Zeno · audit
Paste a forecast.
See how it ends.
Words or numbers. Free.
A real past case, outcome included. Nothing new is assessed.
Zeno's verdict · Market
?
Run the check: Zeno says whether it supports a change, keeps your forecast, needs a range, or has no tested record, computed now by the published model.
Prediction marketsReplay · outcome already known
Will Sam Konstas make a test century?
Case evidence
Zeno did not beat the market price on later data, so it leaves the 69% market estimate alone. The event resolved No.
Served answer
reference kept
Published answer · run it to see today's
Check my forecast · guided
Advanced · edit the raw request
Forecast spectrum
Agreement meets reality.
Zeno replay · Decision · 2024-12-29 00:43 UTC · n 1 forecasts · 1 replay case
evidence +−
panel
69%
Zeno
reference
actual
No
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
Prediction markets
Who predicts best
Not measured. one forecaster per question: no shared questions to rank on
Three skills · this field
Correct the forecast
keeps your forecast unchanged
live model's sealed exam: no measurable difference
Warn: one forecaster wrong
no edge yet
practice test on past data · 13,170 rows · not proven
Warn: whole panel wrong
not measured yet
no validated test yet
proven only after its own judging · run receipts ↗
Weather vs markets
The crowd beat the weather models in every period.
Learning · Jun–Jul · n 651
Check · Aug · n 348
Later test · Sep · n 330
MarketWeather models50/50 blenderror (RPS) · lower is better
both confidently wrong: 12 of 330 later-test days · past cases, not Zeno warnings · 401 days without a pre-decision price left out · price times observed, model times derived
Past record · Jun–Sep 2026 · Offdiagonal v6.11 · n 1,329 city-days · 15 cities
evidence +−
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