Soccer is one of the fastest-growing sports in Canada, but it lacks funding at grassroots level. Vitruvian wanted to level the playing field by helping amateur players train like the professionals.
Vitruvian worked with IBM on key components of its Zone360 platform, including a VR app that helps players learn to react to in-game scenarios. Machine learning models built with IBM Watson Studio help to simulate realistic match situations.
Democratizessoccer by bringing high-tech training tools within every player’s reach
Empowersplayers and coaches to develop their skills outside of team training sessions
Createsa more transparent, efficient ecosystem for scouting talent
Business challenge story
Kicking off a revolution in grassroot coaching
When Matteo Masucci moved to Toronto five years ago, his first experience of local soccer was a bit of a culture shock. As a former professional in Italy, he had been brought up through a well-funded youth system that is designed to feed some of the world’s biggest professional soccer teams with the brightest young players. In Canada, where soccer is a much less established sport, the situation at grassroots level is somewhat different.
“Soccer is one of the fastest-growing sports in Canada—there are more kids playing soccer now than playing hockey,” says Masucci. “But since the soccer tradition here is very new, there is a shortage of senior players and coaches with top-level experience. Most clubs have only one professional coach, who is responsible for the development of 2,000 to 5,000 players and dozens of volunteer coaches.
“With such tight resources, it’s hard for young players to get the guidance they need to improve. There is a knock-on effect on the quality at senior level too: since there are so many players and such limited opportunities for scouting, it’s hard to find and develop the most promising talents.
“I wanted to find a way to translate all the passion and enthusiasm for soccer into better performance at the player, team, and national level—and that’s where Vitruvian Sport Systems was born.”
Masucci teamed up with Glen Rapoport, an experienced entrepreneur and business manager, to found Vitruvian, a company with a mission to bring top-class, high-tech training tools to players at all levels of Canadian soccer.
Rapoport explains: “The big professional clubs can afford cameras, biometric sensors, and all sorts of expensive training equipment—but that’s far beyond the reach of a youth team that may not even have a full-time coach. We realized that by taking advantage of a piece of equipment that most players already own—a standard smartphone—we could give them access to state-of-the-art training tools.”
NOTE: IBM Data Science Experience is now known as IBM Watson Studio
Using your head
The Vitruvian team began developing the Zone360 Player Development and Profiling platform, with a roadmap of building a wide range of tools to help players and coaches measure, track and improve their technical and tactical skills.
One of the most important tools in the first generation of the build is an immersive training simulation that uses an inexpensive headset (such as Google Cardboard) to turn a player’s smartphone into a virtual reality device. It allows coaches to quickly define a realistic match situation, and put the player in a pivotal position on the virtual pitch. The player must use their wits and senses to quickly assess the situation and make the right decision about where to move and what to do. As they make these decisions, the simulation advances and provides visual feedback on the outcome of the play.
Masucci explains: “To make the simulation convincing, we need the simulated players to position themselves in a realistic way, depending on the match situation. It would be too time-consuming to ask the coach to program every player’s movements individually for each play, so we wanted a smarter solution. That’s why we started working with IBM.”
Vitruvian worked with a team from the IBM Machine Learning Hub in Toronto to build a machine learning model, using Jupyter Notebooks and other tools in IBM Watson Studio. The Machine Learning Hub provided expert guidance via a series of workshops to help skill-up the Vitruvian team, and create an initial prototype. Subsequently, the two teams have continued to work together to refine the model.
Steven Astorino, VP of Development, Hybrid Cloud and z Analytics at IBM, adds: “The Machine Learning Hub is a great place for companies to leverage IBM's expertise and talent on machine learning and data science projects. Whether you are a start-up or one of the largest enterprises in the world, the Hub can help you learn about our products such as Watson Studio and understand how you can best leverage machine learning and data science within your organization. We can help you leverage and collaborate with our expert data scientists, take a deep dive into your existing models, create new models, or brainstorm and understand industry-based use cases that can give you a competitive edge. Vitruvian Sport Systems is a great success story showcasing what the Machine Learning Hub can do for you."
