voidly
KZ · per-country backtest

Kazakhstan forecast vs reality

Every Voidly Sentinel shutdown-risk forecast for Kazakhstan, 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
39.3%
44/112 correct
Brier score
0.297
lower is better
Observed positive rate
18%
mean predicted 31%

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 676.6%no event
Aug 568.2%no event
Aug 476.8%shutdown
Aug 375.4%shutdown
Aug 267.9%shutdown
Aug 160.4%shutdown
Jul 3158.8%shutdown
Jul 3056.4%no event
Jul 2937.0%no event
Jul 2838.3%no event
Jul 2758.0%no event
Jul 2661.6%no event
Jul 2561.8%no event
Jul 2462.4%no event
Jul 2365.8%no event
Jul 2264.9%no event
Jul 2158.4%no event
Jul 204.9%no event
Jul 1963.2%no event
Jul 1860.9%no event
Jul 1762.9%no event
Jul 1663.7%no event
Jul 1560.6%no event
Jul 1452.5%no event
Jul 1354.3%no event
Jul 1261.8%no event
Jul 1162.4%no event
Jul 1057.4%no event
Jul 961.2%no event
Jul 861.2%no event
Jul 760.9%no event
Jul 667.9%no event
Jul 567.0%no event
Jul 469.8%no event
Jul 370.6%no event
Jul 272.3%no event
Jul 177.0%no event
Jun 3073.0%no event
Jun 2974.2%no event
Jun 2875.6%no event
Jun 2776.5%no event
Jun 2677.5%no event
Jun 2582.1%no event
Jun 2483.0%no event
Jun 2379.7%no event
Jun 2273.2%no event
Jun 214.6%shutdown
Jun 204.9%shutdown
Jun 195.6%shutdown
Jun 183.9%shutdown
Jun 175.0%no event
Jun 165.2%no event
Jun 155.3%no event
Jun 144.6%no event
Jun 134.7%no event
Jun 125.2%no event
Jun 114.0%no event
Jun 104.8%no event
Jun 94.9%no event
Jun 85.9%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