The pursuit of wisdom in any walk of life quickly reveals that what you think you know is not nearly enough to get you to where you want to go. As I'm starting out in my football journey I challenged myself to reach out to those already working in various roles in football to answer a short list of questions. My goal wasn't to get answers but relevant perspectives on the game within the game.
Here is a conversation with Jan Van Haaren:
Here is a conversation with Jan Van Haaren:
Hi! I am Jan Van Haaren and I am a Belgian football data scientist who currently resides in The Netherlands. A trained computer scientist and machine learning engineer, I have been developing football analytics metrics that capture player performance and playing style for more than ten years. Moreover, I have been an avid football player since my early childhood.
Q: What is your first memory of football?
My early football memories date back to the middle of the 1990s. I remember collecting Panini stickers for the 1994/1995 season in the Belgian league. Around the same time, I also started playing football at my local football club, which I have continued to do ever since. The first football tournament that I remember consciously watching on television is the 1998 FIFA World Cup in France, where Belgium faced The Netherlands, Mexico and South Korea in the group stage. Having grown up in a border village between Belgium and The Netherlands, I particularly remember the atmosphere in our village for the tense derby against The Netherlands!
Q: What attracted you to data? What is more intriguing now: names or numbers?
I got intrigued by football data in the late 1990s and early 2000s while playing football games such as FIFA 99 and Championship Manager 2001/2002 with the latter game still being my all-time favorite football game. However, I quickly got more interested in modding those games than in actually playing them.
Q: What is your first memory of football?
My early football memories date back to the middle of the 1990s. I remember collecting Panini stickers for the 1994/1995 season in the Belgian league. Around the same time, I also started playing football at my local football club, which I have continued to do ever since. The first football tournament that I remember consciously watching on television is the 1998 FIFA World Cup in France, where Belgium faced The Netherlands, Mexico and South Korea in the group stage. Having grown up in a border village between Belgium and The Netherlands, I particularly remember the atmosphere in our village for the tense derby against The Netherlands!
Q: What attracted you to data? What is more intriguing now: names or numbers?
I got intrigued by football data in the late 1990s and early 2000s while playing football games such as FIFA 99 and Championship Manager 2001/2002 with the latter game still being my all-time favorite football game. However, I quickly got more interested in modding those games than in actually playing them.
I got involved in a Dutch website that reported on the latest news in the football games industry and hosted a community of modders for the FIFA series. While most community members were mostly interested in designing new kits, shorts and badges or even building entire stadiums from scratch, I was more interested in building modding tools and updating the in-game data.
By doing so, I even ended up as a Data Editor for the FIFA series! Back in the days, I used to be a fan of players like Ronaldo, Zinedine Zidane, Rivaldo, Dennis Bergkamp, Thierry Henry, Gabriel Batistuta, Andrei Shevchenko, Luis Figo, and Pavel Nedved. However, these days, I am definitely more intrigued by the numbers than by the names.
Q: What is more intriguing to you: cultivating a team vying for European places or building a team that progressively staves off relegation? Why?
I love football clubs that try to punch above their weight by outsmarting their opponents, regardless of their season objectives. Football analytics has so much to offer when it comes to taking better-informed decisions, but most clubs have only scratched the surface! I would not be surprised if a few forward-thinking clubs that currently still reside in the lower divisions but have already discovered the potential of data and analytics ended up at the top of European football in the next decade or so.
Q: What is the biggest misconception or cliché regarding data or recruitment that you have encountered in this space?
Many traditional football scouts appear to believe that football data scientists aim to replace them with computers and numbers. This misconception considerably slows down the adoption of data-driven methods in football unfortunately. Instead, I believe that football scouts should view football metrics as a tool and an additional source of information that enables them to take better-informed decisions. In addition, these tools would enable them to do their work faster, allowing them to analyze and evaluate more potentially interesting transfer targets.
Q: If you could start over, what skill would you build on first?
Unlike many other people who are trying to get into football analytics nowadays, I first built my skill set and then started applying my skills to football data. However, if I were to develop my skill set from scratch, I would focus first on programming skills and then on data science skills. From mentoring more junior football data scientists, I learned that a lack of programming skills often stands in the way of quickly trying out new ideas. Furthermore, I believe that solid programming skills speed up the process of acquiring relevant data science and machine learning skills.
