voidly
MM · per-country backtest

Myanmar forecast vs reality

Every Voidly Sentinel shutdown-risk forecast for Myanmar, plotted against what actually happened. 59 (predicted, observed) pairs from the rolling 30-day evaluation window.

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

Forecasts evaluated
59
since Apr 17
Accuracy @ 0.5
37.3%
22/59 correct
Brier score
0.393
lower is better
Observed positive rate
41%
mean predicted 11%

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 8

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

All 59 predictions (newest first)

Eval dateForecastPred ≥ 0.5?Observed?Correct?
Jun 84.9%no event
Jun 74.9%no event
Jun 63.2%no event
Jun 53.9%no event
Jun 44.8%shutdown
Jun 34.2%shutdown
Jun 23.2%shutdown
Jun 13.9%shutdown
May 3175.8%shutdown
May 313.8%shutdown
May 303.4%shutdown
May 294.2%shutdown
May 2860.0%no event
May 285.0%no event
May 2760.0%no event
May 275.4%no event
May 263.7%no event
May 254.9%no event
May 2413.3%no event
May 243.9%no event
May 2360.0%no event
May 232.9%no event
May 2245.6%no event
May 225.4%no event
May 2181.7%no event
May 206.5%no event
May 196.3%no event
May 187.7%no event
May 175.1%no event
May 163.1%no event
May 153.2%no event
May 145.0%no event
May 134.2%no event
May 122.4%no event
May 113.5%no event
May 105.1%no event
May 95.5%no event
May 85.0%no event
May 73.9%no event
May 63.4%no event
May 53.0%no event
May 45.0%no event
May 33.7%no event
May 23.1%shutdown
May 12.7%shutdown
Apr 304.1%shutdown
Apr 294.1%shutdown
Apr 286.6%shutdown
Apr 275.0%shutdown
Apr 264.0%shutdown
Apr 255.2%shutdown
Apr 243.9%shutdown
Apr 234.5%shutdown
Apr 225.3%shutdown
Apr 214.7%shutdown
Apr 204.7%shutdown
Apr 193.9%shutdown
Apr 184.5%shutdown
Apr 174.9%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