Real-Time Tactical Analysis: Leveraging GNSS Position Data for

Tactical Behavior

Paolo Roberto Gabrielli

1,2

and Stefano D'Ottavio

1,2,3

School of Sports and Exercise Sciences, Faculty of Medicine and Surgery, Tor Vergata University, Rome, Italy

Tor Vergata University of Rome, Department of Clinical Sciences and Translational Medicine, Rome, Italy

Head of Performance Area, AS Roma Women Football, Italy

Keywords: User Acceptance, Ubiquitous Systems, Personal Computers, Personal Data Tracking, Wearable Computers,

Sports/Exercise.

Abstract: In team sports such as football, tactical analysis has become essential to evaluate the collective behavior of

the team and provide immediate and reliable feedback to the coaching staff. Global navigation satellite system

(GNSS) tracking systems are widely used in team sports such as football. From geographic coordinates (e.g.,

latitude and longitude), information about the positions of players during a match can be obtained. Most

GNSS-based systems use location tracking information to evaluate physical performance in terms of external

load and physical stress. This research aims to present a technological innovation called GLF (Geo Live

Football), which integrates tactical measures from raw GNSS data to understand the behavior of the team

during matches, introducing the concept of "live" analysis.

INTRODUCTION

In team sports, performance analysis has become an

essential tool for coaches, athletes, sports

organizations, and researchers. Collecting and

interpreting performance data allows coaches to

improve training programs, athletes to make better

tactical decisions, sports organizations to manage

teams more effectively, and researchers to develop a

better understanding of sports performance. Analysis

can include technical, metabolic, and tactical aspects,

providing a comprehensive view of player

performance (O'Donoghue 2015). In invasive sports

such as soccer, key actions are observed, recorded,

and analyzed through subjective observation,

resulting in the loss of relevant information (Franks

1991). The application of technology in sports has

enabled automatic monitoring of player performance

during play. Automated physical performance

monitoring helps to define the movement

characteristics of elite soccer players during matches

(Lago-Peñas 2012), quantify training load (Buchheit

2013), emphasize athletes' movement during match

activities, and provide a systematic approach to

working with elite players (Strudwick 2013). Over

time, research has highlighted an increasing

correlation between variables associated with

physical performance and tactical behaviors in

soccer. It has been studied how strategic decisions,

collaborative actions between players, positioning in

relation to the game situation, off-ball movements,

passing choices, and interaction with other players

influence overall performance on the pitch (Low,

2020). Tactical analysis in elite soccer has historically

relied on observational data using variables that

discount most of the contextual information. Team

tactical analysis requires detailed data from various

sources, including technical ability, individual

physiological performance, and team formations

(Rein 2016). Recent performance analysis is

increasingly connected to Big Data (Memmert 2018),

defining new approaches to calculate metrics that

help measure and especially identify team and

individual performance of players, the way teams

interact, thus reducing the gap between research, data

collection, and their interpretation and use (Carling

2005). This evolution allows a more precise vision of

reality, providing information on the strengths and

weaknesses of one's team and opponents in various

phases of the game (Castellano 2012).

Gabrielli, P. and D’Ottavio, S.

Real-Time Tactical Analysis: Leveraging GNSS Position Data for Tactical Behavior.

DOI: 10.5220/0012967600003828

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 12th International Conference on Sport Sciences Research and Technology Support (icSPORTS 2024), pages 193-198

ISBN: 978-989-758-719-1; ISSN: 2184-3201

193

USE OF POSITIONAL DATA IN

FOOTBALL

Football teams are increasingly using

microtechnology, incorporating GPS to monitor

physical conditions by quantifying the external load

(EL) in terms of total distance traveled, various speed

thresholds, accelerations/decelerations and composite

variables (Silva 2018; Rago 2020; Brink and

Frencken 2018). This application often leaves out

significant information, especially in the

investigation and evolution of team tactical behavior

and the individual and collective actions that a team

organizes to score a goal and win the match. A study

(Sampaio and Macas 2012) investigated how the

dynamics of player positional data from GNSS

systems could be used to assess tactical behavior by

measuring movement patterns and coordination

between players, opening new research horizons and

reducing the gap between sports science and sports

training. The research addressed in this article

proposes the development of a technological system

“Geo Live Football” (GLF), which integrates and

develops tactical metrics (Figure 1), and introduces

the concept of “live”, offering a complete and real-

time vision of the physical and tactical behavior of the

team during matches.

Figure 1: (A, B, and C), include team CG, team aspect ratio,

team heatmap, and team effective playing area.

