Digitalization in the Coaching Process Across Nordic-Baltic Countries (NPHZ-2022/10033)

Advances in technology have provided new possibilities for coaches to improve their coaching modalities. Overall, this digitalization process was and is strongly necessary during the Covid-19 pandemic period, in which coaches are challenged in changing their regular coaching ways towards new modalities including technologies. Home-based activities supervised by coaching staff through the use of web-based platforms or video-based lessons have been an important factor since the beginning of the Covid-19 pandemic to maintain the physical activity level and continuing monitor, and increasing sport performance.

Moreover, the use of technologically advanced coaching methodologies could increase the coach-athlete and between-athlete interactions trying to keep the social aspect of coaching. Furthermore, newly created apps might play a crucial role in monitoring athlete sport activities and the coaching process, and in turn potentially increase athlete and coaches’ motivation. Their motivation can also be enhanced through the use of new form of technologies such as virtual reality, mixed reality, augmented reality, exergaming, which can offer new form of interactive training that can be included into the traditional coaching process. While strengthening the Nordic-Baltic Sports Coaching network we look for new partners to be involved and enhance the knowledge of young and more experienced coaches. In last  few periods of the projects, the Network focused on the development of coaching skills and competences by creating new study modules, and practical courses embracing several aspects of the coaching process including psychological, social, pedagogical and methodological areas. Overall, the digitalization process can be considered as a facilitator to increase all the above-mentioned areas in the coaching process.

However, the use of digitalization is quite challenging for coaches for many potential reasons: 1) education attained long time ago; 2) differences in the needs and skills of modern-day athletes and kids compared to previous generations; 3) between-coaches difference in technological competences; 4) the ability to manage the huge amount of information received by new technological tools. These are important topics to be considered in sports coaching, which can help coaches to update their knowledge and advance in their everyday work. Therefore, the aim of this project is to have a full understanding of the coaching knowledge and expectation about digitalization for the coaching process, and to offer scientifically-based knowledge about digitalization opportunities in the coaching process through the creation of new shared e-learning modules and theoretical, as well as practical seminars for Nordic Baltic countries.

The project duration is three years. Activity period 09/2022 – 06/2025

Total budget is 186 050 EUR

The amount funded by Nordplus: 92 940 EUR

Coordinator:

Lithuanian Sports University

Project Partners:

LV-Latvian Academy of Sport Education (LV)
University of South-Eastern Norway (NO)
University of Southern Denmark (DK)
Sport School of Kaunas “Gaja” (LT)
VSI Smart Health DIH (LT)
Valmiera Olympic Center (LV)
Be1 National Football Academy (LT)
Reykjavik University (IS)
International Boxing Association Coaches Committee (FI)
National Association of Conditioning  Training NARTA (LT)

Digitalization in sport coaching – introduction
Slides for the introduction presentation

Barriers in the use of digital tools – possibility to overcome them (Thomas Bredahl, Nicklas Stott Venzel)
Video

Artificial intelligence in the coaching process – how to help the coaching process (Peter O’Donoghue)
Preparation task
This document outlines what you need to do before we all meet in Lithuania.  So I am simply asking you to watch a match in a sport of your choice and make some notes about the tactics you can see being applied.  This is important because artificial intelligence will be compared with human intelligence during the session.
Spatio-temporal data
This video (13:50) describes the data that were used in the machine learning exercises.  They are player tracking XY data from 41 professional soccer matches.  A computer program identified basic events (passes, dribbles, etc).  So I have a question that we can discuss.  What are the relative advantages and disadvantages of the following approaches:
(a) Tagging basic events ourselves in a video tagging package like HUDL or Nacsport.
(b) Using events from a readily available source like Opta.
(c) Writing a program to automatically identify these events from XY tracking data (this is the approach we chose).
Spatio-temporal data (slides)
These are the slides used in the above video presentation.
Automatic identification of overlap runs
This video presentation (20:38) is an alternative to machine learning / AI.  It is an example of algorithmic programming to automatically identify the overlap run tactic in soccer.  It integrates match event data (passes, dribbles, etc) and XY data together.  A critical step in developing such a program is to precisely define what we mean by an overlap run.  There are some questions we will discuss about this when we meet in Lithuania.
(a) Should any criteria be added to avoid some of the false positive cases?  A false positive is where the system said an overlap run occurred but a coach would disagree,
(b) Should any criteria be removed or changed?  This is to allow more overlap runs to be recognised and to avoid false negative cases.  A false negative is where there is an overlap according to an expert coach but the system failed to identify it.
(c) What are the benefits of using artificial intelligence / machine learning instead of using this algorithmic programming approach?  This is what the rest of the session will be about.
Automatic identification of overlap runs (slides)
These are the slides used in the above video presentation.
Overlap runs performed by Millwall
This video (2:57) shows four overlap runs performed by Millwall against Cardiff City.  In each case we firstly see the player movement traces during the overlap run (cyan represents Millwall, yellow represented Cardiff City, green represents keepers, red represents the ball).  Then we see the video clip twice, once without telestration and once with telestration.  Consider these examples.  Would a coach find these useful?  Would a coach agree they are overlap runs?
Animation of an overlap run
This video (3:39) is an animation of the fourth overlap run performed by Millwall against Cardiff City.  This was done in Matlab.  Light blue (cyan) is used to represent the Millwall players who play in blue.  Yellow was used to represent the Cardiff City players who played in orange.
Follow-up analysis: Crossfield runs
This is a video (4:26) where I talk about the system to automatically identify crossfield balls.  We discussed this in Lithuania and you gave me some criteria for recognising this tactic which have now been implemented.
Video

Using unexpensive digital tools to simplify the coaching process – reliable and valid tools based on the literature (Marco Pernigoni)
Pernigoni – Tasks (Validity/reliability and video analysis)
Video
Lecture – Validity and reliability of digital tools
Lecture – Introduction to video analysis

Video Analysis (Marco Pernigoni)
Video

Monitoring in Sports (Dominykas Bartusevičius)
PRE-PROJECT PREP. MONITORING
Lecture – Monitoring in Sports
Video

Testing in Sports (Francesco Coletta)
Video
Monitoring tools for performance