Study aim: The aim of the present study was twofold: firstly, to examine the effect of age on a 20 m sprint performance; and secondly, to establish normative data for the 20 m sprint performance by age in football players.

Material and methods: The anthropometric characteristics of 474 football players (aged 16.81 ± 5.35 yrs, range 9.02–35.41 yrs) were examined and their 20 m sprint performance (with 0–10 and 10–20 m splits) was monitored by a photocell system (Brower Timing Systems, Utah, USA).

Results: A one-way analysis of variance revealed significant differences between the yearly age groups with regards to the sprint time (p < 0.001, η2 = 0.584), as well as the 0–10 m (p < 0.001, η2 = 0.361) and 10–20 m split times (p < 0.001, η2 = 0.635). The older groups scored better than the younger groups. The time attained in the 20 m sprint, and the 0–10 m and 10–20 m splits correlated moderately to largely with the athlete’s age (r = –0.53, –0.40 and –0.57, respectively, p < 0.001).

Conclusions: In summary, the speed ability of the football players improved with age until 15 years old, where it reached its peak. On the other hand, the other age groups U16 to U35 revealed no major differences in the speed over a 20 m sprint. The reference values presented in this study might help football coaches and fitness trainers in monitoring training and in the selection of players. Moreover, since this is the first study of this kind to compare adult age groups, sport scientists focusing on relevant topics might use it as a reference in future studies.