voidly
VE · per-country backtest

Venezuela forecast vs reality

Every Voidly Sentinel shutdown-risk forecast for Venezuela, plotted against what actually happened. 66 (predicted, observed) pairs from the rolling 30-day evaluation window.

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

Forecasts evaluated
66
since Apr 17
Accuracy @ 0.5
68.2%
45/66 correct
Brier score
0.758
lower is better
Observed positive rate
100%
mean predicted 16%

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 12

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

All 66 predictions (newest first)

Eval dateForecastPred ≥ 0.5?Observed?Correct?
Jun 125.2%shutdown
Jun 113.7%shutdown
Jun 105.9%shutdown
Jun 94.2%shutdown
Jun 84.7%shutdown
Jun 74.6%shutdown
Jun 64.6%shutdown
Jun 55.3%shutdown
Jun 45.3%shutdown
Jun 35.2%shutdown
Jun 25.4%shutdown
Jun 15.9%shutdown
May 3137.0%shutdown
May 314.9%shutdown
May 3034.9%shutdown
May 303.4%shutdown
May 295.4%shutdown
May 2860.0%shutdown
May 283.4%shutdown
May 2760.0%shutdown
May 275.1%shutdown
May 2618.9%shutdown
May 263.9%shutdown
May 253.4%shutdown
May 2421.8%shutdown
May 2476.6%shutdown
May 2380.0%shutdown
May 2367.9%shutdown
May 2280.0%shutdown
May 2231.8%shutdown
May 2218.5%shutdown
May 2182.1%shutdown
May 2033.5%shutdown
May 1932.7%shutdown
May 1833.0%shutdown
May 175.3%shutdown
May 163.8%shutdown
May 157.9%shutdown
May 147.7%shutdown
May 136.7%shutdown
May 127.1%shutdown
May 118.2%shutdown
May 105.3%shutdown
May 96.4%shutdown
May 86.7%shutdown
May 76.1%shutdown
May 65.7%shutdown
May 57.3%shutdown
May 47.6%shutdown
May 34.9%shutdown
May 25.7%shutdown
May 13.4%shutdown
Apr 305.4%shutdown
Apr 296.1%shutdown
Apr 284.4%shutdown
Apr 277.5%shutdown
Apr 265.4%shutdown
Apr 255.1%shutdown
Apr 244.2%shutdown
Apr 235.0%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