voidly
IR · per-country backtest

Iran forecast vs reality

Compare published forecasts with observed outcomes for this country.

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

Forecast time series

0.000.250.500.751.00threshold 0.50Apr 17Sep 7

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

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

Forecasts evaluated
144
since Apr 17
Accuracy @ 0.5
51.4%
74/144 correct
Brier score
0.415
lower is better
Observed positive rate
56%
mean predicted 23%

All 144 predictions (newest first)

Eval dateForecastPred ≥ 0.5?Observed?Correct?
Sep 732.4%no event
Sep 658.3%no event
Sep 563.8%no event
Sep 471.2%no event
Sep 365.6%no event
Sep 255.3%no event
Sep 137.5%no event
Aug 3127.9%shutdown
Aug 3061.8%no event
Aug 2949.3%no event
Aug 2856.4%no event
Aug 2756.4%no event
Aug 2646.0%no event
Aug 2543.5%no event
Aug 2440.3%no event
Aug 2340.9%no event
Aug 2237.9%no event
Aug 2133.4%no event
Aug 2034.8%no event
Aug 1952.8%no event
Aug 1848.5%no event
Aug 1764.0%no event
Aug 1627.5%shutdown
Aug 1537.5%shutdown
Aug 1432.3%shutdown
Aug 1352.0%shutdown
Aug 1240.5%no event
Aug 1125.4%no event
Aug 1031.0%no event
Aug 924.2%no event
Aug 829.5%no event
Aug 732.6%no event
Aug 633.4%no event
Aug 525.1%no event
Aug 413.9%no event
Aug 319.0%no event
Aug 213.2%no event
Aug 127.0%no event
Jul 3129.8%shutdown
Jul 3012.8%shutdown
Jul 2916.3%shutdown
Jul 2811.8%shutdown
Jul 2718.7%shutdown
Jul 2627.8%shutdown
Jul 2530.4%shutdown
Jul 2424.2%shutdown
Jul 2323.7%shutdown
Jul 2223.7%shutdown
Jul 2125.6%shutdown
Jul 2021.3%shutdown
Jul 1935.0%shutdown
Jul 1827.1%shutdown
Jul 1723.0%shutdown
Jul 1630.3%shutdown
Jul 1519.7%shutdown
Jul 1434.7%shutdown
Jul 1331.9%shutdown
Jul 1245.7%shutdown
Jul 1140.9%shutdown
Jul 1063.7%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.
About this view

Every Voidly Sentinel shutdown-risk forecast for Iran, plotted against what actually happened. 144 (predicted, observed) pairs from the rolling 30-day evaluation window.