voidly
SA · per-country backtest

Saudi Arabia forecast vs reality

Every Voidly Sentinel shutdown-risk forecast for Saudi Arabia, plotted against what actually happened. 112 (predicted, observed) pairs from the rolling 30-day evaluation window.

Updated every 30 min · CC BY 4.0 · Raw JSON · Current forecast →

Forecasts evaluated
112
since Apr 17
Accuracy @ 0.5
40.2%
45/112 correct
Brier score
0.300
lower is better
Observed positive rate
19%
mean predicted 33%

Forecast time series

Blue line: forecast probability. Green ✓ markers: forecast was right. Red ✗ markers: forecast was wrong. Dashed line at 0.5 is the binary decision threshold.

0.000.250.500.751.00threshold 0.50Apr 17Aug 6

Y axis: forecast probability · Faint tick: distance to observed outcome (0 or 1)

All 112 predictions (newest first)

Eval dateForecastPred ≥ 0.5?Observed?Correct?
Aug 679.2%no event
Aug 572.1%no event
Aug 478.5%no event
Aug 381.3%no event
Aug 279.1%no event
Aug 179.8%no event
Jul 3179.5%shutdown
Jul 3078.8%shutdown
Jul 2978.8%shutdown
Jul 2866.3%shutdown
Jul 2763.2%shutdown
Jul 2671.8%no event
Jul 2572.4%no event
Jul 2474.0%no event
Jul 2373.9%no event
Jul 2273.1%no event
Jul 2173.5%no event
Jul 2071.0%no event
Jul 1976.1%no event
Jul 1873.3%no event
Jul 1775.6%no event
Jul 1675.8%no event
Jul 1580.5%no event
Jul 1473.6%no event
Jul 1361.4%no event
Jul 1270.1%no event
Jul 1161.1%no event
Jul 1066.0%no event
Jul 964.0%no event
Jul 856.1%no event
Jul 763.8%no event
Jul 657.6%no event
Jul 557.5%no event
Jul 462.6%no event
Jul 359.4%no event
Jul 260.9%no event
Jul 169.1%no event
Jun 3066.9%shutdown
Jun 2963.4%shutdown
Jun 2861.7%no event
Jun 2765.6%no event
Jun 2656.4%no event
Jun 2565.5%no event
Jun 2464.5%no event
Jun 2371.0%no event
Jun 2264.7%no event
Jun 214.3%no event
Jun 204.4%no event
Jun 193.4%no event
Jun 183.2%no event
Jun 174.8%no event
Jun 163.2%no event
Jun 153.3%no event
Jun 144.7%no event
Jun 134.8%no event
Jun 123.8%no event
Jun 113.6%no event
Jun 105.4%no event
Jun 94.8%no event
Jun 84.7%no event

How to read this

  • Each row is one historical forecast. We made the prediction at eval_date, then waited the 7-day horizon, then graded against the observed outcome.
  • Forecast % is the calibrated probability we published that day. Post-recalibration (2026-05-20) these now match actual observed rates much better — see the refit finding.
  • Pred ≥ 0.5 shows the binary alert decision. Note we usually fire alerts at a lower threshold (see /v1/sentinel/global_heatmap for the live cutoff) — the 0.5 column here is for backtest scoring consistency with the global confusion matrix.
  • Correct = both (pred ≥ 0.5) and (observed) agree.

Related