GNSS - SYSTEM

REQUIREMENTS

GNSS-based performance monitoring systems have

been widely used in recent years to analyze the

performance of sports teams, such as in football, by

providing detailed information during training and

matches-play. However, football clubs are constantly

looking for improvements. Analyzing live tactical

behavior during matches, using tactical metrics,

presents a new challenge. Real-time information

about players' physical condition and overall team

performance statistics, obtained by geolocalization

players on the pitch, are essential for making

decisions and strategic adjustments during matches.

GNSS is a key technology for tracking and

monitoring athlete performance. Compared to its

main alternative, Ultra-Wide-Band technology, it

does not require ground infrastructure near the

football pitch. Most football stadiums are not fully

covered, and those that are use materials such as

polytetrafluoroethylene (PTFE) or glass fiber

composites, or ETFE (ethylene tetrafluoroethylene),

which minimize GNSS signal attenuation. To provide

real added value to live football applications, GNSS

systems must ensure:

• High Accuracy: At least 50cm RMS for tactical

purposes, achievable using real-time differential

corrections and dual-frequency receivers.

• Miniaturization: Devices should be small enough

to integrate into undergarments, impacting

battery and antenna size.

• Integration with Accelerometers: To better

calculate key performance parameters like

metabolic power.

• Low Power Consumption: Necessary due to

dual-frequency receivers and real-time

transmission power requirements. At least 4

hours of autonomy is required.

• Real-Time Data Transmission: Low-power and

customized transmission protocols to reduce

power consumption and avoid packet collisions.

The typical RF interference in football stadiums

should be considered. The transmission medium

will be a 4G/5G modem encapsulating RTCM

V3.1 or V3.2/MSM (Multiple Signal Messages)

standard for differential correction.

• Real-Time Data Processing: To derive specific

metrics.

• An app for configuration and metric viewing.

Figure 2: Operational architecture.

icSPORTS 2024 - 12th International Conference on Sport Sciences Research and Technology Support

194

GEO LIVE SOCCER SYSTEM

The proposed application, GLF (GeoLiveFootball),

represents an innovative tool for "live" real-time

monitoring of football performances. Currently, athlete

performance monitoring does not use players' positions

for tactical analysis, which is primarily conducted

through video analysis at additional costs. GLF enables

both performance and tactical analysis within the same

solution. GLF, a downstream application of PNT User

Segment (GNSS) technology, allows "live" metrics

calculation, usable by coaching staff for match tactics

analysis with accurate positioning, player geolocation,

and real-time data collection and processing (Figure 3).

Accurate PVT of each wearable GNSS device will be

calculated using differential corrections from a GNSS

reference station; PVT will then be transferred to a

cloud server. The software can select differential

corrections from available commercial services (e.g.,

HxGN SmartNet, Trimble RTX) or a reference station

provided by GLF. The GLF metrics calculation

software on the cloud server will provide a general

overview of team performance in geometric terms and

individual players' physical and technical performance

(distance indicators and kinematics such as speed,

intensity, and acceleration) at predefined intervals

(usually 5 minutes). Metrics will be available through

a web-based interface for controlled user access and

system setup for each football club (Figure 3). The

GLF metrics will aid coaching staff in decision-making

during matches, and all metrics are recorded for post-

match analysis. GLF evolved from the MESSI-HP

solution (Monitoring Evolution with Soccer Satellite

Navigator Innovation - High Precision) developed to

monitor players during training and competitions.

GLF's innovative features include:

• Live: Real-time GNSS data transmission and

processing, providing real-time metrics during

events, a unique feature in match analysis

solutions.

• High Accuracy: Target placement accuracy of

about 50cm error, comparable to UWB accuracy,

enabling game tactics evaluation.

• Georeferencing: GNSS devices are used for

kinematic purposes (velocity, acceleration,

metabolic power). GLF georeferences player

positions relative to the football field, providing

insight into player positions during matches and

being competitive with UWB and video

analytics, which are not "live".

• Miniaturization: Devices are extremely

miniaturized, fitting in a bib pocket. Key design

features include:

• Multi-Constellation and dual-frequency GNSS

receiver (e.g., ublox F9P)

• Transmission of GNSS signal phase difference

corrections (RTCM V3.x) RTK and DGNSS

algorithms for high accuracy positioning (RMS

<50cm)

• Hybridization of GNSS and IMU data

• 4G/5G data transmission

Figure 3: Tactical metrics visible through a web-based

interface.

4.1 Georeferencing: Four Corners of

the Field

Figure 4: Pitch location: four pitch corners.

Georeferencing is crucial for accurate player

positioning during a game. This is done by visually

determining the four corners of the field and manually

obtaining the latitude and longitude coordinates by

clicking on those corners.

