voidly
IR · per-country backtest

Iran forecast vs reality

Every Voidly Sentinel shutdown-risk forecast for Iran, plotted against what actually happened. 59 (predicted, observed) pairs from the rolling 30-day evaluation window.

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

Forecasts evaluated
59
since Apr 17
Accuracy @ 0.5
69.5%
41/59 correct
Brier score
0.654
lower is better
Observed positive rate
88%
mean predicted 16%

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 59 predictions (newest first)

Eval dateForecastPred ≥ 0.5?Observed?Correct?
Jun 85.0%shutdown
Jun 74.0%shutdown
Jun 63.9%shutdown
Jun 54.0%shutdown
Jun 43.9%shutdown
Jun 33.8%shutdown
Jun 25.2%no event
Jun 113.7%shutdown
May 3140.4%shutdown
May 3114.8%shutdown
May 3014.8%shutdown
May 2918.0%shutdown
May 2880.0%shutdown
May 2873.7%shutdown
May 2815.6%shutdown
May 2780.0%shutdown
May 2717.7%shutdown
May 2616.1%shutdown
May 2516.9%shutdown
May 2414.2%shutdown
May 23100.0%shutdown
May 2313.0%shutdown
May 2226.0%shutdown
May 2212.0%shutdown
May 2187.0%shutdown
May 2011.4%shutdown
May 1912.2%no event
May 1811.4%no event
May 1710.6%shutdown
May 168.7%shutdown
May 157.1%shutdown
May 146.1%shutdown
May 136.1%shutdown
May 128.4%shutdown
May 117.1%shutdown
May 107.2%shutdown
May 96.4%shutdown
May 86.7%shutdown
May 74.8%shutdown
May 67.3%no event
May 59.4%no event
May 49.5%no event
May 35.8%no event
May 28.6%shutdown
May 18.6%shutdown
Apr 309.1%shutdown
Apr 299.6%shutdown
Apr 289.5%shutdown
Apr 277.9%shutdown
Apr 268.9%shutdown
Apr 257.6%shutdown
Apr 245.1%shutdown
Apr 237.9%shutdown
Apr 226.7%shutdown
Apr 216.4%shutdown
Apr 206.1%shutdown
Apr 194.9%shutdown
Apr 184.0%shutdown
Apr 176.1%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