IBM Watson Studio facilitates the machine learning process for Vitruvian by acting as both a model development tool and a collaborative workspace. Within the cloud-based Data Science Experience interface, IBM and Vitruvian can easily create, edit and share notebooks, data sets and other artifacts, as well as monitoring progress and results.
“The IBM Machine Learning Hub has given us a great head-start, and we are excited about iterating on the model development process with Watson Studio,” says Rapoport. “As our in-house team continues to build up its machine learning skills, we are also planning to use Watson Studio for model deployment and model management.”
The current version of the model is able to analyze player positioning data from historical soccer matches, and automatically generate realistic positions and movement patterns that can be plugged into the Zone360 simulation.
Rapoport adds: “The machine learning model means we can do 90 percent of the setup automatically, so the coach only needs to tweak a few elements to create the kind of training situation they want. With IBM cognitive technology, instead of only having time to create a small number of generic exercises for their whole team, coaches can quickly build realistic soccer scenarios that help each individual player to practice the skills they need.”
The Vitruvian team also sees opportunities to leverage machine learning and IBM Watson™ cognitive solutions in future tools that it plans to integrate into the Zone360 platform.
Rapoport says: “There are all kinds of ways that IBM Watson could help us. Ultimately, we will use machine learning to recognize the potential of players at a younger age, and help clubs do smarter scouting—we’d effectively be creating a professional social network of soccer players, with objective data about their skills and abilities.”
Masucci comments: “IBM’s expertise in machine learning and data science is wonderful. IBM Watson Studio and IBM Watson cognitive services give us the machine learning and visualization tools we need to help players and coaches excel.”
Hitting the back of the net
Masucci is already using the app with a group of players at his own club in Toronto, and the feedback has been excellent.
“What’s interesting is that the youngest kids are the quickest to adapt to the virtual reality environment, but even the older players have said that it’s a useful addition to their training regimen,” he says. “The best aspect is that it lets players practice individual technical skills in their own time, while saving team training sessions for team-focused activities. It makes everyone much more productive.”
Rapoport adds: “One of the typical problems when coaching resources are scarce is that players don’t get feedback, so they can’t develop their game. You can have a kid who has a lot of talent at age 12, but if they don’t work on their left-foot technique, they might be behind the curve at age 17.
“With Zone360, we can capture objective data, build up profiles of each player’s strengths and weaknesses, and give them much better insight into what they need to do to improve. By improving the individuals, you improve the team—and by improving the teams, you raise the general standard of play at professional and international level.”
Masucci concludes: “Zone360 won’t just be good for soccer in Canada—it will help the global game. As Technical Director of the Juventus Academy Toronto, I know that professional clubs around the world are looking for ways to tap into undiscovered talent and streamline their scouting processes. With Zone360 as a cognitive-powered social network for grassroots soccer, we’re giving them new insight into the brightest stars of the next generation.”
About Vitruvian Sport Systems
Vitruvian Sport Systems is an innovative startup based in Toronto, Canada, which focuses on helping soccer players and coaches improve through the use of mobile and artificial intelligence technologies. The company’s Zone360 solution provides a host of tools to help fine-tune technical and tactical skills, without any need to invest in expensive cameras or other hardware: the whole experience is delivered via a standard smartphone.
Take the next step
IBM Analytics offers one of the world's deepest and broadest analytics platform, domain and industry solutions that deliver new value to businesses, governments and individuals. For more information about how IBM Analytics helps to transform industries and professions with data, visit ibm.com/analytics. Follow us on Twitter at @IBMAnalytics, on our blog at ibmbigdatahub.com and join the conversation #IBMAnalytics.
To learn more about IBM Data Science Experience, please visit ibm.com/analytics/data-science-experience
To learn more about the IBM Machine Learning Hub, please visit ibm-ml-hub.com