Forecast track record

Archived, never edited, scored against official data. The first round was archived retroactively.

Scoreboard

Every scored forecast against its ONS outturn.

Last scored ; data store updated 16 Nov 2026.

Live forecasts in range 4 of 5 +4 retroactive · 9 forecasts, 5 outturns
Average missn=9 0.34pp ModelNo change No change: 0.47pp · lower is better
Biasn=9 −0.33pp Ran low on 8 of 9
Rounds scored 4 of 4 Next: 2026Q3 unemployment Each has at least one score

Too few to show skillLimits

The record

Target and roundMedian, archived bands and outturnErrorWhere it fell
Median 2.68%, 68% band 2.24–3.16, 90% band 1.94–3.45, outturn 2.80%.
−0.12pp
Median 1.10%, 68% band 0.55–1.67, 90% band 0.19–1.97, outturn 1.37%.
−0.27pp
2026Q2 UK real GDP, y/y2026-07-29 · BoE SVARh1backcast
Median 1.14%, 68% band 0.59–1.68, 90% band 0.19–1.97, outturn 1.37%.
−0.23pp
Median 4.97%, 68% band 4.94–5.00, 90% band 4.93–5.01, outturn 4.90%.
+0.07pp
Median 2.71%, 68% band 1.95–3.45, 90% band 1.42–3.95, outturn 3.10%.
−0.39pp
2026Q3 UK CPI, y/y2026-07-29 · BoE SVARh2
Median 2.54%, 68% band 1.83–3.29, 90% band 1.40–3.85, outturn 3.10%.
−0.56pp
Median 1.36%, 68% band 0.53–2.23, 90% band -0.09–2.70, outturn 1.94%.
−0.58pp
2026Q3 UK real GDP, y/y2026-07-29 · BoE SVARh2
Median 1.46%, 68% band 0.61–2.27, 90% band 0.11–2.80, outturn 1.94%.
−0.48pp
2026Q3 UK real GDP, y/y2026-11-12 · BoE SVARh1backcastarchived 05:13 UTC, before the 07:00 UTC ONS release
Median 1.57%, 68% band 1.01–2.14, 90% band 0.59–2.52, outturn 1.94%.
−0.37pp

Error = archived median minus outturn, in pp; positive means the forecast ran high.

Against the benchmarks
Absolute error on the same outturns, in percentage points; each benchmark's own value beneath. A benchmark appears only beside rounds fixed after it was published, so a later edition never sits beside an earlier forecast; the October 2026 EFO and November 2026 MPR, published after 2026Q3 ended, are close to nowcasts for it. They show where views differed, not a test of skill.
TargetThis roundNo changeBoE Feb 2026 MPROBR March 2026 EFOBoE survey of external forecasters (Apr 2026)BoE Nov 2026 MPROBR October 2026 EFO
2026Q2 UK CPI, y/y0.120.303.10%0.702.10%0.662.14%0.102.90%——
2026Q2 UK real GDP, y/yround 2026-07-210.270.530.84%0.570.80%0.450.92%———
2026Q2 UK real GDP, y/yround 2026-07-290.230.530.84%0.570.80%0.450.92%———
2026Q2 UK unemployment rate0.070.105.00%no published quarterly path
2026Q3 UK CPI, y/yround 2026-07-210.390.003.10%1.102.00%1.022.08%———
2026Q3 UK CPI, y/yround 2026-07-290.560.003.10%1.102.00%1.022.08%———
2026Q3 UK real GDP, y/yround 2026-07-210.581.100.84%0.941.00%0.721.22%———
2026Q3 UK real GDP, y/yround 2026-07-290.481.100.84%0.941.00%0.721.22%———
2026Q3 UK real GDP, y/yround 2026-11-120.370.571.37%0.941.00%0.721.22%—0.041.90%0.571.37%

No change is the naive baseline: the last outturn at the round's data edge, held flat (read from the current vintage, so revisions can shift it slightly). This is the driftless walk; the limits say what happens against a walk with drift.

