An examination of the training loads within elite professional football
The popularity of soccer throughout the world has led to the demand for a scientific approach to the preparation of players for competitive matches. Although previous researchers have attempted to understand the training demands undertaken by soccer players, limited information is known regarding the structure of training in soccer. At present research has focused on the frequency and duration of soccer training without using both objective and subjective measures of training load to systematically evaluate training practices in elite teams. Little is also known regarding the periodisation strategies employed by elite soccer teams across a competitive season and whether they follow traditional models of periodisation. With this in mind, the primary aim of this thesis is to therefore characterise the current training periodisation practices that exist in elite soccer using applied methods of training load assessment. The aim of the first study (Chapter 3) was to evaluate the use of Global Positioning Devices (GPS) for the measurement of soccer-specific activities to provide objective data for training load assessment. Findings from this study were applied to study 3 (Chapter 5) of the thesis. Firstly, a soccer-specific movement course was designed based on the movements exhibited by an elite soccer player during a competitive match using a multi-camera tracking system (ProZone®). Two moderately trained males performed 10 bouts of the soccer-specific track following familiarisation and a 10 minute standardised warm up. Both subjects wore two 10Hz GPS units inside a custom-made vest during all bouts of the track to determine both reliability and inter-unit reliability of the GPS devices. Data analysis revealed the reliability of the GPS devices was good for distance covered at lower velocities (0 – 4 m/s; CV% = 0.6 – 3.6%). However when the velocity of movement increased (> 4 m/s), the reliability of the units decreased (mean change from 13.8 to 33.6 CV%). Both total distance (mean CV% = 1.1%) and max speed (mean CV% = 2.7%) were both found to be highly reliable variables. However the devices demonstrated high levels of inter-unit reliability error due to an increase in systematic error with random distribution of data points between both devices for all variables measured. The data suggested that 10Hz GPS devices are reliable for the measurement of lower velocity (0 – 4 m/s) running. However, care must be taken when analysing data in higher velocity bands (> 4 m/s) due to the high ii error rates observed. The high inter-unit reliability error also suggests that 10Hz GPS devices cannot be used interchangeably between players in order to minimise the associated error. The aim of the second study (Chapter 4) was to quantify the reliability and validity of a portable vertical jump assessment tool (Optojump®) for use in the applied setting. Vertical jump assessment was utilised as a measurement tool to analyse the effect of training load on the neuromuscular system that was evaluated in study 4 (Chapter 6) of the thesis. Eleven healthy male subjects were familiarised to perform four separate common types of vertical jump test: countermovement with arm swing (CMJ-W), countermovement without arm swing (CMJ-WO), squat jump (SJ) and drop jump (DJ). Contact time, flight time and jump height were selected as variables for the study. For reliability assessment, all subjects performed 3 efforts of each jump type across 5 identical testing sessions (separated by minimum of 2 days). For validity assessment, subjects were asked to perform the same jump modalities as the previous investigation on one occasion while data was simultaneously collected from both a force plate (criterion instrument) and the Optojump photocells. The data revealed the Optojump device was highly reliable for the assessment of jump flight and height for CMJ-W, CMJ-WO, SJ and DJ (all CV% = 3.2 and 5.6%). However reliability of the device was reduced for the measurement of contact time with the DJ (CV% = 13.9%). Validity data revealed that all jump types and variables were highly valid in comparison to the force plate criterion measure (SEE% = < 1%, Pearsons correlation = r > 0.99). This study revealed that the Optojump device is highly reliable and valid for all jump types and variables, with the exception of contact time for DJ. Therefore the Optojump system may be used with confidence to detect within-group changes in applied assessments of vertical jump performance. Due to the high cost and lack of portability of laboratory-based force plates, the Optojump system is a viable alternative for accurate jump measurement and neuromuscular assessment. The CMJ-WO jump assessment was chosen for study 4 for comparison with previous research. The aim of the third study (Chapter 5) was to quantify the periodisation strategies employed by an elite professional soccer team throughout a competitive season. Training load data was collected from 37 elite outfield soccer players at one professional English soccer team over a 45 week period during the 2011-2012 domestic season. All players wore iii global positioning system (GPS) devices, heart rate (HR) belts and were asked to provide a rating of perceived exertion (RPE) for each training session to generate training load data. Players were assigned to one of 5 positional groups: central defender (CD), wide defender (WD), central midfielder (CM), wide midfielder (WM) and attacker (AT). The data was separated into the pre-season (6 weeks duration) and in-season (39 weeks duration) phases in order to investigate specific training periods recognised within the annual plan. The pre-season phase was further separated into weekly blocks for analysis of the structure employed in each specific microcycle. The in-season phase was divided into 6 x 6 week blocks for analysis of mesocycle structure. Within the in-season data, three separate microcycles (weeks 7, 24 and 39) were selected consisting of the same weekly training schedules to determine whether differences in microcycle training load pattern existed. In addition, the training data within a given microcycle was analysed to investigate the loading patterns in relation to number of days away from the competitive match fixture. Linear mixed modelling analysis revealed significant differences for total distance and average HR (P < 0.05) between period 1 with periods 3 and 6 during training mesocycles. However no differences were found for the remaining training variables during both pre-season and in-season microcycles (P > 0.05). Training load variables were significantly reduced on match day (MD) -1 (P < 0.05) but remained similar across MD-2, MD-3 and MD-5 (P > 0.05) during in-season microcycles. CM players generally covered the most total distance compared to other positions. Defenders reported higher internal load values (average HR and RPE) compared to attackers during in-season training phases but such differences were not evident during pre-season. This study revealed that training load doesn’t appear to be systematically periodised across a competitive season in an elite soccer team. This may have practical implications for training planning, as monotonous training load prescription may lead to maladaptation in soccer players during a competitive season. This was the first study to systematically evaluate periodisation strategies in an elite soccer team, but further work is required to determine such practices at different soccer teams. The aim of the fourth study (Chapter 6) was to determine the neuromuscular response to a microcycle of soccer training in elite soccer players using vertical jump assessment via the Optojump device. Nine elite level youth soccer players from an U18 soccer academy team were recruited for the study. The players underwent four separate on-field soccer training sessions following familiarisation of all testing procedures.