Q: What is more important: domain knowledge or curiosity?
I would say that one needs a good mix of domain knowledge and curiosity as a football data scientist. Domain knowledge helps to understand the need of the practitioners and to assess the quality of potential solutions, whereas curiosity helps to come up with potential solutions that might address the need of the practitioners.
Q: What is more impactful for a player: psychology or environment? Why?
I would say that the answer very much depends on the player and varies from one player to another. The impact of psychological and environmental factors on the performances of players are not very well understood yet. These factors definitely deserve more attention in the football analytics community.
Q: What coach or player would you give a lifetime contract to? Why?
Kevin De Bruyne would be my pick as I really enjoy watching him play for Manchester City and the Belgian national team. He is not only a great football player, but also a great example of why the risk versus reward trade-off should be taken into account when developing and interpreting football metrics.
Q: What advice would you give to someone wanting to get into this space?
My advice would be to sufficiently focus on relevant non-technical skills. My impression is that newcomers to the field overvalue hard skills and undervalue soft skills. In many situations, the ability to identify the need of the practitioners and to convert the analysis results in a data-driven story is far more important than programming or data science skills.
Q: What is your favorite quote or saying?
My motto in football analytics is "fewer numbers, more insights". I believe that football data scientists should focus on presenting actionable insights as data-driven stories rather than sets of numbers that are hard to interpret and understand by practitioners.
Please feel free to reach out to me if you are interested in discussing further!
Q: What is more intriguing to you: cultivating a team vying for European places or building a team that progressively staves off relegation? Why?
I love football clubs that try to punch above their weight by outsmarting their opponents, regardless of their season objectives. Football analytics has so much to offer when it comes to taking better-informed decisions, but most clubs have only scratched the surface! I would not be surprised if a few forward-thinking clubs that currently still reside in the lower divisions but have already discovered the potential of data and analytics ended up at the top of European football in the next decade or so.
Q: What is the biggest misconception or cliché regarding data or recruitment that you have encountered in this space?
Many traditional football scouts appear to believe that football data scientists aim to replace them with computers and numbers. This misconception considerably slows down the adoption of data-driven methods in football unfortunately. Instead, I believe that football scouts should view football metrics as a tool and an additional source of information that enables them to take better-informed decisions. In addition, these tools would enable them to do their work faster, allowing them to analyze and evaluate more potentially interesting transfer targets.
Q: If you could start over, what skill would you build on first?
Unlike many other people who are trying to get into football analytics nowadays, I first built my skill set and then started applying my skills to football data. However, if I were to develop my skill set from scratch, I would focus first on programming skills and then on data science skills. From mentoring more junior football data scientists, I learned that a lack of programming skills often stands in the way of quickly trying out new ideas. Furthermore, I believe that solid programming skills speed up the process of acquiring relevant data science and machine learning skills.
Q: What is more important: domain knowledge or curiosity?
I would say that one needs a good mix of domain knowledge and curiosity as a football data scientist. Domain knowledge helps to understand the need of the practitioners and to assess the quality of potential solutions, whereas curiosity helps to come up with potential solutions that might address the need of the practitioners.
Q: What is more impactful for a player: psychology or environment? Why?
I would say that the answer very much depends on the player and varies from one player to another. The impact of psychological and environmental factors on the performances of players are not very well understood yet. These factors definitely deserve more attention in the football analytics community.
Q: What coach or player would you give a lifetime contract to? Why?
Kevin De Bruyne would be my pick as I really enjoy watching him play for Manchester City and the Belgian national team. He is not only a great football player, but also a great example of why the risk versus reward trade-off should be taken into account when developing and interpreting football metrics.
Q: What advice would you give to someone wanting to get into this space?
My advice would be to sufficiently focus on relevant non-technical skills. My impression is that newcomers to the field overvalue hard skills and undervalue soft skills. In many situations, the ability to identify the need of the practitioners and to convert the analysis results in a data-driven story is far more important than programming or data science skills.
Q: What is your favorite quote or saying?
My motto in football analytics is "fewer numbers, more insights". I believe that football data scientists should focus on presenting actionable insights as data-driven stories rather than sets of numbers that are hard to interpret and understand by practitioners.
Please feel free to reach out to me if you are interested in discussing further!
No comments:
Post a Comment