Section 3 of 6
Pooling of Vertical-Jump Metrics
Ibnu Noufal Kambitta Valappil, Karuppasamy Govindasamy, Gavoutamane Vasanthi, Masilamani Elayaraja, Cain C. T. Clark, Koulla Parpa, Borko Katanic, Hüseyin Şahin Uysal, Hassane Zouhal, and Urs Granacher · about 1 minutes
The commentary’s most substantive concern is with the inclusion of the countermovement jump (CMJ), Sargent jump, drop jump, squat jump and Bosco index within a single meta-analysis for vertical jump height. We have considered this argument carefully and, on reflection, maintain that the original pooling is defensible.
Each test included is, by design, a measure of vertical jump height as a proxy for lower-limb muscular power. Cochrane guidance explicitly supports combining such measures using the SMD when they capture the same conceptual domain. This approach is consistent with established practice in meta-analyses [6, 19, 20].
The commentary’s own jump-type-specific analyses yield point estimates that consistently favour FIFA 11 + (SMD = 0.28–0.32). These estimates fall within a consistent small-to-moderate range, albeit with some attenuation relative to the pooled estimate (SMD = 0.56). This attenuation is expected when analyses are disaggregated, as statistical power is reduced and confidence intervals widen.
Notably, the magnitude of these estimates clearly falls within a plausible range for training-induced adaptations. Methodological reviews have highlighted that implausibly large effect sizes (e.g., > 3 standard deviations) are frequently attributable to calculation errors rather than true effects [15]. The estimates observed here do not exhibit such characteristics, further supporting the validity of the pooled findings.
We acknowledge that each jump metric places different relative demands on the stretch–shortening cycle, on arm-swing coordination and on concentric versus reactive muscle action [21, 22]. This is precisely why heterogeneity was reported and why subgroup analyses were used to probe moderators. But the existence of mechanical differences between test modalities does not preclude their aggregation under a shared functional construct, any more than aggregating different pain scales or different measures of aerobic capacity would be precluded by their differing item content.