After several postponements due to different reasons (e.g., COVID lockdowns, mismatch between research team availability and club's availability), this study will no longer be performed.

The goal is to assess inter- and intraindividual variation in response to different warm-up protocols in a randomized, cross-over study with U16 and U17 Portuguese soccer players.

Trial design We are following the Consolidated Standards of Reporting Trials (CONSORT) 2010 statement: extension to randomized crossover trials (Dwan, Li, Altman, & Elbourne, 2019). Randomized crossover designs increasing statistical power, as participants act as their own controls (Dwan et al., 2019). Here, a randomized crossover design with two arms is elaborated. A two-week familiarization period with the two interventions and with testing will be performed, during which the participants will make four visits, in all of which they will go through pre-testing, protocols A and B (on alternate days), followed by post-testing. This period will be followed by two-weeks of performing each intervention, separated by a two-week wash-out period to avoid a carryover effect (Higgins et al., 2019), which is superior to that used in similar research (Barbosa et al., 2019). In each period, the groups will perform four visits in terms of intervention, and in two of those visits they will be tested with the goal of assessing within-group, within-intervention consistency. Allocation ratio will be ~1:1. The protocol will be sent to approved by the Ethics Committee of the Faculty of Sport of the University Porto, and then registered at http://www.clinicaltrials.gov. Participants will be informed of the goals and methods of the research and their legal tutors will sign informed consent forms (as they will be <18 years-old), following the updated version (2013) of the Declaration of Helsinki.

Participants Participants will be 40 to 50 healthy football players from the U16 and U17 teams of a club from the Oporto city, engaged in three weekly training sessions with an average weekly training volume of 270 minutes. Following similar research (Beato, De Keijzer, et al., 2019), participants will have to be free from injuries or illness at the start of the study. Data will be collected in the facilities of the club where the participants play, and therefore no familiarization to the environment will be required.

Interventions In line with previous research (Beato, De Keijzer, et al., 2019; Faude, Roth, Di Giovine, Zahner, & Donath, 2013), a 10-minute general warm-up will precede the implementation of the specific warm-up protocols. The general warm-up will consist of three minutes running at low to moderated intensity, followed by three minutes of active, dynamic mobility work, followed by approximately four minutes of 20-m accelerations at 70% of maximum perceived intensity, where each player will perform 5 repetitions of increasing distances (i.e., 10 m, 15 m, 20 m, 25 m, 30 m). During the pre-tests, general warm-up, interventions and post-tests, players will wear Global Positioning System (GPS, VX Sport, Model VXLOG 175).

The interventions will consist of two different warm-up strategies. In the first two weeks of implementation, one group will perform intervention A while the other will perform intervention B. After the wash-out period, the groups will crossover (AB/BA design). In each two-week period of the interventions, the participants will make two weekly visits, in a total of four visits per period. In the first visit of each week, the participants will only perform the interventions, while in the second visit of each week, the participants will also engage in pre- and post-testing. Therefore, two tests per period will be applied.

Intervention A is based on football-specific small-sided games (SSGs), consisting of 4 versus 4 matches (with mini-goals), performed in a 30 m by 24 m court (Giménez et al., 2018), using five sets of 90 seconds of work interspersed with 30 seconds of pause, in a total duration of 10 minutes. Intervention B is based on speed, COD and plyometrics (i.e., a "potentiation" protocol), inspired by ideas presented by Howe, Coward, and Price (2017). Participants will perform five repetitions of the following exercises, in this order: unilateral hurdle hops and lateral hurdle hops with each lower limb; countermovement jump; broad jump; forward in and out; slalom jumps; lateral scissor jumps; single leg linear hops. Each intervention will last ~10 minutes and will be applied in the football field. Two experienced football coaches with a degree in Sports Sciences will deliver and monitor all the sessions. The interventions were designed to be easily applicable by coaches without the need for sophisticated facilities or material. All interventions will be implemented at the same time of the day and in the same days of the week for both groups. Due to the nature of the interventions, participants and implementers will not be blinded.

Sample size The software G*Power (version 3.1.9.6. for Mac) was used to assess a priori sample power. Considering a repeated-measures ANOVA, for within factors, with a power of 0.95, α = 0.05, correlation coefficient of 0.5, non-sphericity correction of 1, effect size of 0.35, two groups and four measures (i.e., outcomes), minimum sample size was established in 20 subjects. For also analyzing within-between interactions, 36 subjects were required, which is still below our estimated sample of 40-50 participants.

Randomization In this crossover study, some athletes will perform an AB sequence, while others will perform a BA sequence. It is predetermined that each group will have ~50% of the participants, without any further rules being stipulated for randomization. Randomization will be performed only after the two-week familiarization periods and the pre-tests. The random allocation sequence will be performed by a researcher not involved in testing or in monitoring the interventions. Two opaque bags will be used: (i) the first bag containing folded papers with the names of the participants; (ii) the second bag containing the intervention (i.e., A or B). In each round, the researcher will remove a paper from the second bag, followed by a paper from the first bag. Allocation will be concealed from all other researchers until the start of the intervention. The two main testers will not be provided with this information, to ensure they remain blinded. If allocation suggests problems with baseline values, creating imbalances in the groups with respect to the values obtained in the four tests, the randomization process will be repeated.

Inclusion Criteria:

• Participants will be 40 to 50 healthy football players from the U16 and U17 teams of a club from the Oporto city, engaged in three weekly training sessions with an average weekly training volume of 270 minutes.

Exclusion Criteria:

• Following similar research (Beato, De Keijzer, et al., 2019), participants will have to be free from injuries or illness at the start of the study.

Testing will be blinded to the interventions and to the goals of the current research. Two external professionals, with a master's degree in Sports Science and experienced in collecting the data will be responsible for collecting the tests at all time points. In each test, they will be further blinded to the previous results. Additionally, two researchers will assist in operationalizing the tests; these researchers will be blinded to the participants' allocation, but not to the study's goals.