{"resourceId":"nature-cmj-pipeline-variation-ncaa-di-2026","versions":[{"version":"external-3c51094a43cc226436c7ac2be50fb4df5710a35c0a83c3640c8cf28f8879c57d","resource":{"id":"nature-cmj-pipeline-variation-ncaa-di-2026","title":"Scientific Reports: CMJ processing pipeline choice exceeds 2 cm SWC in ~1/3 of 32,952 NCAA DI trials","organization":"Auburn University Athletics collaborators / Scientific Reports (Nature)","sector":"NCAA Division I strength & conditioning / sports science measurement","geography":"United States (NCAA Division I; Auburn Athletics testing dataset)","publishedAt":"August 19, 2026","publicationDate":"2026-08-19","eventDate":null,"sourceName":"Scientific Reports","sourceLabel":"Peer-reviewed large-scale measurement study (HTML)","sourceUrl":"https://www.nature.com/articles/s41598-026-65211-2","evidenceClass":"academic-research","outcomeClass":"cautionary","topics":["infrastructure","data-security","operating-model","developers-agents"],"finding":"Goodlett, Roper, Aakur et al. applied 30 factorial CMJ pipelines (5 body-weight × 6 onset methods) to 32,952 countermovement-jump trials from 579 NCAA Division I athletes across 15 teams, 12 sports, and five seasons (11,410 sessions). Variance-minimization body-weight methods were interchangeable (95% LoA within ±5 N). First-derivative onset was most consistent across signal-quality tiers; threshold onsets showed quality-dependent timing bias. Pipeline choice produced jump-height divergence exceeding a 2 cm smallest worthwhile change in about one-third of trials, while matched-pair between-method bias on plausible trials stayed ≤0.4 cm. Authors recommend variance-minimization BW with derivative onset for arms-akimbo field CMJ on consistency—not criterion-validated accuracy—grounds.","sledRelevance":"Why now: Same-week stream vigilance on wearable/injury ML metrics makes this unarchived Aug 2026 DI-scale finding newly actionable—measurement pipeline drift can dominate claimed AI readiness signals before models are even trained.","evidence":"Opened Nature HTML reports abstract metrics (32,952 trials; 579 athletes; 30 pipelines; 2 cm SWC exceedance ~1/3; ≤0.4 cm matched-pair bias), publication date 2026-08-19, Auburn Athletics acknowledgments. PDF fetch returned non-PDF challenge page; HTML used as primary inspection.","architectureImplications":"Pin and version force-plate feature pipelines; store raw traces; treat vendor jump height as method-dependent; feed models only with pipelineId metadata.","governanceImplications":"Require change control when onset/BW methods change mid-season; disclose processing to sports medicine before using CMJ AI for return-to-play.","securityPrivacyImplications":"Large multi-season force-plate datasets are identifiable performance records—govern access and retention like other athlete biometrics.","caveats":"Consistency recommendation without criterion force-plate gold standard; arms-akimbo CMJ only; single athletics department data source; HTML inspection (PDF binary blocked).","streamIds":["college-athletics"],"roles":{"sales":"Sales framing: Strength staff buying jump-testing AI or readiness scores should hear that pipeline choice alone moved NCAA DI CMJ jump height beyond a 2 cm smallest worthwhile change in about one-third of 32,952 trials. Sell a processing-standardization engagement before model procurement. Bound claims to measurement consistency, not injury prediction. Unsupported: that any single commercial pipeline is criterion-validated here, or that AI on inconsistent CMJ features is clinically safe.","engineering":"Engineering framing: Fit is force-plate CMJ feature engineering governance for DI athletics. Architecture: store raw force traces; pin body-weight estimation (variance-minimization) and derivative-based onset; version pipelines; reject silent vendor reprocessing. Prerequisites: arms-akimbo protocol fidelity, signal-quality tiers, change-log. Constraints: recommendation is consistency-based not criterion-validated; Auburn Athletics dataset; arms-akimbo only. Security: athlete performance biometrics under RBAC. Proof of value: within-trial pipeline delta distribution and SWC exceedance rate under locked config.","delivery":"Delivery framing: Freeze CMJ processing SOP, reprocess a historical sample under 2–3 pipelines, brief S&C on SWC impacts, and gate any ML until pipeline ID is immutable in the feature store. Dependencies: S&C testing staff, sports science, data platform. Ownership: S&C director owns protocol; analytics owns pipeline versioning. Governance checkpoints: change-control board for onset/BW methods. Risks: cross-season incomparability and false readiness alerts from silent pipeline drift."},"roleDetails":{"sales":{"customerProblem":"Departments feed CMJ metrics into readiness/injury AI without knowing processing choices can exceed smallest worthwhile change within the same trial.","stakeholders":["Head S&C","Sports science","Team physicians","Analytics"],"discoveryQuestions":["Which onset and body-weight methods are locked today?","Do vendors reprocess historical jumps silently?","What SWC do you use for jump height?"],"valueHypothesis":"Standardizing variance-minimization body-weight with derivative onset may reduce within-trial metric noise that otherwise masquerades as performance change.","potentialEngagement":"Pipeline audit on one season of force-plate CMJ with SWC exceedance report.","unsupportedClaims":["Criterion gold-standard accuracy","Injury-prediction improvement","Transfer beyond arms-akimbo CMJ"]},"engineering":{"fit":"Measurement-layer control for DI force-plate CMJ before AI features.","architectureIntegration":"Raw force store to pinned BW/onset modules to versioned metrics to model features tagged with pipelineId.","prerequisites":["Force-plate export","SOP","Quality tiers","Change log"],"constraints":["Consistency not criterion accuracy","Single university dataset source","Arms-akimbo only"],"security":"Protect raw force and derived metrics as athlete performance data with RBAC.","proofOfValue":["SWC exceedance rate under alternatives","LoA for BW methods","Immutable pipelineId in exports"]},"delivery":{"work":"Document current pipeline, reprocess sample, publish SOP, block ML jobs without pipeline pin.","dependencies":["S&C testing","Platform engineering","Vendor cooperation"],"ownership":"S&C owns field protocol; sports science owns analytic pipeline.","skillsAdoption":"Train staff that jump-height deltas below SWC may be processing artifacts.","governanceCheckpoints":["SOP sign-off","Quarterly pipeline drift check"],"acceptanceCriteria":["Written BW+onset standard","Historical reprocess memo","Feature-store pipeline tag"],"risks":["Vendor lock-in","Broken longitudinal norms","Overtrust in AI on noisy labels"]}},"retrievedAt":"2026-10-03T22:16:22-05:00","enrichedAt":"2026-10-03T22:18:14-05:00","enrichmentBasis":"retrieved source","accessibilityWorkforceImplications":"S&C staff need measurement literacy to challenge vendor black-box jump scores; reduces false athlete-decline narratives that harm trust.","procurementImplications":"Contract for raw-export rights, locked processing versions, and notification of pipeline changes—not only dashboard licenses.","operatingModelImplications":"Insert a measurement-governance step before any CMJ-derived AI readiness or load decision.","sourceVerification":{"responseId":"curl-nature-cmj-20261003T221622","openedUrl":"https://www.nature.com/articles/s41598-026-65211-2","referenceExcerpt":"exceeding the 2 cm smallest worthwhile change in approximately one-third of trials","promptVersion":"sled-research-v4.0","model":null,"basis":"agent-reported inspection"}}}]}