Forecast accuracy

Forecast accuracy scorecard

Name: World Monitor forecast resolution scorecard
Creator: World Monitor

World Monitor scores every forecast it publishes once the outcome is knowable, over a rolling 180-day window. This is the standing record: the scores, the calibration, the sample sizes, and the parts that are not measurable yet.

Record status

Availability: Captured: Read from the live scoring service on 2026-09-10 and frozen into this page.
Freshness: Current: The scoring service generated these numbers 12 hours before the capture read them. The threshold is 36 hours, matching the one the scoring service applies to itself.
Coverage: Measurable: The headline cohort has 180 scored forecasts, enough to publish a score.

Brier score, headline cohort0.118 180 scored forecasts

Log score, headline cohort0.375 180 scored forecasts

Brier score, every scored entry0.192 490 scored forecasts

Scored entries490 of 772 resolved

Over the current 180-day window, World Monitor's headline cohort scores a Brier of 0.118 across 180 scored forecasts, against 0.25 for answering 0.5 to everything.

Lower Brier is better. A Brier score is the mean squared error of a probability forecast, so 0 is perfect and answering 0.5 to everything scores 0.25. Log score is harsher on confident mistakes, and lower is better there too.

What the headline number counts

The headline Brier and log score cover 180 scored forecasts: every scored entry except those from the origins the scorecard excludes, currently bet_engine, state_derived. That exclusion costs 310 scored entries. An entry that carries no origin at all is filed as unknown and is counted, so this cohort is defined by what it leaves out and not by any property of the entries it keeps. The all-scored figure beside it is the same window with every origin put back, which is why the two numbers differ.

Resolution ledger

Resolution ledger totals for the rolling window. Voided entries are counted for coverage and excluded from every score below. A judging backlog is an ordinary state of the ledger, not a fault.
Ledger stage	Entries
Entries in the rolling window	958
Resolved	772
Scored	490
Voided	36.5% of 772 resolved entries (282 entries)
Awaiting a judge	96
Still open, not yet resolvable	90
Scored share of the ledger	51.1% of 958 entries

Brier/log score over resolved YES/NO published forecast windows; VOID and pending entries are counted for coverage but excluded from accuracy math.

Calibration

Calibration by predicted probability, computed over all 490 scored entries rather than the narrower headline cohort. Average predicted and Brier are probabilities, not rates. Buckets that scored nothing are omitted rather than shown as zero.
Predicted probability	Forecasts	Average predicted	Actually happened	Brier
0% to 10%	40	0.060	0.0% of 40 forecasts	0.005
10% to 20%	53	0.143	1.9% of 53 forecasts	0.034
20% to 30%	80	0.254	32.5% of 80 forecasts	0.227
30% to 40%	168	0.351	35.1% of 168 forecasts	0.227
40% to 50%	106	0.412	36.8% of 106 forecasts	0.234
50% to 60%	29	0.531	27.6% of 29 forecasts	0.266
60% to 70%	13	0.640	46.2% of 13 forecasts	0.274
90% to 100%	1	0.930	100.0% of 1 forecasts	0.005

Accuracy by domain

Accuracy by forecast domain, over every scored entry in that domain. A domain that resolved entries but scored none is marked as an insufficient sample, never left blank.
Domain	Resolved	Scored	Voided	Brier	Log score
conflict	36	12	66.7% of 36 resolved	0.273	0.730
cyber	146	144	1.4% of 146 resolved	0.076	0.278
energy	24	24	0.0% of 24 resolved	0.192	0.567
geopolitical	6	6	0.0% of 6 resolved	0.242	0.645
infrastructure	89	11	87.6% of 89 resolved	0.286	0.764
macro	5	5	0.0% of 5 resolved	0.235	0.662
market	346	272	21.4% of 346 resolved	0.242	0.679
military	20	6	70.0% of 20 resolved	0.270	0.731
political	6	0	100.0% of 6 resolved	Insufficient sample	Insufficient sample
supply_chain	94	10	89.4% of 94 resolved	0.237	0.666

Accuracy by generation origin

Accuracy by generation origin, and whether each origin counts toward the headline cohort. An entry with no recorded origin is filed as unknown and does count.
Generation origin	In the headline cohort	Resolved	Scored	Voided	Brier	Log score
bet_engine	No, excluded	299	299	0.0% of 299 resolved	0.236	0.664
legacy_detector	Yes	362	176	51.4% of 362 resolved	0.114	0.366
state_derived	No, excluded	78	11	85.9% of 78 resolved	0.241	0.676
unknown	Yes	33	4	87.9% of 33 resolved	0.289	0.773

Against prediction markets

Measured over all scored entries that overlapped a liquid market, not over the narrower headline cohort. On 78 such resolved questions the forecast Brier was 0.155 and the market Brier was 0.073. The published delta is the market Brier minus the forecast Brier, and lower is better, so a negative delta means the market scored better. Here the delta is -0.081, so on this sample the market scored better.

What this page does not publish

Calibration buckets that scored nothing are omitted from the table rather than drawn as a zero: 70-80, 80-90 are empty in this window.
No confidence intervals. Each figure is published with the number of forecasts behind it instead, because stating a sample size without an interval is honest and inventing an interval is not. Tracking: issue #7072.
No accuracy for the 24-hour, 7-day and 30-day projections shown in the product. Those horizons are not scored yet, so nothing here describes them. Tracking: issue #7075.
No individual forecasts, resolution evidence, judge inputs or archive locations. This page publishes aggregates only.

Open the live forecast panel in World Monitor →

Related reference

Download: scorecard.json. Source: docs/snapshots/crawlable-live-pulse-2026-09-09.json. Numbers generated 2026-09-10 06:02:24 UTC and read on 2026-09-10. Live results come from the credentialed forecast scorecard endpoint, which this page freezes so it can be read without one.