Monitoring changes in jump and sprint performance: best or average values?

This study compares different approaches to monitor changes in jump and sprint performance while using either the best or the average performance of repeated trials. One hundred two highly trained young footballers (U13 to U17) performed, in 2 different testing sessions separated by 4 mo, 3 countermovement jumps (n = 87) and 2 sprints (n = 98) over 40 m with 10-m splits to assess acceleration (first 10 m) and maximal sprinting speed (best split, MSS). Standardized group-average changes between the 2 testing periods and the typical error (TE) were calculated and compared for each method. The likelihood of substantial changes in performance for each individual player was also calculated. There was a small increase in jump performance (+6.1% for best and +7% for average performance). While 10-m time was likely unchanged (+~1.2% for both best and average performance), MSS showed likely small improvements (+~2.0% for both best and average performance). The TEs for jumping performance were 4.8% (90% confidence limits 4.3;5.6) and 4.3% (3.8;5.0) for best and average values, respectively; 1.8% (1.6;2.1) and 1.7% (1.5;1.9) for 10-m time and 2.0% (1.8;2.3) and 2.0% (1.8;2.3) for MSS. The standardized differences between TE were likely unclear or trivial for all comparisons (eg, 10-m, 0.01 [-0.09;0.10]). The numbers of players showing a likely increase or decrease in performance were 30/0 and 29/0 for best and average jump performances, 9/4 and 12/2 for 10-m times, and 33/4 and 33/4 for MSS. In conclusion, the 2 monitoring approaches are likely to provide similar outcomes.