voidly
PK · per-country backtest

Pakistan forecast vs reality

Every Voidly Sentinel shutdown-risk forecast for Pakistan, plotted against what actually happened. 61 (predicted, observed) pairs from the rolling 30-day evaluation window.

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

Forecasts evaluated
61
since Apr 17
Accuracy @ 0.5
57.4%
35/61 correct
Brier score
0.549
lower is better
Observed positive rate
100%
mean predicted 40%

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 17Jun 8

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

All 61 predictions (newest first)

Eval dateForecastPred ≥ 0.5?Observed?Correct?
Jun 895.0%↑shutdown✓
Jun 795.0%↑shutdown✓
Jun 695.0%↑shutdown✓
Jun 595.0%↑shutdown✓
Jun 495.0%↑shutdown✓
Jun 395.0%↑shutdown✓
Jun 295.0%↑shutdown✓
Jun 195.0%↑shutdown✓
May 3126.1%↑shutdown✓
May 3195.0%↑shutdown✓
May 3095.0%↑shutdown✓
May 2995.0%↑shutdown✓
May 28100.0%↑shutdown✓
May 2824.7%↑shutdown✓
May 2895.0%↑shutdown✓
May 27100.0%↑shutdown✓
May 2795.0%↑shutdown✓
May 2695.0%↑shutdown✓
May 2522.6%↑shutdown✓
May 2595.0%↑shutdown✓
May 2414.8%↑shutdown✓
May 2495.0%↑shutdown✓
May 23100.0%↑shutdown✓
May 2395.0%↑shutdown✓
May 2295.0%↑shutdown✓
May 2180.0%↑shutdown✓
May 2181.5%↑shutdown✓
May 207.6%↑shutdown✓
May 196.9%↑shutdown✓
May 186.3%↑shutdown✓
May 173.3%—shutdown✗
May 164.4%—shutdown✗
May 155.0%↑shutdown✓
May 143.8%—shutdown✗
May 133.9%—shutdown✗
May 124.9%—shutdown✗
May 114.5%—shutdown✗
May 104.4%—shutdown✗
May 93.6%—shutdown✗
May 85.1%↑shutdown✓
May 74.0%—shutdown✗
May 64.5%—shutdown✗
May 55.3%↑shutdown✓
May 44.6%—shutdown✗
May 33.9%—shutdown✗
May 24.0%—shutdown✗
May 13.1%—shutdown✗
Apr 304.6%—shutdown✗
Apr 294.9%—shutdown✗
Apr 284.4%—shutdown✗
Apr 274.9%—shutdown✗
Apr 264.5%—shutdown✗
Apr 255.1%↑shutdown✓
Apr 244.6%—shutdown✗
Apr 235.3%↑shutdown✓
Apr 224.9%—shutdown✗
Apr 214.8%—shutdown✗
Apr 204.4%—shutdown✗
Apr 193.2%—shutdown✗
Apr 184.0%—shutdown✗

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