Background and experience
I did not follow the traditional path that leads to where I am today. It's not widely known, but I started my career as a physiotherapist and got my first taste of the football world in that capacity. My first job was as a physio at OH Leuven, a professional club in Belgium. As I became more interested in data analysis, I started creating dashboards for the club using data to make decisions.
This experience, coupled with my desire to learn more about data, led me to pursue an introductory course in Computer Sciences at the University of Harvard while working at the club. There were many early mornings and late evenings but it was definitely worth it. During my free time, I assisted several tech startups, and that's when I realized that my future was in the tech startup industry.
Why did you start MyGamePlan?
MyGamePlan was born out of my observation that staff members were struggling each week with manual tasks needed for performance analysis; like tagging, coding, sending clips to players, and recording, … they live in a constant rush from game to game together with the whole coaching staff.
MyGamePlan was born out of my observation that staff members were struggling with manual performance tasks used for performance analysis. I researched other teams in Belgium and around the world and discovered that they were all struggling with the same issues. This realization led me to team up with Dries and Milan, we found a starting budget and MyGamePlan was born.
Could you explain the mission and vision of MyGamePlan?
The ultimate goal is to automate the main part of football analysis, make it smooth, and ensure
that all team members, whether it’s the coaches, assistants, and players, can easily access all the data and video needed to make decisions without any difficulty.
Imagine the game is played and you immediately have a scorecard with all your principles as a coach, all your videos in the right playlists as an analyst, and all clips about what was good and what could improve for the next game as a player. Or even have that live during the game.
It will make life easier for the staff and it will give ownership to the players in a way they can digest the information, which ultimately leads to something we all want: better football.
How is MGP fixing this problem now?
With our MyGamePlan app, teams can now automate this process already. Teams can set up the
video clips they want to see for their next opponent or send individual clips to their players just once in the app, and it will repeat itself automatically each week.
Based on their feedback combined with the ever-growing possibilities of tracking data, we are able to track more complex principles each week.
Think about it, if you have to send 3–5 clips about the strengths and weaknesses of the next opponent on their position to each of your 15 players that will play next week, how long will it take you? 30–40 hours? With our application, it’s done minutes after the game.
Players receive a notification after the game, where they can analyze all their clips and data related to the game and also see how they scored on the tasks of the coach.
Any important milestones to share?
Since our launch after the World Cup, we reached a couple of milestones. In three months’ time we started working together with 38 professional teams and over 200 players use our PlayerApp each week to take a look at their last game video and data or analyze their next opponent.
Now it’s a really busy period talking to clubs and leagues all over the world who heard about our app and want to discuss a partnership for next season.
What are the football analytics trends and how is MGP embedding this in its vision?
We believe that automation is crucial, because it enhances individual player development, creates more time for a detailed analysis, and therefore you see the results in performance.
To make sure teams and players embrace
automation, we have to make sure that they can create any possible metric they want to track. Therefore, you need to be able to create an easy-to-use tool that can combine event and tracking data to link it with video. Because of the amount of input we get from users, it will grow stronger every day.
The next big step is to use ML models to value actions in a way that it is relevant and understandable for the coaches and players. If you hand them insights based on the most powerful ML model, but it’s not tailored to their strategy, formation, or style of play, it’s not really useful.
I always make a comparison to an xG model. You can have the most powerful model, but if it doesn’t take the individual player performance into account it’s almost completely useless. A shot inside the penalty area of Kevin De Bruyne should not have the same xG as a shot of Brandon Mechele (all respect to Brandon of course).
The next challenge that a lot of companies are working on is to increase the size of the data set we have available. If we are able to track the same insights out of training footage and academy football as we do now with professional football games, then we are really going to see the full potential of data analytics in football.
Liked this Blogpost? You can learn more on www.mygameplan.ai or send us a message.
