voidly
TM · per-country backtest

Turkmenistan forecast vs reality

Every Voidly Sentinel shutdown-risk forecast for Turkmenistan, plotted against what actually happened. 73 (predicted, observed) pairs from the rolling 30-day evaluation window.

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

Forecasts evaluated
73
since Apr 17
Accuracy @ 0.5
65.8%
48/73 correct
Brier score
0.100
lower is better
Observed positive rate
10%
mean predicted 6%

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 28

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

All 73 predictions (newest first)

Eval dateForecastPred ≥ 0.5?Observed?Correct?
Jun 280.1%—no event✓
Jun 270.1%—no event✓
Jun 260.1%—no event✓
Jun 250.1%—no event✓
Jun 240.1%—no event✓
Jun 230.1%—no event✓
Jun 220.1%—no event✓
Jun 214.8%—no event✓
Jun 204.3%—no event✓
Jun 194.0%—no event✓
Jun 184.8%—no event✓
Jun 174.9%—no event✓
Jun 164.3%—no event✓
Jun 154.1%—no event✓
Jun 143.9%—no event✓
Jun 132.9%—no event✓
Jun 124.3%—no event✓
Jun 112.7%—no event✓
Jun 103.7%—no event✓
Jun 95.0%↑no event✗
Jun 84.0%—no event✓
Jun 75.5%↑no event✗
Jun 63.6%—no event✓
Jun 53.5%—no event✓
Jun 45.1%↑no event✗
Jun 33.5%—no event✓
Jun 24.2%—no event✓
Jun 14.4%—no event✓
May 313.7%—no event✓
May 303.5%—no event✓
May 293.6%—no event✓
May 283.4%—no event✓
May 275.5%↑no event✗
May 263.0%—no event✓
May 254.5%—no event✓
May 2459.5%↑no event✗
May 2361.7%↑no event✗
May 2219.5%↑no event✗
May 216.6%↑no event✗
May 207.3%↑no event✗
May 197.8%↑no event✗
May 187.8%↑no event✗
May 175.1%↑no event✗
May 162.0%—no event✓
May 152.4%—no event✓
May 143.8%—shutdown✗
May 135.4%↑shutdown✓
May 124.4%—shutdown✗
May 114.3%—shutdown✗
May 103.4%—shutdown✗
May 94.5%—shutdown✗
May 84.9%—shutdown✗
May 75.2%↑no event✗
May 64.7%—no event✓
May 55.2%↑no event✗
May 45.0%↑no event✗
May 34.0%—no event✓
May 24.5%—no event✓
May 12.9%—no event✓
Apr 302.7%—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