Public forecast 012 · permanent record

Will the ONS headline UK unemployment rate for the three months to July 2026, as first published, be between 4.7% and 5.1% inclusive?

RESOLVED — scored against the source named before the outcome was known.

Reality replay

The forecast as it was frozen, then what happened, then the score. The first panel shows only what was fixed when the forecast was frozen.

Predicted then

frozen 2026-08-30 17:00 UTC

86–94%

the chance, given at issue, that the frozen criterion resolves TRUE

Known now

resolved 2026-09-16 18:06 UTC

TRUE

4.9%, as first published by the source named in advance

The score

  • This forecast: Brier 0.01, scored against band midpoint 0.90
  • Always saying 50%: 0.25, a reference and not a forecast anyone made
  • The scale runs from 0, a perfect forecast, to 1, certain and wrong

The scored probability, 90%, leaned towards TRUE and the outcome was TRUE, so it scores better than always saying 50%.

One outcome is not a track record. No accuracy figure is claimed until 10 outcomes have resolved in a domain; the full register keeps the count.

Question
Will the ONS headline UK unemployment rate for the three months to July 2026, as first published, be between 4.7% and 5.1% inclusive?
Probability band
86–94%
Issued
2026-08-30 17:00 UTC
Data cut-off
2026-08-30 17:00 UTC
Frozen criterion
Resolves TRUE if the headline UK unemployment rate (aged 16 and over, three months May to July 2026), in the figure first published by the ONS in the Labour market overview bulletin scheduled for 15 September 2026, is greater than or equal to 4.7% and less than or equal to 5.1%. Resolves FALSE otherwise. Scored against the figure as first published, not later revisions. If the bulletin is postponed beyond 30 September 2026 the forecast is marked withdrawn with that reason.
Resolves by
2026-09-15 08:00 UTC
Origin
Editorial forecast — operator judgement, scored on the editorial board; not evidence for the decision models — reasoning recorded at issue, unchanged since
Ledger id
lxf_external_012
Sealed receipt
2c84cf290170da82e82578c120af7a879d3e4dd888e56c185babf5bac5e04931 — SHA-256 over the entry’s frozen fields at issue; how to check it
Outcome
TRUE — 4.9%
Evidence, as first published
The UK unemployment rate for people aged 16 years and over was estimated at 4.9% in May to July 2026. (source, published 2026-09-15)
Brier score
0.01 — band midpoint 0.90. 0 is perfect, 0.25 is what always saying 50% scores. What does this Brier score mean?
Reasoning recorded at issue — unchanged since, including where it turned out to be wrong: The headline rate was 4.9% for April to June 2026, as first published on 18 August 2026 and resolved on this register (lxf_external_004); the same bulletin notes the rate down 0.1pp on the quarter, with vacancies at 707,000 and falling — a cooling but not collapsing market. The May-to-July window shares two of three months with the April-to-June window, so month-to-month movement in the rolling headline rate is structurally small; one-quarter changes of more than 0.2pp are rare in the modern series. The band is the July print ±0.2pp. The residual risk is methodological rather than economic: the Labour Force Survey remains in transformation and its month-on-month volatility has been higher than historic norms, which is why this is not issued at 90%+.
If ten forecasts were issued at this band, the band claims roughly 9 of them should come true. Calibration — not certainty — is what the register is scored on.
What would prove this system wrong: a forecast on this register edited after issue, a resolution scored against anything but the source named in advance, or an accuracy claim made before ten outcomes have resolved in a domain. Every one is checkable from the raw ledger — the full falsification list.