4.2 Live Match

The match session allows detection of both physical

and technical-tactical metrics during training or

official matches and viewing only some metrics.

Technical-tactical metrics are aimed at the team,

while physical metrics target individual players,

groups, or the entire team. During live matches, data

is displayed at 5-minute intervals. The metrics chosen

to be displayed during the live session include:

Real-Time Tactical Analysis: Leveraging GNSS Position Data for Tactical Behavior

195

4.2.1 Real-Time Tactical Metrics

Figure 5: Players who occupy the offensive and defensive

phase, Center of gravity of the team, Heat map, Team length

and width, Occupied playing area.

Tactical metrics (Figure 5) offer an overview of how

the team moves on the pitch and are particularly

useful for analyzing the formation and collective

behavior during the most important phases of the

game, such as attack and defense. This information

allows you to refine strategies in real time, favoring

greater tactical effectiveness.

• Players involved in the offensive and defensive

phases: this metric highlights the number of

players who participate in the attack and defense

phases. It is an important indicator for evaluating

the balance of the team and identifying any

tactical imbalances.

• Team Center of Gravity: represents the

geometric point that corresponds to the average

value of the spatial distribution of players on the

field, providing information on the overall

structure of the team. This metric allows you to

evaluate whether the team tends to lean too much

towards attack or to remain excessively compact

in defense.

• Heat Map: graphically represents the areas of

the field most occupied by players. This

visualization is useful for identifying which areas

are used most frequently and which areas are

under-used, providing fundamental information

to optimize field coverage and implement

strategies.

• Team Length and Width: these metrics

measure how much the team extends on the field,

both in terms of depth (length) and width (width).

Length concerns the distance between the most

advanced players and those furthest back, while

width describes the use of lateral spaces. This

information helps to evaluate whether the team is

too compact or too dispersed.

• Occupied Playing Area: defines the space

covered by the team during the various phases of

the game. It is used to determine the team's

ability to effectively cover the pitch and limit the

spaces available to opponents.

4.2.2 Real-Time Physical Metrics

Figure 6: Total distance traveled, distance traveled at high

intensity, distance covered in sprints.

Physical metrics (Figure 6), on the other hand, focus

on individual player performance, monitoring

parameters such as endurance, speed and intensity.

This data is essential to understand the physical state

of players during the match and to make informed

decisions regarding substitutions or changes of pace.

• Total distance traveled: measures the total

number of kilometers covered by an individual

player or the entire team during the match. It is

an indicator of physical endurance and can be

useful for assessing levels of fatigue.

• High-intensity distance traveled: calculates the

distance covered by players at high speed. This

metric is crucial for analyzing physical effort

during the most intense moments of the game,

such as during pressing or counterattacks.

• Sprint distance traveled: shows the distance

covered at maximum speed. This metric is

important for evaluating the acceleration

capacity of players and their contribution during

decisive moments, such as rapid offensive or

defensive actions.

USING METRICS DURING THE

GAME

Analyzing physical and tactical metrics at regular

intervals, such as every 5 minutes, provides a real-

time overview of team and individual player

performance. Tactical metrics allow you to assess the

team’s position on the pitch and make immediate

icSPORTS 2024 - 12th International Conference on Sport Sciences Research and Technology Support

196

strategic adjustments. In contrast, while physical

metrics allow you to monitor fatigue levels and

optimize substitutions or changes of pace. Together,

this data is a powerful tool for making quick,

informed decisions during the game, improving your

chances of success.

DISCUSSION

In the discussion section, it is essential to compare

and contextualize the GLF system with other similar

technologies to highlight its distinctive features and

the specific contributions it offers to the field of sports

analytics. Here is how we can integrate the Catapult

Sports and STATSports systems into the comparison.

The GLF system represents significant advances in

the field of sports performance analytics, especially

when compared to existing technologies. For

example, Catapult Sports and STATSports are widely

used systems that use Global Navigation Satellite

System (GNSS) technology to track player

movements and measure performance in real time.

These systems rely on GPS data to evaluate physical

metrics such as speed, distance and acceleration,

while also offering tools for tactical analysis.

However, the GLF system stands out for its ability to

provide “live” data in real time, allowing teams to

make tactical adjustments during matches, a feature

that represents a significant improvement over the

post-match analysis offered by traditional systems.

While systems like Catapult and STATSports are

more focused on monitoring physical performance,

the GLF system integrates both physical and tactical

metrics with superior accuracy. With a margin of

error of approximately 50 cm, the GLF provides a

higher level of accuracy than many GNSS-based

platforms currently available. This superior accuracy

translates into more detailed and actionable

information to improve tactical performance.

