Zeno · audit
Paste a forecast.
See how it ends.
Words or numbers. Free.
Type your own question and who forecast it. A fixed rulebook (not the model) spots copied opinions.
Illustrative values · not a real case
Rule-based assessment
Risk
elevated
Declared independent groups
3 → 2
● shares a model · | consensus
consensus
76%
2 declared source group(s) of 3
validated correction
none
no proven adjustment
track record
0/3
calibration known
vendor-a and vendor-b share an underlying model or family: counted once
most useful next check · Agree on the resolution rule: what exactly counts as the outcome, and who decides.
no proven adjustment for this field: original forecasts preserved · scorer: Brier / log score · original forecasts preserved
Assessment scope
- range supplied · judged
- point + range · judged
- point alone · kept, support named
Show the evidence · past misses, inside a check, review budget
Explore a field · which confident forecasts missed
01 / Explore
02 / Investigate
ECB
Who predicts best
108 ranked of 116 forecasters
- 1. ecb · 055+15%
- 2. ecb · 135+12%
- 3. ecb · 022+4%
- 4. ecb · 039+5%
- 5. ecb · 063+4%
skill vs panel median · orange = beats the median, grey = worse
Past record · frozen 21 Oct 2026 · n 58,362 scored forecasts
evidence +−
Three skills · this field
Correct the forecast
promising · your forecast returned for now
Passed, but only 7 distinct outcomes behind 82 test cases.
Warn: one forecaster wrong
ahead of most rivals · +16.8%
practice test on past data · 12,040 rows · not proven
Warn: whole panel wrong
not measured yet
no validated test yet
proven only after its own judging · run receipts ↗
No market comparison, shared-miss pairs or replay cases recorded for this field yet.
03 / Live signals
Open questions
…
Next close
…
Live AI entrants
…
Next model
step 1 of 5: reading all the data · GPU training not started
reading… · SurpriseBench →
past replay · already resolved
What a check looks inside.
systems
6
distinct families
5
own ranges missed
6 of 6
still missed after correction
2 of 6
ECMWF AIFS
family ecmwf
no recordstill misses
ECMWF IFS
family ecmwf
widened ×1.9now covers
GEM · Canada
family cmc
widened ×2.1now covers
ICON · Germany
family dwd
widened ×2.0now covers
GEFS · US
family ncep
widened ×2.3now covers
UKMO · UK
family ukmo
widened ×2.3still misses
Read: 5 independent developers agreed, and all 6 ranges missed on the same side. Agreement here was one shared mistake, not five confirmations.
Thin bar: the system's own 90% range. Wide bar: the range after the correction its past record earned (none where no record exists). Teal line: observed 23.1°C. Source Offdiagonal failure-atlas-v1/weather/weather-ensemble-residuals.parquet.
review budget
How many alerts are worth it?
of 100 alerts
62 real misses · 38 false alarms
share of all confident misses caught
33%
The rest still slip through. Alerts fire above the level that flagged the riskiest 10% of training cases.
- ensemble weather only
- confirmation period · v0.57
- retrospective, not prospective