Comparison of a smartphone App with other GPS devices

A Comparison of a smartphone App with otherGPS tracking type devices employed in footballPeter Tierney 1, and Neil Clarke 2

1Birmingham City Football Club, and 2Coventry University

Mobile App | GPS device | Tracking | Football

Headline

Individually worn GPS tracking devices have become com-monplace in professional football (1) to track player per-

formance. However they are restricted to few teams due tocost and complexity (2). Football activity has been shown tohave many health benefits (3). With an estimated 500 mil-lion participants worldwide, being able to quantify more ofthis populations activity involvement would help inform par-ticipants and practitioners alike. Any App developed needsto be validated due to consumer wearable technology beingquestioned (4).

Aim.To evaluate a mobile application in a range of mobiledevice as a reliable method to gather physical data and tocompare against other tracking devices used in football. Fur-thermore to evaluate use and comparison with other trackingdevices in a football environment

MethodsExperimental Approach of the Problem.The study was de-signed in two parts, firstly to validate a range of popularsmartphones (Table 1) both Android and Apple IOS type de-vices with an application installed that tracks movement thatis available at the APP store and Play store entitled Quantrax() and a range of GPS tracker type devices; Viper (Stat sportsNewry, Northern Ireland), Playertek (Catapult group Aus-tralia) Polar (Polar Electro Warwick, UK), Axsys (AxSys per-formance Canberra, Australia).These all transported over a 20meter shuttle course to a total distance 200 meter to quan-tify total distance covered in meters, various speeds in km/hrclassified as walking ≤6.5km/hr, jogging ≥6.6km/hr and an

Table 1. Mean (90% CI) values for perceived fatigue at restbefore (Pre), after the overload training period (Mid) and atthe end of the taper (Post) in both groups.

SmartphoneBrand

Manufacturedetails

Year ofmanufac-

ture

HTC ONEHTC corporation

Taiwan2014

Iphone 5se Apple inc USA 2017

Iphone 5 Apple inc USA 2016

Samsung FameSamsung Electronics

South Korea2013

Samsung FameSamsung Electronics

South Korea2015

Samsung Young 2Samsung Electronics

South Korea2015

Fig. 1. Garment pocket for housing of devices

Fig. 2. Pocket displaying a GPS tracker device and smartphone in situ

Fig. 3. Positional heat map performing shuttle test activity

activity positional heat map (fig 3). Secondly a Smartphoneand GPS tracker were tested in a small sided football environ-ment on all-weather 4th Generation AstroTurf playing surfacesmall sided football Pitch measuring 36 x 27meters boardedall round with a goal at each end. Results evaluated for com-parison of common metrics used in football of total distancecovered, various speeds, maximum speed reached and an ac-tivity positional heat map.

Athletes. Full time professional football players (n = 10, witha mean age 19 ± 1.8 years, height 177 ± 6 cm, body mass

sportperfsci.com 1 SPSR - 2018 | Nov | 39 | v1

Comparison of a smartphone App with other GPS devices

Table 2. Smartphone and GPS tracking device shuttle test.

Walking Jogging

DeviceTotaldistance(m)

Totaldistancewalking (m)

Totaldistancejogging (m)

Totaldistancerunning (m)

Max speed(km/hr)

Totaldistance(m)

Totaldistancewalking (m)

Totaldistancejogging (m)

Totaldistancerunning (m)

Max speed(km/hr)

HTC one 98.00±3.41 72.17±5.49 24.83±5.27 0.00 6.67±0.52 99.83±2.23 7.83±2.64 45.67±10.56 46.17±8.75 14.17±0.75

Iphone 5 se 95.00±7.81 61.67±6.51 33.33±4.04 0.00 7.00±0.63 91.67±18.77 12.33±8.39 34.00±6.56 45.33±33.38 14.33±0.63

Iphone 5 96.67±2.31 54.33±13.2 41.67±10.6 0.00 6.90±1.00 100.67±3.79 9.67±5.51 53.00±22.07 38.33±30.09 13.60±0.58

Fame a 88.33±9.50 68.33±12.66 20.00±16.37 0.00 7.40±0.41 94.33±9.07 8.67±2.08 41.33±5.13 43.67±16.17 13.67±0.48

Fame b 88.00±7.55 60.33±15.89 28.00±16.09 0.00 6.67±0.78 94.67±8.96 8.33±1.53 39.67±22.74 47.00±28.62 14.25±1.25

Young 2 94.67±4.93 58.33±24.68 36.00±28.83 0.00 6.60±0.58 86.33±5.51 7.33±3.21 42.00±36.72 37.00±34.70 13.12±1.00

