voidly
TR · per-country backtest

Turkey forecast vs reality

Every Voidly Sentinel shutdown-risk forecast for Turkey, 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
50.7%
37/73 correct
Brier score
0.154
lower is better
Observed positive rate
16%
mean predicted 7%

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 280.3%—no event✓
Jun 270.3%—no event✓
Jun 260.3%—no event✓
Jun 250.5%—no event✓
Jun 240.7%—no event✓
Jun 230.7%—no event✓
Jun 220.7%—no event✓
Jun 213.3%—no event✓
Jun 204.7%—no event✓
Jun 193.6%—no event✓
Jun 183.1%—no event✓
Jun 174.2%—no event✓
Jun 164.6%—no event✓
Jun 154.5%—no event✓
Jun 144.7%—no event✓
Jun 133.3%—no event✓
Jun 123.8%—no event✓
Jun 114.1%—no event✓
Jun 105.1%↑no event✗
Jun 94.7%—no event✓
Jun 84.5%—no event✓
Jun 73.9%—no event✓
Jun 63.4%—no event✓
Jun 54.9%—no event✓
Jun 43.2%—no event✓
Jun 34.9%—no event✓
Jun 24.0%—no event✓
Jun 114.7%↑no event✗
May 3115.8%↑no event✗
May 3017.6%↑no event✗
May 2917.2%↑no event✗
May 2818.5%↑no event✗
May 2717.8%↑no event✗
May 2616.8%↑no event✗
May 2519.8%↑no event✗
May 2413.7%↑no event✗
May 2314.5%↑no event✗
May 2212.9%↑no event✗
May 2113.1%↑no event✗
May 2011.6%↑no event✗
May 1912.7%↑no event✗
May 1812.5%↑no event✗
May 178.8%↑no event✗
May 167.5%↑no event✗
May 157.9%↑no event✗
May 146.9%↑no event✗
May 138.5%↑no event✗
May 127.1%↑no event✗
May 118.2%↑no event✗
May 105.9%↑no event✗
May 97.9%↑no event✗
May 86.1%↑no event✗
May 76.1%↑no event✗
May 66.8%↑no event✗
May 57.0%↑no event✗
May 45.3%↑no event✗
May 36.3%↑no event✗
May 26.6%↑shutdown✓
May 16.5%↑shutdown✓
Apr 307.0%↑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