Agility, the ability to change direction rapidly and accurately in response to a stimulus, is a decisive physical component in football. Preliminary observation of the football extracurricular activity at MTsN 5 Tanah Laut indicated that the students’ agility was still relatively low and that the training provided tended to be conventional and monotonous. This study aimed to determine the effect of game-based agility training on the improvement of students’ agility. A pre-experimental one group pretest-posttest design was used. The sample consisted of 18 students selected through total sampling, comprising the entire population of the activity. Agility was measured using the Illinois Agility Test, and each participant performed two trials with the fastest time recorded. The treatment, namely game-based agility training, was delivered for four weeks at three sessions per week with a progressive load. Data were analysed using descriptive statistics, the Shapiro-Wilk and Levene tests, and a paired sample t-test, complemented by the effect size (Cohen’s d) and the 95% confidence interval of the mean difference. The mean completion time decreased from 16.95 seconds in the pretest to 15.66 seconds in the posttest, a mean difference of 1.29 seconds (95% CI = 1.07 to 1.51). The paired sample t-test yielded t = 12.324 with a significance of 0.000 (p < 0.05) and a very large effect size (Cohen’s d = 2.90). Therefore, there was a significant effect of game-based agility training on the improvement of students’ agility. Within the limits of a single-group design, this training can serve as an effective and enjoyable alternative for developing the agility of school-age football players.

Keywords: agility; extracurricular; football; game-based agility; Illinois Agility Test