Data confidence
Confidence ≠ calm. Confidence ≠ danger. Confidence = coverage strength.
It measures how much the system can see, not what it concludes. Pair confidence with pressure scores for interpretation.
📡 Coverage depth
🧾 Source reliability
⏱️ Recency
What confidence means
- High confidence: broad, recent, and reliable coverage.
- Low confidence: sparse or stale signals, even if the score looks calm.
Coverage inputs
- Source reliability: verified vs emerging feeds.
- Recency: fresh signals indicate active observation.
- Density: multiple independent sources reduce blind spots.
Example interpretations
Calm + low confidence → not enough visibility to confirm stability.
Unstable + high confidence → sustained pressure is well observed.
Tiers
| Tier | Meaning | How to use it |
|---|---|---|
| High | Strong coverage across sources and time | Trust the pressure signal as representative. |
| Medium | Partial coverage or mixed reliability | Use as directional guidance, confirm with local context. |
| Low | Sparse coverage or limited reliability | Treat pressure as under-observed and verify externally. |
Limits
- Confidence does not correct for total news blackouts.
- It does not infer intent or sentiment.
- It only reflects observable coverage.
Confidence helps avoid whiplash from single-source spikes and flags coverage gaps.
Back to the live map
Compare confidence tiers with current pressure signals.