*2.3. Intervention*

The control condition of the typical warm–up (CON), lasted approximately 8 min and consisted of 3 min jogging at a low–medium tempo, followed by dynamic stretching exercises for the lower limbs (Table 1). More specifically, the first 7 exercises were performed for a 10-m distance and after the end of each exercise the participant walked back to the starting point. Dynamic stretching was preferred to static, to eliminate any potential adverse e ffect in performance [36].

**Table 1.** The dynamic stretching exercises performed for 10 m after the 3 min jogging.


The two other conditions consisted of three additional sets of 10 m maximal BwR or FwR sprints. Participants returned to the starting position running forward at a low pace. Subjects received verbal encouragemen<sup>t</sup> during BwR and FwR to ensure that the conditioning stimulus was maximal.

#### *2.4. 20 m Sprint Test*

Sprint time over 5, 10 and 20 m was measured during the 20 m sprint (Table S1). For this purpose, three photocell timing gates (Witty Wireless Training Timer, Microgate, Bolzano, Italy) were placed at 5, 10, and 20 m. Photocells were adjusted to the pelvis height of each participant [37]. Participants were instructed to start after the verbal signal "ready, go". They stood on an upright stride stance position, with their preferred foot forward, placed on the starting line over a pressure pad. Timing started when participants' foot was detached from the pressure pad. The assessor ensured no false steps before starting and correct starting posture before the start. During the sprints verbal encouragemen<sup>t</sup> motivated for maximal e ffort. Sprint speed was analyzed for the distances 0–5, 5–10, 0–10, 10–20, and 0–20 m and was calculated by dividing the running distance by the time.

#### *2.5. Rate of Perceived Exertion (RPE)*

RPE was measured immediately after the completion of each intervention (Table S1), using the 10-degree Children's OMNI scale [38]. Participants replied to the question "*how tired do you feel?*", while the investigator showed them the schematic OMNI scale. Participants had to declare the requisite exertion by indicating a number on the scale from 0 (not tired at all) to 10 (very, very tired). During the last two sessions at the sports club they were familiarized with the scale. This included a thorough description and explanation of the scale and responding to any questions or doubts that they had.

#### *2.6. Statistical Analysis*

All data are presented as means and standard deviations. The dependent variables were the sprint speed for the distances of 0–5, 5–10, 0–10, 10–20, and 0–20 m and the RPE. The Shapiro-Wilk test was used to confirm the normal distribution of the data (*p*-values ranging from 0.126 to 0.850 among all variables), and Levene's test for the equality of variances (*p* = 0.368–0.787). Furthermore, Mauchly's test was performed to confirm that the assumption for sphericity was satisfied (*p* = 0.067–0.641). One-way Analysis of Variance (ANOVA) for repeated measurements was used for the statistical assessment to examine the e ffect of warm-up protocol (three levels: CON, BwR and FwR). The level of significance α was set at 0.05. Statistically significant e ffects were assessed with the Sche ffé's post-hoc test. The e ffect sizes were calculated using eta squared (η2). The one sample t-test was used to examine the change in percent of sprint performance during the BwR or FwR relative to the CON condition compared to baseline zero. Confidence intervals at 95% confidence level (CI95%) were constructed. Statistical analysis was performed with SPSS for Windows, version 25 (IBM Corp., Armonk, NY, USA) and custom scripts in R, version 3.6.1 (R Foundation for Statistical Computing, Vienna, Austria).
