Athletes received immediate feedback on their own jump height on a small tablet placed directly in front of them. No other team members could see the display. Standard setup, as commonly used in most sport science monitoring contexts.
Athletes received feedback on their tablet plus a large team leaderboard on a TV screen visible to all athletes and coaches. The leaderboard showed performance rankings and individual jump scores for the entire team in real time, introducing social comparison and competition dynamics.
Raincloud plots comparing IVF vs. SCF conditions for 6 significant CMJ metrics. Pink circles = female athletes; blue circles = male athletes. Black density distributions = SCF; grey = IVF. In all cases SCF showed significantly higher performance. ***p < 0.001, **p < 0.01.
Metric-specific density plots of intra-day coefficient of variation (CV%) by feedback condition. Blue = IVF; grey = SCF. The dashed vertical line = CV of 10% (commonly accepted cutoff for good relative variability). SCF distributions are consistently tighter (left-shifted), indicating less within-session variability.
Left: Athlete-specific random intercepts for jump height compared to the sample mean (±1 SD, grey band). Shows the natural between-athlete spread in jump ability. Right: Athlete-specific slope coefficients showing how much each athlete’s jump height changed between IVF and SCF conditions compared to the average within-session SEM (grey band). Most athletes showed a positive SCF response, but the magnitude varied considerably.
The left plot shows each athlete’s baseline jump height relative to the team average. Athletes whose bars fall within the grey band are near the team mean. Bars extending beyond show above- or below-average jumpers. Random component ICC (ICCR) = 0.95, meaning 95% of variance in jump height was due to differences between athletes — far more than the feedback condition itself.
This is a critical reminder: athlete identity is the dominant driver of jump performance, not feedback type.
The right plot shows the athlete-specific SCF benefit — how much each individual improved during SCF vs. IVF, relative to the average SEM band. Athletes whose bars extend beyond the grey band showed a response exceeding typical noise.
Coaching application: Some athletes are much more responsive to the competitive leaderboard than others. Coaches can use this to identify who benefits most from SCF and who performs consistently regardless of context.
① Be consistent with your feedback type. Mixing IVF and SCF sessions across time will inflate data variability and reduce the sensitivity of your longitudinal monitoring. Pick one approach and stick with it — do not implement feedback conditions randomly.
② Use SCF to maximise intent on test days. If the goal is to elicit peak performance (e.g., pre-season profiling, return-to-play clearance), the leaderboard condition will systematically produce higher jump outputs with more consistent data. However, be mindful of the psychological context — rank-based SCF can also increase shame if not framed around effort and improvement.
③ SCF produces more stable data — especially for monitoring. For longitudinal monitoring applications (fatigue tracking, season trends), SCF may be the preferred condition because tighter within-session CVs mean smaller genuine changes are detectable. The monitoring use case may benefit more than the profiling use case.
④ Account for sex differences in feedback response. mRSI showed a significant sex × condition interaction — men benefited more from SCF than women. Coaches should not assume feedback affects both groups identically, especially for jump strategy metrics.
⑤ Statistical vs. practical significance — use both lenses. Seven metrics were statistically significant, but only jump height and propulsive net impulse exceeded the SEM (practically meaningful). When athletes ask “was my jump actually better?”, the SEM comparison is the honest answer — not the p-value.