Additionally, the GLF’s real-time geotagging

capabilities rival other solutions, such as Ultra-

Wideband (UWB) and video analytics, but with the

added benefit of providing immediate feedback,

essential for making in-game adjustments. Another

area where the GLF excels is the miniaturization of

GNSS devices, designed to fit discreetly into a chest

pocket without compromising player performance.

Featuring multi-constellation, dual-frequency

receivers, these devices represent a more

technologically advanced solution than those

typically employed by systems like Catapult and

STATSports, which tend to use bulkier, less

sophisticated hardware. By integrating these

features—high accuracy, real-time data processing,

and advanced miniaturization—the GLF system

addresses some of the most critical challenges in

sports analytics, such as data accuracy, stadium

coverage, and efficient data management. Compared

to established systems like Catapult and STATSports,

the GLF offers a more comprehensive solution for

analyzing both physical and tactical performance,

marking a significant advancement in the way team

dynamics and in-game strategies are evaluated.

CONCLUSIONS

This research presented an innovative system

leveraging GNSS technology to capture and analyze

players' positional data during matches. By

integrating these data into tactical metrics, the system

offers a comprehensive evaluation of both physical

and tactical performance. The "live" analysis

capability of the GLF system represents a significant

advancement in sports analytics. For this system to be

reliable and accurate, it must adhere to specific

technical requirements. The GLF system meets these

standards, ensuring precise and real-time data

collection, which is crucial for detailed tactical

performance evaluation and future tactical planning.

REFERENCES

O'Donoghue, P. 2015. Introduction to Sports Performance

Analysis, Routledge.

Franks, I.M. & Miller, G. 1991. Training Coaches to

Observe and Remember. Sports Science Journal, 9,

285-297.

Lago-Peñas, C., Rey, E. & Lago-Ballesteros, J. 2012. The

influence of actual playing time on the physical

demands of elite soccer players. The Open Sports

Science Journal, 5.

Buchheit, M., Racinais, S., Bilsborough, J.C., Bourdon,

P.C., Voss, S.C., Hocking, J., Cordy, J., Mendez-

Villanueva, A. & Coutts, A.J. 2013. Monitoring fitness,

fatigue, and run performance during a pre-season

training camp in elite soccer players. J Sci Med Sports,

16, 550-5.

Sampaio, J.E., Lago, C., Goncalves, B., Macas, V.M. &

Leite, N. 2014. Effects of rhythm, state, and imbalance

in motion variables of tempo, heart rate, and tactical

behavior during soccer play to 5 small row games.

Journal of Science and Medicine in Sports, 17, 229-33

Strudwick, A. 2013. Contemporary issues in the physical

preparation of elite players. Science and Football.

Routledge.

Low, B., Coutinho, D., Goncalves, B., Rein, R., Memmert,

D. & Sampaio, J. 2020. A systematic review of

Real-Time Tactical Analysis: Leveraging GNSS Position Data for Tactical Behavior

197

collective tactical behaviors in soccer using positional

data. Sports Medicine, 50, 343-385.

Rein, R. & Memmert, D. 2016. Big data and tactical

analysis in elite football: future challenges and

opportunities for sports science. Springer Plus, 5, 1410.

Memmert, D. & Rein, R. 2018. Match Analysis, Big Data

and Tactics: Current Trends in Elite Football. Deutsche

Zeitschrift Fur Sportmedizin, 2018, 65-72.

Carling, C., Williams, A.M. & Reilly, T. 2005. Handbook

of Soccer Match Analysis a Systematic Approach to

Improving Performance, London; Routledge.

Castellano, J., Casamichana, D. & Lago, C. 2012. The use

of match statistics discriminating between successful

and unsuccessful soccer teams. Journal of Human

Kinetics, 31, 139-47.

Silva, P., Santos, E.D., Grishin, M. & Rocha, J.M. 2018.

Validity of heart rate-based indices for measuring

training load and intensity in elite soccer players.

Journal of Strength and Conditioning Research, 32,

2340-2347.

Clemente, F., Couceiro, M., Martins, F. & Mendes, R.

2013. An online tactical metric applied to the game of

soccer. Research Journal of Applied Sciences,

Engineering and Technology, 5, 1700-1719.

Shergill, A.S., Twist, C. & Highton, J. 2021. Importance of

GNSS data quality assessment with new control criteria

in professional soccer matches. International Journal of

Performance Analysis in Sports, 21, 820-830.

icSPORTS 2024 - 12th International Conference on Sport Sciences Research and Technology Support

198