voidly
TR · per-country backtest

Turkey 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 4

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
141
since Apr 17
Accuracy @ 0.5
50.4%
71/141 correct
Brier score
0.314
lower is better
Observed positive rate
33%
mean predicted 4%

All 141 predictions (newest first)

Eval dateForecastPred ≥ 0.5?Observed?Correct?
Sep 41.6%shutdown
Sep 31.8%shutdown
Sep 21.8%shutdown
Sep 11.5%shutdown
Aug 315.0%shutdown
Aug 301.9%shutdown
Aug 292.3%shutdown
Aug 282.1%shutdown
Aug 271.5%no event
Aug 261.5%no event
Aug 251.5%no event
Aug 241.3%no event
Aug 230.7%no event
Aug 220.5%no event
Aug 210.7%no event
Aug 201.2%no event
Aug 190.6%no event
Aug 180.6%no event
Aug 170.6%no event
Aug 161.1%no event
Aug 151.1%no event
Aug 141.1%no event
Aug 131.4%no event
Aug 120.7%no event
Aug 110.7%shutdown
Aug 100.8%shutdown
Aug 91.0%shutdown
Aug 80.4%shutdown
Aug 70.3%shutdown
Aug 60.1%no event
Aug 50.0%no event
Aug 40.0%shutdown
Aug 30.0%shutdown
Aug 20.1%shutdown
Aug 10.2%shutdown
Jul 310.3%shutdown
Jul 300.5%shutdown
Jul 290.9%shutdown
Jul 280.9%shutdown
Jul 270.8%shutdown
Jul 261.0%no event
Jul 253.1%shutdown
Jul 243.1%shutdown
Jul 233.3%shutdown
Jul 222.9%shutdown
Jul 212.3%shutdown
Jul 205.3%no event
Jul 191.4%no event
Jul 181.3%shutdown
Jul 171.6%shutdown
Jul 161.7%shutdown
Jul 151.6%shutdown
Jul 141.9%shutdown
Jul 132.1%no event
Jul 122.0%no event
Jul 111.3%no event
Jul 101.1%no event
Jul 91.6%no event
Jul 80.9%no event
Jul 70.6%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.
About this view

Every Voidly Sentinel shutdown-risk forecast for Turkey, plotted against what actually happened. 141 (predicted, observed) pairs from the rolling 30-day evaluation window.