REF-825 · Knechtle et al., 2011 · The Physician and Sportsmedicine, Vol. · Paper · PubMed

Anthropometric and Training Variables Related to Half-Marathon Running Performance in Recreational Female Runners

Blimited evidenceMeasured on recreational peopleWritten before October 2026, being re-checked

In 42 recreational female half-marathon runners, race time correlated in bivariate analysis with body weight (r, 0.60), body mass index (0.48), body fat percentage (0.56), several skinfold thicknesses (0.49 to 0.69), mean speed of training sessions (r, -0.68) and half-marathon personal best (0.69).

After multivariate analysis, body weight (P = 0.0054), pectoral skinfold (P = 0.0068) and mean training speed (P = 0.0041) remained significant. Mean training speed was itself related to several skinfold thicknesses (r, -0.31 to -0.44) and to percent body fat (-0.31).

The authors conclude that high running speed in training, as opposed to extensive training, may both reduce upper-body skinfolds and improve race performance in recreational female half-marathoners.

Participants

n=42 recreational, non-professional female half-marathon runners (21.0975 km). The abstract does not give age or training volume.

Protocol

Anthropometric measurements (body weight, body mass index, body fat percentage, eight skinfold thicknesses), pre-race experience and training variables recorded.
Bivariate, then multivariate analysis against half-marathon race time.

Key Results

VariableCorrelation with race time (r)
Body weight0.60
Body mass index0.48
Body fat percentage0.56
Skinfolds: pectoral 0.61, mid-axilla 0.69, triceps 0.49, subscapular 0.61, abdominal 0.59, suprailiac 0.55, medial calf 0.530.49 to 0.69
Mean speed of training sessions-0.68
Personal best half-marathon time0.69
Still significant after multivariate analysis: body weight (P = 0.0054), pectoral skinfold (P = 0.0068), mean training speed (P = 0.0041).
Mean training speed correlated with mid-axilla (-0.31), subscapular (-0.38), abdominal (-0.44) and suprailiac (-0.41) skinfolds, the sum of 8 skinfolds (-0.36) and percent body fat (-0.31).

Practical Application

In 42 recreational female half-marathoners, the women who ran faster in their training sessions ran faster races (r -0.68 between training speed and race time), and training speed stayed a significant predictor after adjusting for body weight and skinfolds; training speed, not how much they trained, was the training variable that held up. For a recreational half-marathoner, the authors suggest that high training speed, as opposed to extensive training, may improve race performance — an association, not a tested effect.
Body weight and skinfold thickness also predicted race time. Inference, not tested here: this is a correlation in a cross-section; it does not justify a weight-loss target, and chasing leanness carries its own risk (low energy availability).
The authors' claim that fast training "may" reduce upper-body skinfolds is an interpretation of correlations, not a tested effect.
Not from this paper; the base's own practice: add quality gradually on top of an easy base, as in SYNTH-64, and keep weekly volume increases moderate (REF-623).

Limitations

Cross-sectional correlations in 42 women; causation cannot be inferred in either direction (faster runners may simply train faster).
How training variables were collected is not described in the abstract; intensity distribution was not reported.
One race; age, training history and physiological measures (aerobic capacity, threshold) are not in the abstract.
Written from the abstract and record metadata only (full text not read).
How this entry was checked and changed
Why this level

cross-sectional / correlational, n=42, one race

Content correction 2026-09-27
2026-10-01: independent check against the abstract — wording brought back to what the abstract states.