Evaluation sample
47 forecast points · 26 exact global reset events · 7 positive 24-hour outcomes · 13 positive 48-hour outcomes.
RESEARCH
CNR-0.4.2 research results
CNR-0.4.2 is a research-only forecast being compared with the current public forecast. It does not replace the public CNR-0.2 forecast.
24-hour result
CNR-0.4.2 Brier: 0.133382 vs baseline 0.133834. Log loss: 0.462051 vs 0.465475. Brier skill: +0.34%.
48-hour result
CNR-0.4.2 Brier: 0.210662 vs baseline 0.221653. Log loss: 0.625665 vs 0.666093. Brier skill: +4.96%.
What the result means
The research forecast performed slightly better over 24 hours and more clearly better over 48 hours in this historical sample.
COMPARISON
Measured against a simple baseline
Both forecasts predict the same type of broad reset. Brier score and log loss measure how closely the predicted probabilities matched what later happened. Lower scores are better.
LIMITS
The sample is still small
The available history may be incomplete, and a good result on past events does not prove future accuracy. More live outcomes are needed.
ONGOING CHECK
New forecasts are tracked over time
The research forecast continues to be scored as new 24-hour and 48-hour windows finish.
Current research forecast
— within 24h · — within 48h
Loading current research forecast…
Evidence collected
Loading observation time and completed reset cycles…
24-hour score
No completed score loaded yet.
48-hour score
No completed score loaded yet.
These results are published for transparency. The live homepage continues to use the current public forecast.