Maximal Strength Training in High-level Female Football Players

In male football players, studies have reported a favorable effect on both 1RM, sprint and jumping performance following MST. Although females and males possess diverse levels of anabolic hormones, they do in general respond similarly after training interventions in most strength outcomes. However, there are reports of a larger relative increase in females compared to males when the same ST is applied.

In this cluster-randomized controlled trial, two football teams (playing at level 2 and 3 in Norway) was invited to participate. The study were conducted during the last part of the pre-season preparations, ending one week before first seasonal competition. The training group (TG) performed MST training carried out as free-barbell squats twice a week over five weeks in addition to the planned pre-season training, while the control group (CG) was instructed to perform their originally planned pre-season training.

The total sample comprised 46 players aged 15-26 years, where two separate football teams were cluster-randomized to either TG or CG. The two teams played at level two and three in Norway, where level two is a national league and level three a regional league in Northern Norway. Inclusion criteria was that the players perceived themselves as injury free and able to complete the strength training. Players were only excluded if having injuries that made strength training, running and jumping unachievable. The players carried out ̴ 6.5 h training per week with their team. Four players were injured, two did not complete the required amount of training, one withdrew due to time limitations and five withdrew without providing any reason resulting in 19 participants in TG and 15 in CG that completed both pre- and posttests, and were included in the analyses for training effect.

According to the declaration of Helsinki, all participants were fully informed of the potential benefits and risks of the study, both orally and written, before signing an informed consent. For participants under 16 years, both the players and their parents gave their written informed consent. The participants were fully informed of their rights to withdraw from the study at any time without providing any reason.

All testing and training sessions were conducted in an exercise training laboratory at Alfheim Stadium, Tromsø. Prior to the intervention, the players underwent baseline tests over two test days, with a 72 hour washout period to avoid any detrimental effects from the preceding test day: day 1) measurement of body mass and body height, 5-, 10- and 15m sprint time and a counter-movement jump (CMJ), day 2) 1RM in a free-barbell squat exercise with partial 90° knee angle range of motion (ROM).

Prior to the tests, the participants were asked to refrain from heavy training the preceding day, and to arrive in the laboratory well-hydrated. All tests and training sessions started in the afternoon, with the same general warm-up routine: 7 min of self-selected low intensity cycling on an ergometer bike (Pro/Trainer, Wattbike Ltd, Nottingham, UK) followed by 7 min low intensity running of self-selected speed on artificial grass.

On day 1, following the general warm-up and three 15 m strides on a sprinting field, a 15-m sprint test was carried out. Data were assessed in 5 m splits by photocells mounted to the floor and walls (ATU-X, IC control AB, Stockholm, Sweden) using single-beam electronic barriers. The within-subject coefficient of variation is 2% for this measurement. The surface consisted of artificial grass, and the players wore their own running shoes. The sprints started with the players in a static position placing their front foot 30 cm behind the starting line. A timer was triggered by the participant breaking the initial sensor. The rest interval between the single sprint trials was 180 seconds. The fastest sprint time of three trials was carried forward for further analyses.

Thereafter, the players rested for 5 minutes prior to performing the CMJ test. CMJ was assessed by a portable force platform, with a validity within 1 centimeter (2%) when compared with the gold standard mounted floor force platform, and a within-subject coefficient of variation of 2.8%. Force data were recorded by a software. This device records only the vertical ground reaction force at a sampling frequency of 1200 Hz and jump height is automatically calculated by software applying double integration of the force signal through Simpson's rule of integration. The players were instructed to keep their hands placed on the hips and the feet shoulder-width apart. Each player performed two trials with a ≥180 second rest between sets. The highest jump was carried forward for further analysis. Day 1 was ended with a familiarization trial for the squat exercise with low loads.

On day 2, the players returned to the laboratory for the assessment of maximal strength as 1RM. The session was initiated with the same general warm up routine as mentioned above. An Olympic barbell and a suitable rack was applied for testing of 1RM. The ~ 90° knee angle of each participant was measured during every repetition using a goniometer, and the players were given an orally "go" when being allowed to start the concentric phase of the lift. Prior to starting their 1RM attempts, the participants warmed up with 10 repetitions with a low load of - 50% 1RM (subjectively assessed by the instructor). The starting 1RM attempt was an initial acceptable load decided by the instructor. Each 1RM attempt was carried out by a single repetition, with increasing load of 5-10 kg until they failed to execute the 1RM attempt, which on average was five trials. Each attempt was interspaced by ≥180 s of rest. The within-subject coefficient of variation for squat 1RM is 2.9%.

The players attended supervised training in the laboratory twice a week for five weeks. The training session started with the general warm-up routine described above, before starting the strength training. The program consisted of 90° squats, carried out in the same way as in the 1RM test. The squat training was initiated with three sessions of three sets of six repetitions, followed by seven sessions of four sets of four repetitions. The repetitions were carried out with a slow eccentric movement followed by maximal mobilization in the concentric phase. 180 s of recovery was given between each set. The load was initially set at 85% of pre-test 1RM, which the participants increased with 2.5-10 kg if they could manage more than six or four repetitions, depending on their scheduled program, resulting in a consistent overload during the whole intervention. Weight lifted for each repetition was logged continuously during the study. Additionally, for ethical reasons, in order to avoid hamstring strains due to an anticipated large agonist-antagonist strength ratio following the intervention, three sets of six repetitions of the Nordic hamstring exercise were performed after the squat exercise for each session with a ≥180 s rest period between sets.