Open rounds

On the record. Dots are published ONS outturns, scored above where the round came first; later quarters wait.

GDP growth from the 2026-11-12 round: our median and bands beside the OBR's path.

UK real GDP, % change on a year earlier

BoE SVAR UK real GDP growth forecast vs OBR October 2026 EFO Fan chart of UK real GDP, year-on-year growth, percent. The BoE SVAR archived median with its 68 and 90 percent bands, empirically re-calibrated for measured under-coverage, is shown beside the OBR October 2026 EFO path over the overlapping quarters. ONS outturn for 2026Q3: 1.94 percent.
  • BoE SVAR median
  • 68% band (re-calibrated)
  • 90% band (re-calibrated)
  • OBR October 2026 EFO
  • ONS outturn

Source: BoE SVAR round, 12 Nov 2026 · OBR October 2026 EFO · ONS

Details

Numbers by quarter

BoE SVAR UK real GDP growth forecast, round 2026-11-12, beside the OBR October 2026 EFO: % change on a year earlier. Bands are re-calibrated, as drawn.
QuarterMedian68% band (re-calibrated)90% band (re-calibrated)EFO
2026Q31.57%1.09%–2.06%0.54%–2.58%1.37%
2026Q41.88%1.13%–2.64%0.38%–3.45%1.51%
2027Q11.57%0.26%–2.67%−0.63%–3.66%1.11%
2027Q21.28%−0.01%–2.64%−1.44%–4.04%0.95%
2027Q31.16%−0.20%–2.61%−1.50%–4.27%0.96%
2027Q41.16%−0.26%–2.66%−1.42%–4.06%1.16%
2028Q11.24%−0.21%–2.79%−1.64%–4.19%1.37%
2028Q21.29%−0.11%–2.78%−1.82%–4.47%1.58%
2028Q31.30%−0.16%–2.73%−1.85%–4.37%1.75%
2028Q41.40%−0.17%–2.69%−1.92%–4.49%1.81%
2029Q11.35%−0.28%–2.74%−1.94%–4.47%1.83%
2029Q21.34%−0.15%–2.86%−1.90%–4.65%1.82%
2029Q31.36%−0.19%–2.96%−1.88%–4.94%1.80%

Sources

Source: archived round 2026-11-12/boe-svar.json; OBR October 2026 EFO path: four-quarter growth of Table 1.1 real GDP, detailed economy tables, stored in forecasts/official/obr_efo_oct_2026.csv; outturns from the data store

Same round, CPI: the model tracks the EFO through 2027Q3, then stays near 3% while the EFO returns towards 2%. Nothing in the model pulls inflation back to target (a unit root, no inflation anchor), so its later quarters extend recent inflation (limits).

UK CPI, % change on a year earlier

BoE SVAR UK CPI inflation forecast vs OBR October 2026 EFO Fan chart of UK CPI, year-on-year inflation, percent. The BoE SVAR archived median with its 68 and 90 percent bands, empirically re-calibrated for measured under-coverage, is shown beside the OBR October 2026 EFO path over the overlapping quarters. ONS outturn for 2026Q3: 3.10 percent.
  • BoE SVAR median
  • 68% band (re-calibrated)
  • 90% band (re-calibrated)
  • OBR October 2026 EFO
  • ONS outturn

