.
• Geopolitical conflict (GDELT): news coverage of armed conflict over 7 days compared with its own 90-day normal. Normal scores 15; double the normal scores 75.
• Economic stress (U.S. Treasury OFR Financial Stress Index): 0 is average stress and scores 20; 5.5 scores 75; 8 or more scores 100.
• Cyber (CISA): actively exploited vulnerabilities added to CISA's catalogue in the last 30 days, 2 points each, capped at 100.

Limits

• It counts reports and alerts, not harm. One huge disaster and three small ones can score alike.
• Conflict is measured by news coverage, which rises and falls with media attention.
• The economic and cyber drivers use U.S. government sources, so they lean toward the U.S.
• The rules and caps are judgement calls, published so anyone can question them.