The collection of performance data in sport has been widespread for many years now, but with advances in technology and the process used to collect information, there is now an abundance of objective data available to both athlete and coach. The more information that can now be collected on performance can lead to a more accurate measurement of physiology pertaining that performance; although what also needs to be considered is that not everything that can be counted counts, and not everything that counts can be counted. In this respect it is essential to avoid paralysis through analysis by extracting only the information that is relevant, and that can be used to improve performance.
Global Positioning Systems (GPS) have been introduced to athletes in predominately team sports in attempt to obtain an objective analysis of both conditioning and performance training loads. This works via a small device tracking micro-movements in real time, providing information on accelerations, decelerations, velocity travelled, change in direction and total distance covered. A training load is then calculated from these variables, and can be used to manipulate training intensities that closely reflect game conditions. The device is fitted into a compression garment and sits between the athletes’ shoulder blades. Other methods used to collect training and game data include time motion analysis, through the use of video footage, but this is considered to be a time consuming method of collecting data, and may not always provide an accurate analysis of performance; therefore may be limited to providing players with performance feedback rather than physiological data. When conditioning athletes there exists a balance between minimum training load that elicits the desired physiological adaptation, and the maximum training load that results in unwanted results such as overtraining, illness and injury. In designing conditioning and rehabilitation programs that mimic game intensity, the physical requirements of the game must be measured; this can assist the sport scientist in identifying individual training load thresholds which may help prevent injury from overtraining, prescribing intensity levels for training sessions in order to periodise conditioning and targeting training methods that mimic the positional requirements of each player. This review of the current literature will focus on how GPS data can be used to quantify physical player loads in team sports and also the reliability and validity of GPS devices in collecting this data.
There is a fine balance between training, detraining and overtraining which needs to be considered when conditioning athletes for team sports. Colby, Rogalski & Gabbett (2014) found that excessive training load may increase injury risk, but also insufficient loads will have a similar effect. This study comprised of forty-six elite Australian Football players, calculating weekly cumulative training loads to then establish any links between increases in training loads with intrinsic injury incidence (soft tissue only, no impact injuries included). Training load was established through a combination of GPS data gathered from game and training sessions, combined with heart rate data and RPE from every game and training session. From GPS data collected, total three week cumulative distance covered of between 73km-86km during preseason increased injury risk by 5.5 times compared to distances <73km, and that sprint distance of between 864m-1453m had a lower injury risk than sprint distances >1453m. This indicates excessive load is indicative to injury but it must also be acknowledged that an insufficient load is likely to react the same. Therefore a certain level of load is required to protect against injury. The strongest relationship with intrinsic injury incidence was a monitoring of three weekly cumulative training load in preseason and two weekly load in season. Although GPS may quantify total distance covered in game, it may not always be a true reflection of game physiology. Coughlan, Green, Pook, Toolan & O’Connor (2011) successfully quantified the physical demands of an international Rugby Union game, but concluded that the data from GPS may underrepresent the overall demands due to the GPS device unable to collect data on static exertions of energy common in rugby, such as pushing, pulling wrestling etc. This study, although limited by the number of participants (two international elites) was also the first to report impact loads from tackles, via an accelerometer, with heavy (7-8G) to severe (10+G) loads being noted. This may be significant from a safety perspective in that these high loads can be monitored by medical staff, who can then prioritize medical assessment of these players when they leave the pitch. Using an identical procedure, Cunniffe, Proctor, Baker & Davies (2009) established that there was a change in tempo every 3-4 seconds, and maximum speed reached was 7.8km/h. The GPS data collected from this match indicated that a back covered more total distance and also entered the high speed zone (+20kmh) on more occasions than the forward. This data may indicate the relevance of individual conditioning programs that are position specific.