Source: BoE SVAR round, 12 Nov 2026 · OBR October 2026 EFO · ONS

Details

Numbers by quarter

BoE SVAR UK CPI inflation forecast, round 2026-11-12, beside the OBR October 2026 EFO: % change on a year earlier. Bands are re-calibrated, as drawn.
QuarterMedian68% band (re-calibrated)90% band (re-calibrated)EFO
2026Q32.97%2.43%–3.56%1.94%–4.02%3.05%
2026Q43.48%2.40%–4.50%1.54%–5.63%3.52%
2027Q13.62%2.12%–5.16%0.57%–6.74%3.79%
2027Q23.58%1.38%–5.81%−0.50%–7.96%3.64%
2027Q33.30%0.69%–5.94%−1.52%–8.16%3.38%
2027Q43.15%0.72%–5.81%−1.86%–8.40%2.89%
2028Q13.07%0.54%–5.82%−1.64%–8.07%2.37%
2028Q23.08%0.78%–5.47%−2.61%–8.44%2.14%
2028Q33.07%0.80%–5.53%−1.98%–8.84%1.99%
2028Q43.15%0.80%–5.53%−2.30%–8.95%1.91%
2029Q13.06%0.72%–5.50%−2.30%–8.62%2.02%
2029Q23.02%0.72%–5.48%−2.02%–8.65%2.13%
2029Q33.06%0.63%–5.55%−2.34%–8.69%2.14%

Sources

Source: archived round 2026-11-12/boe-svar.json; EFO path from Table 1.7 of the October 2026 EFO detailed economy tables, stored in forecasts/official/obr_efo_oct_2026.csv; outturns from the data store

Unemployment, satellite round 2026-07-28; no official path to overlay. Bands carry GDP uncertainty only and run narrow: in a backtest excluding Covid quarters, 68% bands built this way held 22–48% of outturns and 90% bands 30–61% (coverage: the share inside a band).

UK unemployment rate, %

BoE SVAR unemployment satellite: UK unemployment rate forecast, median with 68% and 90% bands Fan chart of the UK unemployment rate, percent, from the BoE SVAR unemployment satellite round of 2026-07-28: the archived median with its 68 and 90 percent bands, 2026Q2 to 2027Q1. The median runs from 4.97% to 4.93%. ONS outturn for 2026Q2: 4.90%.
  • BoE SVAR unemployment satellite median
  • 68% band (GDP uncertainty only)
  • 90% band (GDP uncertainty only)
  • ONS outturn

Source: BoE SVAR unemployment satellite, 28 Jul 2026 · ONS

Details

Numbers by quarter

BoE SVAR unemployment satellite, round 2026-07-28: UK unemployment rate, %. Bands as archived (GDP uncertainty only).
QuarterMedian68% band90% band
2026Q24.97%4.94%–5.00%4.93%–5.01%
2026Q34.96%4.89%–5.03%4.85%–5.08%
2026Q44.94%4.81%–5.07%4.74%–5.16%
2027Q14.93%4.74%–5.14%4.61%–5.27%

Sources

Source: archived round 2026-07-28/okun-unemployment.json; outturn, ONS series MGSX from the data store

Band calibration and the EFO comparison
  • The October 2026 EFO was fixed on 28 Oct, before the 12 Nov round, on less data: the charts show where views differ. The forecast evaluation is on the BoE SVAR validation page.
  • A round archived after an outturn is drawn beside it but never scored on it.
  • GDP and CPI bands are rescaled per horizon to measured coverage: on average they under-cover, but UK GDP over-covers at short horizons (68% factor 0.85 one quarter ahead, 1.13 at eight). Factors: band_calibration.json.
  • Factors come from the model shown (with its Covid dummies) over 50 overlapping origins, so they are imprecise beyond two decimals.
  • The EFO growth path is four-quarter growth of Table 1.1 real GDP in the October 2026 EFO detailed economy tables.
  • Raw quantiles stay untouched in the round files. Re-calibrated numbers are tabled under each fan; raw medians and bands are on the Growth and Inflation tabs. The BoE MPR and the survey of external forecasters are in the official CSVs.

Archive

What was claimed, and when. Each round is frozen when archived.

