Toxicity-MM fill-hazard policy slice
negative-resultstrategy
Question
Test whether a causal fill-risk gate improves the frozen pessimistic strategy economically.
Verdict
Negative stop: no fill-hazard gate passed the strict economic-improvement rule.
In plain language (2 terms)
This site's own explanation of the words above. The verdict itself is published exactly as the research index records it and is not rewritten here.
- negative stop
- The measurement came back negative and the work stopped there, by a rule agreed before the measurement ran. It was not retried under different settings until it came out differently.
- fill-hazard
- An attempt to predict how likely a resting order is to fill in the next stretch of time, so that orders unlikely to fill could be skipped.
- Population
- Frozen development-only five-minute candidate universe.
- Resolution
- Candidate-level fill labels and aggregate policy evaluation.
- Cutoff
- Final 90-day holdout unread.
- Baseline
- Unchanged frozen reprice baseline and strict economic-improvement gate.
Limitations
No model-family widening or follow-up sweep is authorised after the negative stop.
Notes
No additional notes published for this entry.
Source
- Citation
- po-research:toxicity-mm-fill-hazard:d711712f0084a667
- Write-up
- docs/research/toxicity-mm-fill-hazard.md
- Data
- docs/research/toxicity-mm-fill-hazard.json
- Artifact
research-index.jsonv1 — sha256:cb3d976779a5cbcb4188039a8d26d5e558a2d844030a638c2f16fe5a8eb0a2fe- Build data cutoff
- 2026-08-12
- Issues
- #175