Single-team models using machine learning identify influential predictors of athlete availability in elite female athletes, with differing predictors between lacrosse and soccer teams.
This study used elastic net regression to identify predictors of athlete availability (percentage of unmodified practices and games) in 52 NCAA Division I female lacrosse and soccer athletes. Combined model RMSE was 17.8%, while single-team models showed RMSEs of 8.9% (lacrosse) and 17.8% (soccer), with no force plate variables selected for lacrosse and no body composition variables for soccer.
This article is automatically generated from the structured evidence profile behind the claim above. Scores reflect the quality and quantity of available research, not clinical advice.
Single-team models using machine learning identify influential predictors of athlete availability in elite female athletes, with differing predictors between lacrosse and soccer teams. The current body of evidence comprises 1 study. EvidenceHub rates the overall confidence at 32/100 (low).
The Claim
Single-team models using machine learning identify influential predictors of athlete availability in elite female athletes, with differing predictors between lacrosse and soccer teams.
This conclusion is most relevant to: 52 National Collegiate Athletic Association Division I elite female athletes (28 lacrosse, 24 soccer)..
What the Research Shows
The conclusion draws on 1 linked study. Highlights from the cited literature:
- ▸Using Machine Learning for Identification of Athlete Availability Predictors in a Multisport Elite Female Athlete Cohort. (Journal of strength and conditioning research, 2026) —
How It Works
The proposed biological pathway:
- ▸Collect training load, recovery, wellness, body composition, and force plate data
- ▸Apply elastic net regression to identify influential predictors
- ▸Compare combined vs single-team models
- ▸Result: Single-team models yield lower RMSE and more consistent predictors
Who Might Benefit
Evidence fit by population:
- ▸52 National Collegiate Athletic Association Division I elite female athletes (28 lacrosse, 24 soccer).
Recommended Dose
N/A
Limitations & Caveats
Important context when interpreting this evidence:
- ▸Small sample size (n=52) limits generalizability
- ▸Single-season data may not capture year-to-year variability
- ▸No causal inference possible from observational design
Frequently Asked Questions
What machine learning method was used?▼
Elastic net regression was used to identify influential predictors of athlete availability.
Which predictors were most important for lacrosse athletes?▼
Training load, recovery, and wellness variables were selected; no force plate variables were selected for lacrosse athlete availability.
How was athlete availability defined?▼
Athlete availability was defined as the percentage of unmodified practices and games.
Why might single-team models be better than combined models?▼
Single-team models showed lower RMSE (8.9% for lacrosse vs 17.8% combined) and identified different predictors, suggesting team-specific factors are important.
References
- 1.Moore SR, Cantú EI, Brantner CL, Britton ME, DelBiondo GM, Blue MNM, Bruinvels G, Hackney AC, Register-Mihalik JK, Smith-Ryan AE. “Using Machine Learning for Identification of Athlete Availability Predictors in a Multisport Elite Female Athlete Cohort..” Journal of strength and conditioning research, 2026. PMID: 42413555 DOI: 10.1519/JSC.0000000000005611