Polar 104.63±2.41 74.50±28.45 30.13±30.52 0.00 6.47±0.43 100.70±4.91 8.80±3.36 58.08±11.19 33.82±8.30 14.37±0.34

Viper 99.18±2.15 73.73±15.65 25.44±14.39 0.00 6.50±0.48 100.87±3.04 8.07±2.41 64.57±30.12 28.23±30.37 14.29±0.95

Playertek 97.00±3.46 6.37±0.74 97.00±3.46 13.8±0.78

Axsys 98.09±1.27 67.80±17.69 30.29±18.95 0.00 6.98±0.54 102.16±3.77 8.54±2.18 52.77±11.49 40.85±7.07 14.68±0.64

Data are presented as mean ± SD.

Table 3. GPS tracker device and smartphone tracking in 5 aside football match mean and SD over 4 games

DeviceTotaldistance(m)

Totaldistancewalking (m)

Totaldistancejogging (m)

Totaldistancerunning (m)

TotaldistanceHigh speedrunning (m)

Maxvelocity(km/hr)

GPS Device 484.83±68.99 83.29±82.88 131.29±46.82 229.82±173.43 39.78±15.85 22.27±190

HTC one 476.88±54.12 100.33±90.61 159.25±65.99 202.25±170.95 15.05±13.17 23.34±1.91

Fig. 4. Comparison of GPS Device and Smartphone in 5 aside football match play over 4 x 5 minute games

77.6 ± 4.3 kg estimated body fat percentage of 7.2 ± 1.2%)respectively participated in this study. Participants completeda health screen questionnaire prior to the study, in additioneach participant’s capabilities to participate in physical activ-ity was assessed by club Doctor and qualified Physiotherapist.

Design. Part 1: Six mobile smartphones with same applicationinstalled and 4 GPS tracking devices. On a clear sunny day ona hard standing area 2 cones were placed 20 meters apart Asubject carrying all in a purpose built aid that housed all de-vices, first walked between each cone 5 times turning at each

cone 180°, travelling in total 100 meters. This was then re-peated at a faster pace to replicate jogging, the test was thenrepeated on 2 more occasions and an average for each deviceover the 3 sessions was calculated. On completion of the testeach smartphone and device had data collected which was thencropped in order to only collect data from when performingthe test for later analysis (Table 2).

Part 2: 10 professional football players were grouped into 2teams of 5, with each team defined by the wearing of a colourtop for each team. A player was then selected from one ofthe teams to wear the GPS tracker unit with a smartphone

sportperfsci.com 2 SPSR - 2018 | Nov | 39 | v1

Comparison of a smartphone App with other GPS devices

in purpose built garment as used in previous tests (fig 2). Allparticipants were experienced in the wearing of the vests andunits as they wear for all football training as well as Matchplay, player wearing the two devices did not complain of anyissues nor did it impede in any way their normal range ofmovement or performance from the result of wearing of vestsand fitted units. Play commenced in a 5v5 small sided gameformat, 4 x 5 minute periods with 2 minutes rest betweeneach period. On completion, data was uploaded for analy-sis. Data was cropped so as to only use the actual in gamedata for each 5 minute period. Data collected for analysis fromboth devices included: total distance covered measured in me-ters, Total distance covered at various speeds measured in me-tres, classified as walking ≤6km/hr, jogging 6.1-12km/hr, run-ning 12.1-18km/hr, high speed running ≥18km/hr, maximumspeed (km/hr) and an activity positional heat map.

Statistical analysis. Analyses were conducted to determine thereliability and validity of distances travelled in classified speedzones. Data was presented as mean ± SD and 95% confidenceintervals. First, to analyse the reliability of the smartphoneapp used for measuring the total distances and various speedzones employed, and to analyse with other types of smart-phone coefficient of variation (CV) were used. Secondly toanalyse the smartphone app with GPS tracking device for sig-nificant differences. Third to calculate the concurrent validity,the bivariate Pearson product moment correlation coefficient(r) was used.

ResultsThere were no significant differences between types of smart-phones in running (partial eta squared: 0) and high-speed run-ning (partial eta squared: 0) distance, only small differencesin walking (partial eta squared: 0.22) and jogging (partial etasquared: 0.26) distance and a small difference in total distance(partial eta squared: 0.35). Furthermore, only trivial differ-ences in maximum speed were observed (partial eta squared:0.11). Smartphone app and GPS tracking device in shuttleruns showed a high degree of correlation (r = 0.94-0.99) andin match play (r = 0.90-0.99).

