← Research hub
NHL ·
Evidence-first read: conclusion first, then methodology, measured deltas, and sample limits.
Conclusion
Evgeny Romasko leans unders vs closing goal totals
Totals go unders 25.0% (9-22-5) across 31 lined games.
Rule and category
ou edge
O/U ATS uses estimated closing totals where sportsbook data is unavailable. Min 30 decisive games required.
Evidence
Measured deltas in sample
- O/U ATS: 9-22-5 (25.0%)
- Sample: 31 (Min 30 games)
- Edge vs 50%: 21.0 pts (Leans unders)
Related context
Profiles and splits to verify
Follow linked officials and teams to see whether adjacent samples support or contradict this pattern. Unresolved splits stay descriptive until sample gates clear.
Uncertainty
Sample and limits
Sample: 31 games over 10 seasons (2016–2026)
Historical context
Where this sits in the dataset
Ranked with other NHL findings by effect size and sample depth. Use the tendency index and team-with-official history table for broader crew history after reading the measured deltas above.
- O/U ATS
- 9-22-5
- 25.0%
- Sample
- 31
- Min 30 games
- Edge vs 50%
- 21.0 pts
- Leans unders
Methodology
- Findings ranked by effect size × √sample size with category deduplication.
- Sample gates: 30+ ref games, 8+ team splits, 30+ ATS decisions where applicable.
- Estimated closing lines disclosed where applicable, not betting advice.
- Full methodology →