In a team sport environment, external training load tends to be same due to the prescription of group exercise for training. In this sense internal training load becomes very significant, as each individual will respond differently to differing physical demands. Over four games and nineteen training sessions Henderson, Cook, Kidgell & Gastin (2015) compared the difference in competition to training loads in junior elite Australian footballers. Quantifying training load through GPS, heart rate and RPE they found that players were exposed to higher external and internal loads during competition compared to training sessions. The data indicated that the players covered more distance and at a higher intensity during game play compared to training. This may be due to some sessions being lower intensity in order to facilitate acquisition of new skills and technical and tactical drills. The external load from GPS data collected during game play can assist in the facilitation in the optimum weekly physical load prescription, which can taper off intensity as game day approaches with the knowledge that external and internal loads will be maximized in game.
In order to obtain an accurate measure of the physical demands of game play, GPS devices used must be both reliable and valid in collecting data. Peterson, Payne, Portus & Dawson (2009) looked at the reliability and validity of two GPS devices with sampling frequencies of 1Hz and 5Hz, to monitor cricket specific movements. The average distance covered by an Australian Academy Cricket fast bowler was used to validate the device against a 400m athletics track. Over twenty trials the single participant completed the locomotion patterns involved in cricket (walking, jogging, running, striding and sprinting) over the average distance covered of 13000m. Both devices consistently underestimated short sprint distances, and over reported low intensity efforts. As distances increased then so did the accuracy of the device sampling at 5Hz. This may lead to an under representation of physical demands if high intensity changes of direction are being underestimated, leading to possible inaccurate training intensities being prescribed during conditioning. In a similar vein Rawstorn, Maddison, Ali, Foskett & Gant (2014) found that 5Hz sampling frequency devices were inefficient for quantifying distance during rapid direction change. This could misinterpret critical aspects of match play in that 340-380m was thought to be lost through rapid changes of direction during the study. Using the Loughborough Intermittent Shuttle Test (LIST) the six participants completed six blocks of fifteen minutes activity on a curvilinear track covering walking, jogging, running and sprinting, designed to replicated activity patterns in football. With a similar protocol but testing straight line running and change of direction running (zig-zag course) Jenkins, Cormack, Coutts, Boyd & Aughey (2014) used electronic timing gates to compare actual time with data collected from GPS devices ranging from 1Hz-5Hz. Twenty elite Australian Football players took part in straight line sprint tests of 10, 20 and 40m and a 40m 90 change of direction course. Reliability improved as distance increased but decreased as speed increased, indicating that care needs to be taken when interpreting data on single sprints or small changes in direction and velocity.
A higher sampling frequency of 10Hz was used by Varley, Fairweather & Aughey (2011) which provided sufficient accuracy to quantify acceleration, deceleration and constant velocity. Limitations to this study were that the low number of participants involved, were only measured in straight-line acceleration, deceleration and constant velocity. The accuracy observed in quantifying acceleration from this study would lead to the assumption that change of direction would also be accurately measured. Akenhead, Hayes, Thompson & French (2012) also used 10Hz sampling frequency GPS device to describe the distances covered during acceleration and deceleration in professional football. Significant findings revealed that 18% of total distance was covered while accelerating at a rate greater than 1m/s/s, with 7.5%, 4.4% and 3.3% of total distance covered at 1-2m/s/s, 2-3m/s/s and >3m/s/s respectively. This data indicates the importance of power and rate of force development work during conditioning sessions and an emphasis on eccentric strength to control high loads during deceleration. This data could also be helpful in setting intensity and volumes of sprint sessions that would closely mimic the game demands.
As previously mentioned accelerometers have been incorporated into certain GPS devices to attempt to extract more information from performance. As well as measuring impact forces from tackles, Boyd, Ball & Aughey (2011) also found accelerometers can be effective in positively detecting differences in physical activity levels. Testing was at 0.5-3.0G which account for 96% of acceleration values obtained in Australian Football, with devices presenting consistent data in between devices - indicating that players can be issued with any device rather than having to make sure the same one is consistently used.
