voidly
MM · per-country backtest

Myanmar forecast vs reality

Every Voidly Sentinel shutdown-risk forecast for Myanmar, plotted against what actually happened. 111 (predicted, observed) pairs from the rolling 30-day evaluation window.

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

Forecasts evaluated
111
since Apr 17
Accuracy @ 0.5
54.1%
60/111 correct
Brier score
0.302
lower is better
Observed positive rate
42%
mean predicted 28%

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 111 predictions (newest first)

Eval dateForecastPred ≥ 0.5?Observed?Correct?
Aug 669.9%no event
Aug 567.0%no event
Aug 465.8%no event
Aug 371.2%no event
Aug 279.1%no event
Aug 174.1%no event
Jul 3166.4%no event
Jul 2960.3%no event
Jul 2866.0%shutdown
Jul 2770.0%shutdown
Jul 2668.5%shutdown
Jul 2568.6%shutdown
Jul 2468.9%shutdown
Jul 2364.3%shutdown
Jul 2272.9%shutdown
Jul 2176.6%shutdown
Jul 204.7%shutdown
Jul 1965.6%shutdown
Jul 1864.4%shutdown
Jul 1762.2%no event
Jul 1662.9%no event
Jul 1563.0%shutdown
Jul 1461.8%shutdown
Jul 1358.8%shutdown
Jul 1259.7%shutdown
Jul 1168.4%shutdown
Jul 1069.5%no event
Jul 984.8%no event
Jul 85.8%no event
Jul 767.6%no event
Jul 670.2%no event
Jul 568.8%no event
Jul 461.7%shutdown
Jul 360.7%shutdown
Jul 264.9%shutdown
Jul 162.1%shutdown
Jun 3058.0%shutdown
Jun 2949.9%shutdown
Jun 2846.5%shutdown
Jun 2746.1%shutdown
Jun 2658.8%shutdown
Jun 2548.2%no event
Jun 2446.3%no event
Jun 2356.5%no event
Jun 2246.4%no event
Jun 212.7%no event
Jun 203.8%no event
Jun 194.4%no event
Jun 183.8%no event
Jun 174.1%no event
Jun 164.4%no event
Jun 153.6%no event
Jun 143.1%no event
Jun 134.7%no event
Jun 124.4%no event
Jun 114.4%no event
Jun 104.4%no event
Jun 93.6%no event
Jun 84.9%no event
Jun 74.9%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