voidly
NG · per-country backtest

Nigeria forecast vs reality

Every Voidly Sentinel shutdown-risk forecast for Nigeria, 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
33.9%
38/112 correct
Brier score
0.113
lower is better
Observed positive rate
0%
mean predicted 23%

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 663.8%no event
Aug 579.0%no event
Aug 463.0%no event
Aug 353.3%no event
Aug 274.2%no event
Aug 178.7%no event
Jul 3177.0%no event
Jul 3081.0%no event
Jul 2973.8%no event
Jul 2880.4%no event
Jul 2777.8%no event
Jul 2650.9%no event
Jul 2540.3%no event
Jul 2447.3%no event
Jul 2351.7%no event
Jul 2246.6%no event
Jul 2131.3%no event
Jul 2010.0%no event
Jul 1914.5%no event
Jul 1814.4%no event
Jul 1713.9%no event
Jul 1620.0%no event
Jul 1526.3%no event
Jul 1435.5%no event
Jul 1352.2%no event
Jul 1239.7%no event
Jul 1140.5%no event
Jul 1052.1%no event
Jul 953.8%no event
Jul 834.8%no event
Jul 716.7%no event
Jul 629.1%no event
Jul 523.3%no event
Jul 432.4%no event
Jul 338.9%no event
Jul 231.9%no event
Jul 129.2%no event
Jun 3047.3%no event
Jun 2912.6%no event
Jun 2814.6%no event
Jun 2728.4%no event
Jun 2617.7%no event
Jun 2518.6%no event
Jun 2425.0%no event
Jun 2340.5%no event
Jun 2217.3%no event
Jun 216.1%no event
Jun 206.0%no event
Jun 194.0%no event
Jun 184.7%no event
Jun 174.5%no event
Jun 165.0%no event
Jun 154.8%no event
Jun 143.6%no event
Jun 135.3%no event
Jun 124.0%no event
Jun 113.8%no event
Jun 103.9%no event
Jun 94.1%no event
Jun 86.1%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