DiscussionThe major findings of this study support studies that reportedthat a smartphone application can reliably monitor physicalactivity (5,6). This present study has gone further than pre-viously published work in the area by assessing the validityand reliability across a range of smartphones, on what differ-ences there are and comparison with a range of current GPStracking devices used in sport and in a sport specific envi-ronment. In keeping with results in previous studies (7) thepresent study found that GPS tracking type devices are a re-liable and valid method in quantifying physical activity, butwith varying levels of accuracy (8).

The use of smartphone technology to measure physical ac-tivity is nothing new as there has been objective studies thatsupport the use of in this way and reported that these were re-liable and valid (9). Therefore with smartphones, along withGPS tracker devices, continually developing more advancedhardware and sensors (10,11), this method will only becomemore able to better report more types of physical activity inaddition to those detailed within this present study. By nothaving the need for additional technology, other than the in-stallation of a suitable application, makes a compelling argu-ment for the smartphone to be used as a reliable alternative

or to supplement GPS tracker devices already in use. Thelatter of these being of value where team systems are usedand not allowing for individual training sessions away from acentralised hub.

Football continues to increases in popularity recent esti-mates that over 500 million worldwide participate (3). TheUK alone has over 11 million registered participants (12) atvarious levels from recreational to Elite level. There has beena large numbers of studies that have identified the physical de-mands imposed at the Elite 11 aside level (2, 13) yet there isstill very little known in comparison to other formats and lev-els across the participation landscape. Small sided football isone of the most popular formats of the game as it is employedacross the football landscape (14). Elite level teams use it asa tactical and physical conditioning training tool and there isevidence that it is utilised at all levels (15). In a recent surveyit was reported that there are over 1.5 million adults partici-pating in England each week and over 30,000 registered teams(14). It is estimated that number has increased due in part tothe growth of numerous facilities housing commercial leisureleagues. One example of the many commercial enterprisespromoting this type of football in the UK has over 400 venueswith in excess of 5,000 registered league teams and over 70,000regular users. There has been some studies that have looked atphysical activity in non-elite football (3) these conclude thatwhilst participation does improve health there is a need forquantitative data to better understand the activities involvedand the exact health benefits (16). Using the same softwareapplication for both GPS tracker device and smartphone al-lowed for consistency in the present study in the presenta-tion of data resulting in better reliability (17) in reporting onthe different speed zones when comparing GPS tracker deviceto smartphone. Previous studies have highlighted significantdifferences across different methods and systems provided totrack movement specifically in football (18).

Practical Applications� Having this ability to easily monitor physical activity en-

ables more people to engage in physical activity related tofootball (19).

� Increased motivation from engagement of fans in smart-phone applications has been shown to have health benefits(20).

� The capability to more accurately measure and the capacityto report on more physical activity in diverse populationscan help health organisations and governing bodies to bet-ter understand participation (21).

� A simple and affordable method (being free) to quantifyphysical activity and help coaches and practitioners to bebetter informed.

DatasetDataset available on SportPerfSci.com

References1. Hennessy, L., & Jeffreys, I. (2018). The Cur-rent Use of GPS, Its Potential, and Limitations in Soc-cer. Strength and Conditioning Journal, 40(3), 83-94.doi:10.1519/ssc.00000000000003862. Buchheit, M., & Simpson, B. M. (2017). Player-TrackingTechnology: Half-Full or Half-Empty Glass? InternationalJournal of Sports Physiology and Performance, 12(Suppl 2),S2-35-S2-41. doi:10.1123/ijspp.2016-0499