  1. BoE SVAR

    Data edge
    2026Q1
    Periods
    13
    Scored
    4
    Archived 25 Jul 2026Round fileHistory
  2. 7 days later

    BoE SVAR unemployment satellite

    Data edge
    2026Q1
    Periods
    4
    Scored
    1
    Archived 28 Jul 2026Round fileHistory
  3. 1 day later

    BoE SVAR

    Data edge
    2026Q1
    Periods
    13
    Scored
    3
    Archived 29 Jul 2026Round fileHistory
  4. 106 days later

    BoE SVAR

    Data edge
    2026Q2
    Periods
    13
    Scored
    1
    Archived 12 Nov 2026Round fileHistory

First round weaker: archived after the CPI release from a file committed ~16 h before it, with thinner provenance.

Data vintages and scoring inputs

The data edge is the last observation the model was allowed to see; outturns are read from the dated, never-edited snapshots in the data store.

A CI job fails any pull request that edits or deletes a round, so a correction creates a new round and the superseded one stays scored.

Scores are computed from outturns.json, which is vintage-versioned so revisions never silently rewrite a score, and written to scorecard.json; the method is in the README.

Method and limits

What this record does not show, and how it is kept.

Limits

  • Too few to judge: no skill claim before ten or more outturns, a year or more away.
  • Bands run narrow in hindsight tests, worse with horizon; open fans are re-calibrated.
  • Unemployment bands are too narrow: the archived round carries GDP uncertainty only (measured coverage). The model now adds the satellite’s own uncertainty; no round with those bands is archived yet.
  • No demonstrated skill against a drifting random walk: no gain on any variable survives multiple-testing correction.
  • Three variables, no fixed calendar; coefficients are not re-estimated every round.
  • Every row here is BoE SVAR; the microsimulation does not forecast. How a score is built

How we score six rules, each enforced in code

  1. 1Archive first. The round's JSON is committed before the outturn is published (the first round is the exception).
  2. 2Never edit. CI rejects any edit or deletion; a correction is a new round.
  3. 3No hindsight. A round archived after the outturn's snapshot is never scored on it.
  4. 4Dated outturns. Read from a dated ONS snapshot; revisions are appended, never overwritten.
  5. 5Error and coverage. Signed error of the median; coverage of the bands as archived.
  6. 6Benchmarks. Beside a no-change forecast and any published path for that quarter.
The full method
  • Archive and freeze
    • One JSON file per model under forecasts/rounds/<date>/: median, 68% and 90% bands, data edge, source hash.
    • The forecast-archive workflow fails any PR that edits, renames or deletes one.
    • So the Git timestamp proves when each claim was made.
  • No hindsight
    • score.py scores a period only if the round was archived on or before the outturn’s snapshot date.
    • Later rounds covering the same quarter are never scored on it.
    • The test is by date; it matters only for the first round, archived after the CPI release from a file committed before it (archive).
  • Outturns
    • ingest_outturns.py converts stored ONS series to the forecasts’ units: GDP growth from the ABMI level; CPI and unemployment as published.
    • Each value is appended with its snapshot date; revisions are appended, never overwritten.
    • The scoreboard uses the newest; earlier scores stay reproducible from outturns.json.
  • Errors and bands
    • Error is the archived median minus the outturn; positive means it ran high.
    • Kept signed, because bias and random misses differ.
    • Coverage uses the bands as archived, not the re-calibrated bands in the open-round charts.
  • Benchmarks
    • A no-change forecast: the last outturn at the data edge, held flat.
    • Any stored quarterly path published before the round: BoE MPR (Feb and Nov 2026), OBR EFO (Mar and Oct 2026), BoE survey of external forecasters (Apr 2026; the Oct 2026 survey is stored but scores no quarter yet), from the official CSVs.
    • Everything lands in scorecard.json, generated, never edited by hand.
  • Why publish this early
    • It makes early failures part of the evidence.
    • Bank Rate comes closest to an accuracy gain, which does not survive the multiple-testing adjustment; it is not forecast here yet.
    • Rounds follow data-edge refreshes, not a calendar.

Release notes

One written note and the automated digests of each stored data release. Neither is evidence of skill.