voidly
Forecast

Atlas forecast

Follow the
signal.

Explore possible changes in OONI measurements.
Choose a country. Inspect its record.

Seven-day event estimates

Source JSON

Unusual changes in a country’s OONI signals. An event estimate does not establish a full national shutdown or its exact date.

Board datehonest_forecast_v1
  1. EthiopiaET Elevated
    85.2%
  2. BelarusBY Elevated
    79.0%
  3. EgyptEG Elevated
    78.8%
  4. PakistanPK Elevated
    78.2%
  5. NigeriaNG Elevated
    75.3%
  6. SyriaSY Elevated
    69.0%
  7. Saudi ArabiaSA Watch
    65.1%
  8. KazakhstanKZ Watch
    60.3%
Highest 8 of 23 returned records for this model. Full 0–100% scale.Marks show each record’s supplied threshold. Bands come from the source.

The board does not supply per-country observation dates. Open a country’s source response to check its “as of” date.

All 23 event-model records
All returned honest_forecast_v1 records, sorted by the original event estimate.
CountryEvent estimateModelSource bandThresholdThreshold flagSource
EthiopiaET85.2%honest_forecast_v1Elevated67.5%AboveCountry JSON
BelarusBY79.0%honest_forecast_v1Elevated67.5%AboveCountry JSON
EgyptEG78.8%honest_forecast_v1Elevated67.5%AboveCountry JSON
PakistanPK78.2%honest_forecast_v1Elevated67.5%AboveCountry JSON
NigeriaNG75.3%honest_forecast_v1Elevated67.5%AboveCountry JSON
SyriaSY69.0%honest_forecast_v1Elevated67.5%AboveCountry JSON
Saudi ArabiaSA65.1%honest_forecast_v1Watch67.5%Not aboveCountry JSON
KazakhstanKZ60.3%honest_forecast_v1Watch67.5%Not aboveCountry JSON
UzbekistanUZ56.8%honest_forecast_v1Watch67.5%Not aboveCountry JSON
MyanmarMM45.4%honest_forecast_v1Watch67.5%Not aboveCountry JSON
CubaCU44.0%honest_forecast_v1Watch67.5%Not aboveCountry JSON
VenezuelaVE24.1%honest_forecast_v1Low67.5%Not aboveCountry JSON
SudanSD16.4%honest_forecast_v1Low67.5%Not aboveCountry JSON
VietnamVN8.8%honest_forecast_v1Low67.5%Not aboveCountry JSON
RussiaRU3.3%honest_forecast_v1Low67.5%Not aboveCountry JSON
IranIR3.1%honest_forecast_v1Low67.5%Not aboveCountry JSON
TurkeyTR1.8%honest_forecast_v1Low67.5%Not aboveCountry JSON
BangladeshBD1.3%honest_forecast_v1Low67.5%Not aboveCountry JSON
ChinaCN1.2%honest_forecast_v1Low67.5%Not aboveCountry JSON
IndiaIN0.8%honest_forecast_v1Low67.5%Not aboveCountry JSON
IndonesiaID0.3%honest_forecast_v1Low67.5%Not aboveCountry JSON
ThailandTH0.3%honest_forecast_v1Low67.5%Not aboveCountry JSON
TurkmenistanTM0.1%honest_forecast_v1Low67.5%Not aboveCountry JSON
7 records from other or unspecified models

These records remain separate because their model identity or forecast basis differs. Use the corresponding source notes and validation, not the event-model evaluation below.

Separate source records. Their outputs are not a comparable extension of the event-model ranking.
CountrySource valueModelSource bandThresholdThreshold flagSource
LebanonLB27.1%Not suppliedBand not supplied35.0%Not aboveCountry JSON
BrazilBR8.7%Not suppliedBand not supplied5.0%AboveCountry JSON
EritreaER5.1%Not suppliedBand not supplied5.0%AboveCountry JSON
NicaraguaNI4.8%Not suppliedBand not supplied27.5%Not aboveCountry JSON
PhilippinesPH4.2%Not suppliedBand not supplied5.0%Not aboveCountry JSON
North KoreaKP4.1%Not suppliedBand not supplied5.0%Not aboveCountry JSON
MalaysiaMY3.1%Not suppliedBand not supplied22.5%Not aboveCountry JSON

The earlier Sentinel v1 forecast is a current-regime signal with unreliable new-shutdown onset skill. A low output does not show that a country has an open internet.

Read the legacy forecast evaluation

Data service checked at .

Put the model
beside its baselines.

Performance on later data matters more than a good fit to the past.

Inspect the model card

Rolling-origin validation · precision–recall AUC

Event model0.279
Country baseline0.218
Recent-event baseline0.153

Higher is better on the same evaluation. PR-AUC is a ranking measure, not the percentage of correct predictions. The recent-event baseline counts event days in the seven-day input window.

Model card generated
Look across a region

See the wider
pattern.

Inspect regional aggregates and their model scope.

Follow the evidence

Find what
changed.

Explore recent observations and source records.

Understand the record.

What this model’s event means

The event model uses OONI measurement signals. Its target is an anomaly-rate spike relative to a country’s own recent baseline within the next seven days. It is not trained on the same target as the separate KeepItOn shutdown-risk model.

The model card specifies a 28-day baseline, a z-score threshold of 2, an anomaly-rate floor of 8.0%, and a minimum event-day measurement count of 50.

Stable blocking can produce a low change estimate. Sparse observations and differences between countries limit interpretation. Read standing censorship separately, and verify any incident against its evidence.

Evaluation and uncertainty

The comparison above uses strict temporal rolling-origin evaluation for honest_forecast_v1. Country history and measurement coverage matter; transferring to a new country is a different test.

Rolling-origin ROC AUC
0.824
Mean validation precision
29.5%
Mean validation recall
39.0%
Brier score
0.153
Leave-one-country-out mean AUC
0.742

Precision and recall are averages across validation folds, using thresholds selected from each fold’s training data. They do not describe a single current threshold.

These are evaluation results, not guarantees for an individual country. The heatmap supplies no confidence interval. No interval or exact-day trajectory is inferred for the dots above.

Existing country pages include legacy daily trajectories, attributions and interval fields. Check the model identity and source notes before applying them to the event-model headline.

Archived legacy-model movement

This is a separate comparison of archived v1 scores. It does not measure movement in the event-model probabilities above. Prior and current values use the same legacy basis.

Earlier snapshot:

Later snapshot
Returned legacy v1 movers across 7 days. Only records with consistent source arithmetic are shown. The source request filters changes smaller than 2 percentage points.
CountryEarlier v1 scoreLater v1 scoreChange
Myanmar73.5%31.7%-41.8 pp
Ethiopia29.1%70.4%+41.3 pp
Russia36.9%1.0%-35.9 pp
Cuba76.5%44.0%-32.5 pp
Iran32.9%3.6%-29.2 pp
Brazil32.0%8.7%-23.3 pp
Kazakhstan66.9%50.4%-16.5 pp
Uzbekistan43.5%58.8%+15.3 pp
Nigeria52.7%67.3%+14.6 pp
Pakistan67.4%81.7%+14.3 pp
Sudan5.5%16.4%+10.9 pp
Belarus69.2%79.8%+10.6 pp

Model recalibration can change scores without establishing a change on the network. Raw source notes distinguish the legacy deltas from the newer headline values.

Related forecasts, alerts and reuse

Election context, case studies and legacy trajectories have their own scope. A worked alert example does not establish forecast timing skill.

Voidly-original records and annotations are CC BY 4.0. Raw upstream measurements retain their source-specific license. Read the data terms before reuse.

Design previewExplore alert settings