voidly
MY · per-country backtest

Malaysia forecast vs reality

Every Voidly Sentinel shutdown-risk forecast for Malaysia, 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
61.6%
69/112 correct
Brier score
0.166
lower is better
Observed positive rate
19%
mean predicted 12%

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 64.9%no event
Aug 55.4%no event
Aug 44.1%shutdown
Aug 34.3%shutdown
Aug 25.3%shutdown
Aug 13.3%shutdown
Jul 3134.2%shutdown
Jul 3032.8%shutdown
Jul 2933.3%shutdown
Jul 2834.9%shutdown
Jul 2729.4%shutdown
Jul 2633.1%shutdown
Jul 2533.3%shutdown
Jul 2434.5%shutdown
Jul 2334.4%no event
Jul 2230.8%no event
Jul 2132.8%no event
Jul 2031.0%no event
Jul 1933.9%no event
Jul 1832.9%no event
Jul 1732.2%no event
Jul 1633.4%no event
Jul 1530.5%no event
Jul 1433.2%no event
Jul 1330.2%no event
Jul 1232.6%no event
Jul 1129.4%no event
Jul 1030.3%no event
Jul 95.4%no event
Jul 84.6%no event
Jul 728.1%no event
Jul 629.4%no event
Jul 53.9%no event
Jul 45.1%no event
Jul 35.1%no event
Jul 25.9%no event
Jul 15.5%no event
Jun 304.5%no event
Jun 294.3%no event
Jun 285.7%no event
Jun 274.4%no event
Jun 264.4%no event
Jun 253.4%no event
Jun 244.2%no event
Jun 233.9%no event
Jun 224.5%no event
Jun 214.1%no event
Jun 204.1%no event
Jun 195.7%no event
Jun 184.0%no event
Jun 174.6%no event
Jun 165.1%no event
Jun 154.1%no event
Jun 144.6%no event
Jun 134.5%no event
Jun 125.2%no event
Jun 113.7%no event
Jun 105.3%no event
Jun 93.8%no event
Jun 83.3%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