Turkey — 13% peak risk
Calibrated via isotonic refit on 810 live outcomes (2026-05-20). See calibration drift → · Turkey country profile →
The Voidly Sentinel model forecasts a peak shutdown probability of 12.5% for Turkey over the next 7 days, with the highest risk on day 7 (Thu, May 28). Average risk across the window: 8%.
90% conformal interval: [0%, 48%] — empirical coverage 90.4% (target 90%).
Per-day risk
Dashed line: 50% threshold. Bars colored by risk band.
Per-day drivers
| Date | Risk | Drivers |
|---|---|---|
| Thu, May 21 | 3% · Low | Turkish Local Elections 2026 in 11d |
| Fri, May 22 | 6% · Low | Turkish Local Elections 2026 in 10d |
| Sat, May 23 | 5% · Low | Turkish Local Elections 2026 in 9d |
| Sun, May 24 | 7% · Low | Turkish Local Elections 2026 in 8d |
| Mon, May 25 | 8% · Low | Turkish Local Elections 2026 in 7d |
| Tue, May 26 | 10% · Low | Turkish Local Elections 2026 in 6d |
| Wed, May 27 | 10% · Low | Turkish Local Elections 2026 in 5d |
| Thu, May 28 | 13% · Low | Turkish Local Elections 2026 in 4d |
What drove the prediction
Top features by signed contribution to today's risk score. Positive (↑) features push the prediction higher; negative (↓) features push it lower.
- ↑
gdelt_unrest_30d+0.081 - ↓
block_rate_lag14-0.052 - ↓
high_urgency_signals_7d-0.045
Similar past event
The model found a similar event in the corpus. Reading the incident report tells you what “13% risk for Turkey” looked like when it actually happened.
disruption in TR · Sun, May 3
Recent evidence
Live measurements from upstream networks supporting this forecast. Click any link to view the raw data on the source.
- ooniheader-manipulationhttps://explorer.ooni.org/search?probe_cc=TR&test_name=http_header_field_manipulation&since=2026-05-20&until=2026-05-20&only=anomalies
- oonimiddlebox-detectionhttps://explorer.ooni.org/search?probe_cc=TR&test_name=http_invalid_request_line&since=2026-05-20&until=2026-05-20&only=anomalies
- oonitor-blockinghttps://explorer.ooni.org/search?probe_cc=TR&test_name=tor&since=2026-05-20&until=2026-05-20&only=anomalies
Honest caveats
- The model's base AUC on a time-based holdout is
0.50(random on novel events). LOCO median AUC is0.91. The calibrated number you see above is the most-honest operational figure but it's NOT a guarantee. - Calibration was refit on 2026-05-20 on 810 live outcomes — it's good for the 30 watched censoring countries but will drift if the world changes. Nightly recalibration runs at 04:30 UTC.
- A 70% forecast doesn't mean a shutdown is guaranteed; a 20% forecast doesn't mean it's safe. The 90% conformal interval is the honest uncertainty band.
- For long-form methodology see /methodology and the calibration refit finding.
Model version: v1@2026-05-21T04:20:38.966951
Related
- /tr — full country profile with recent incidents + ISP breakdown
- /atlas/elections — all 90-day election forecasts ranked by risk
- /sentinel/calibration — live model-honesty dashboard + feature importance
- /v1/sentinel/current_risk/TR — raw JSON