sportperfsci.com 3 SPSR - 2018 | Nov | 39 | v1

Comparison of a smartphone App with other GPS devices

3. Krustrup P, Krustrup BR. (2018). Football is medicine - itis time for patients to play! Br J Sports Med. Epub ahead ofprint: Retrieved from https://doi.org/10.1136/bjsports-2018-0993774. Piwek L, Ellis DA, Andrews S, Joinson A (2016) The Riseof Consumer Health Wearables: Promises and Barriers. PLoSMed 13(2): e1001953.5. Mitchell, E.; Monaghan, D.; O’Connor, N.E. (2013) .Clas-sification of Sporting ActivitiesUsingSmartphoneAccelerome-ters. Sensors 2013, 13, 5317-5337.6. Benson, A. C., Bruce, L., & Gordon, B. A. (2015). Re-liability and validity of a GPS-enabled iPhoneTM“app” tomeasure physical activity. Journal of Sports Sciences, 33(14),1421-1428. doi:10.1080/02640414.2014.9946597. Marathon Performance (2014). Technical report. . Re-trieved from ¡www.marathoncenter.it/.../16235d38494ef26b1d35985ef92ce44d636¿.8. Jennings, D., Cormack, S., Coutts, A. J., Boyd, L., &Aughey, R. J. (2010). The Validity and Reliability of GPSUnits for Measuring Distance in Team Sport Specific RunningPatterns. International Journal of Sports Physiology and Per-formance, 5(3), 328-341. doi:10.1123/ijspp.5.3.3289. Adamakis, M. (2017). Comparing the Validity of a GPSMonitor and a Smartphone Application to Measure PhysicalActivity. Journal of Mobile Technology in Medicine, 6(2), 28-38. doi:10.7309/jmtm.6.2.410. Zou, H., Chen, Z., Jiang, H., Xie, L., & Spanos, C. (2017).Accurate indoor localization and tracking using mobile phoneinertial sensors, WiFi and iBeacon. 2017 IEEE InternationalSymposium on Inertial Sensors and Systems (INERTIAL).doi:10.1109/isiss.2017.793565011. Skourletopoulos, G., Mavromoustakis, C. X., Mastorakis,G., Batalla, J. M., Dobre, C., Panagiotakis, S., & Pallis, E.(2016). Towards Mobile Cloud Computing in 5G Mobile Net-works: Applications, Big Data Services and Future Opportu-nities. Studies in Big Data, 43-62.12. The Football League (2015) 11 Million Foot-ballers In England Cannot Be Wrong! Retrieved from¡www.thefa.com/news/2015/jun/10/11-million-playing-football-in-england13. Hennessy, L., & Jeffreys, I. (2018). The Cur-rent Use of GPS, Its Potential, and Limitations in Soc-cer. Strength and Conditioning Journal, 40(3), 83-94.doi:10.1519/ssc.0000000000000386

14. The football League (2017) Five-A-Side and Futsal Re-trieved From: http://www.thefa.com/get-involved/player/5-a-side-and-futsal15. Owen, A. L., Wong, D. P., Paul, D., & Dellal, A.(2014). Physical and technical comparisons between various-sided games within professional football. International Jour-nal of Sports Medicine, 35(4), 286–292.16. Oja, P., Titze, S., Kokko, S., Kujala, U. M., Heinonen, A.,Kelly, P., Foster, C. (2015). Health benefits of different sportdisciplines for adults: systematic review of observational andintervention studies with meta-analysis. British Journal ofSports Medicine, 49(7), 434-440. doi:10.1136/bjsports-2014-09388517. Portas, M. D., Harley, J. A., Barnes, C. A., & Rush, C.J. (2010). The validity and reliability of 1-Hz and 5-Hz globalpositioning systems for linear,multidirectional, and football-specific activities. International Journal of Sports Physiologyand Performance, 5(4), 448–458.18. Hoppe MW, Baumgart C, Polglaze T, FreiwaldJ (2018) Validity and reliability of GPS and LPS formeasuring distances covered and sprint mechanical prop-erties in team sports. PLoS ONE 13(2): e0192708.https://doi.org/10.1371/journal.pone.019270819. Tao, Z., Yang, X., Lai, I. K., & Chau, K. (2018). A Re-search on the Effect of Smartphone Use, Student Engagementand Self-Directed Learning on Individual Impact: China Em-pirical Study. 2018 International Symposium on EducationalTechnology (ISET). doi:10.1109/iset.2018.00056

20. Kuska, M., & Zukowska, H. (2017). Smartphone ap-plications as a source of motivation for engaging in physicalactivity. doi:10.1063/1.501245521. Hulteen, R., Smith, J., Morgan, P., Barnett, L., Hallal,P., Colyvas, K., & Lubans, D. (2017). Global participation insport and leisure-time physical activities: A systematic reviewand meta-analysis. Journal of Science and Medicine in Sport,20, e38.

Copyright: The articles published on Science Performance and Science

Reports are distributed under the terms of the Creative Commons Attribu-

tion 4.0 International License (http://creativecommons.org/licenses/by/4.0/),

which permits unrestricted use, distribution, and reproduction in any medium,

provided you give appropriate credit to the original author(s) and the

source, provide a link to the Creative Commons license, and indicate if

changes were made. The Creative Commons Public Domain Dedication

waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the

data made available in this article, unless otherwise stated.

sportperfsci.com 4 SPSR - 2018 | Nov | 39 | v1