No. 04What Actually Predicts a Rugby Injury (And What Doesn't)
Two seasons, one squad, and a model that catches 4 in 10 injuries in its top risk bucket. The findings: recurrence is the dominant signal, body regions have distinguishable pre-injury fingerprints, contact injuries can't be predicted from load, and the standard 4-week chronic window is the inferior choice.
12 min
May 2026
No. 03How We Trained to Stop Fading: A Year of Periodisation in Division 2 Pro Rugby
The training programme behind the closing-twenty number. Annual periodisation, weekly volumes, position-group splits, and one finding I almost didn't publish: deload weeks were associated with a 56% higher injury rate than non-deload weeks.
9 min
Apr 2026
No. 02The Closing Twenty
Across 56 Division 2 matches in 2025, six of eight clubs lost the closing twenty minutes on points. Kintetsu's net was +8.1 per match, the largest in the league. A descriptive look at the pattern, points only.
5 min
Apr 2026
No. 01Rugby Moneyball: What Actually Predicts Wins (Two Seasons, Two Leagues)
Across 302 team-matches in Japan Rugby League One (D1 23/24 plus D2 24/25), fifteen of sixteen candidate features hold their direction in both leagues. Points per ball carry remains the single largest non-outcome signal. The replication tells you what to track, what to drop, and which one-season findings should never have been published.
10 min
Apr 2026