From the literature reviewed it can be assumed that the use of GPS data can be used accurately to quantify data in regard to player training and game demands in team sports. What is evident though is that the higher the sampling frequency of the device (+10Hz), the more accurate data will be at short intense movements and changes of direction. Caution must be taken when interpreting GPS data in that total distance covered may not be a true reflection of game intensity. Peterson, Payne, Portus & Dawson (2009) established average fast bowler distances covered in cricket can be up to 13000m; in contrast to this Akenhead, Hayes, Thompson & French (2012) found that professional footballers covered a similar distance throughout performance. Both of these sports, although presenting similar distances covered may entail totally different intensity levels - therefore require different training protocols. A more relevant calculation would be ratio of high intensity bouts of work to lower intensity, indicating work to rest periods, which will aid in conditioning programs. Other limitations to using GPS are that the technology may not work so well under a roof. Therefore training session played inside or gym strength sessions may be hard to quantify. In this respect GPS should be used along with other variables to measure training load. Heart rate and RPE are the most logical. Using a modified Borg scale of 1-10 athletes can rate the intensity of the session, multiplying that score by the duration of the session in minutes. This will quantify a work load, and by calculating mean and standard deviation values, training strain and monotony values can be established to guide volume/intensity levels of sessions. What also must be considered is that GPS data does not necessarily measure global in-game demands, but rather player output. In this sense all data collected is unique to that individual and the respected team, so essentially a database of information would need to be gathered (for example over 1-2 seasons) including GPS data, heart rate and RPE to give a global picture of athlete activity profile. Only then can inferences be made in regard to training load and prediction of injury.
References
Colby,M., Dawson, B., Heasman, J., Rogalski, B., Gabbett, T. J. (2014) Accelerometer and GPS derived running loads and injury risk in elite Australian footballers. Journal of Strength and Conditioning Research, 28 (8) p2244-2252
Coughlan, G.F., Green, B.S., Pook, P.T., Toolan, E., O'Connor, S. P. (2011) Physical Game Demands in Elite Rugby Union: A Global Positioning System Analysis and Possible Implications for Rehabilitation. Journal of Orthopaedic & Sports Physical Therapy, 41(8) p600-605
Cunniffe, B., Proctor, W., Baker, J.S., Davies, B. (2009) An evaluation of the physiological demands of elite rugby union using global positioning system tracking software. Journal of Strength and Conditioning Research, 23 (4) p1195-1203
Henderson, B., Cook, J., Kidgell, J., Gastin, B. (2015) Game and Training Load Differences in Elite Junior Australian Football. Journal of Sports Science and Medicine 14 (3) p494-500
Peterson, C., Payne, D., Dawson, B., Portus, M. (2009) Validity and reliability of GPS units to monitor cricket-specific movement patterns. International Journal of sports Physiology and Performance. 4 p381-393
Rawstorn, J. C.,Maddison, R., Ali, A., Foskett, A., Gant, N. (2014) Rapid Directional Change Degrades GPS Distance Measurement Validity during Intermittent Intensity Running. PLoS ONE 9(4)
Jennings, D. Cormack, SJ. Coutts, AJ. Boyd, L. Aughey, RJ. (2010). The validity and reliability of GPS units for measuring distance in team sport specific running patterns. International Journal of Sports Physiology and Performance, 5 (3) pp328-341
Varley, M. Fairweather, I. Aughey, RJ. (2011). Validity and reliability of GPS for measuring instantaneous velocity during acceleration, deceleration and constant motion. Journal of Sport Siences, 1 (7)
Akenhead, R. Hayes, PR. Thompson, KG. French, D. (2013) Diminutions of acceleration and deceleration output during professional football match play. Journal of Science and Medicine in Sport, 16 (6) pp556-61
Boyd, L. J. Ball, K. Aughey, R. J. (2011) The Reliability of MinimaxX Accelerometers for Measuring Physical Activity in Australian Football. International Journal of Sports Physiology and Performance, 6 (3) pp311-321