Determinants of Cardiovascular Load Variability During Small-Sided Games: Influence of Game Format, Bout Intensity, and Yo-Yo Intermittent Recovery Performance

Using a randomized, counter-balanced cross-over design in under-19 male soccer players, this study examined whether within-player cardiovascular load variability during small-sided games (SSGs) is more strongly associated with game format (3v3 vs 5v5), intermittent aerobic capacity, or the mean heart-rate intensity achieved across bouts. Sixty players (17-19 years) completed two weeks of SSGs, performing four 3-min bouts per format with 2-min recovery. Heart rate (HR) was recorded continuously. Cardiovascular variability for each format was quantified for each player as the coefficient of variation of bout-level HRmean across the four 3-min bouts (CV% = SD/mean × 100). Intermittent aerobic capacity was assessed using the Yo-Yo Intermittent Recovery Test Level 1 (Yo-Yo IR1). Linear mixed-effects modelling examined the independent associations between format, Yo-Yo IR1 performance, mean HR intensity, and CV% was log-transformed to address skewness and improve model residuals. Paired comparisons showed that HRmean was higher in 3v3 than in 5v5 (p < 0.001), whereas between-player variability (CV%) was higher in 5v5 than in 3v3. Specifically, CV% was 1.97 ± 0.81% in 3v3 versus 2.61 ± 1.21% in 5v5, while mean HR intensity was 180.9 ± 2.8 bpm in 3v3 compared with 173.6 ± 3.6 bpm in 5v5. The mixed model indicated that mean HR intensity was the only significant predictor of cardiovascular variability (β = -0.027, 95% CI = -0.050 to -0.005, p = 0.019). Neither SSG format (β = 0.056, p = 0.618) nor Yo-Yo IR1 performance (β = -0.00008, p = 0.385) were significant predictors. The fixed-effects structure explained 13.6% of the variance (marginal R² = 0.136). Within the training conditions studied, higher mean HR intensity was associated with lower bout-to-bout HR variability. However, the fixed effects explained a modest proportion of variance, indicating that additional unmeasured factors likely contribute to cardiovascular load variability.