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 scouting 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 Koen Vossen:
How did your first opportunity in football come about?
Well this is actually quite interesting.. we were expecting the birth of our son and we had to stay in the hospital for a couple of days before it would happen. I brought my laptop to the hospital because I had some time to kill. Exactly at that moment Joe Mulberry was looking for some help with rewriting notebook code to a flask application. I was able to work for 2 full days on a fun project with tracking data, and the result is the SituSearch project. In other sports I have around 8 years experience working with data and video.
What attracted you to scouting/analytics? What’s more intriguing now names or numbers?
I have a background as software developer (several companies, currently CTO of TeamTV), and as instructor for coaching courses at the Dutch Korfball Association. Furthermore I like sport psychology. And I think the combination attracted me: what’s possible from software perspective and what’s useful from a coaching perspective. Maybe not only coaching perspective, but also instructor perspective: how does goal setting work and what kind of goals can you set, and are there ways to measure progression on those goals. What kind of data do you need, and is it possible to derived it from available data or do we need to generate new data?
So I believe in numbers and in subjective data.
What player or concept you fascinates you? What caught your eye?
I really like to idea of making the subjective more objective. How can we measure performance of a player in subjective terms. In a situation a player can make a lot of choices and which one is the best? That depends on ball position, opponent, team mates and the available space (thanks Joe ;-)). But was we know situations are never the same. Therefore we need to look at the result of the choice (and the action) to determine if it was the right know. It’s even possible that the decision was the best one, but the execution was bad. How can we measure that? Mariette van Maarseveen did some great research on this in her thesis “Reading the game”.
She uses a notation system to attach scores to all kind of outcomes of actions, on-ball and off-ball! A coach could create such a notation system derived from the playing style of the team.
Furthermore I like the work of Laura Jonker and Nico van Yperen on the XOET scan. The scan tries to measure the learning climate that a player needs, and how the current situation is different from the ideal situation. (More info at https://www.xoet.nl/xoetscan/ in Dutch).
What is the biggest misconception you’ve found in this space?
Misconception isn’t the right term but I think people often forget what it takes to make a product that can be used by other people. When I create a cool notebook with interesting plots which is useful for coaches, but with all paths hardcoded and using some local files I can’t distribute it to others. This is not soccer/sport specific but data science in general. So it’s important to know about DevOps as well. Also it’s important to learn about coding practices itself. It’s totally fine when an algorithm takes 2 hours to finish but it’s not acceptable when your end-users tries to interact with your application. How can you split your algorithm into running your model and storing the results, and the interface that queries that data? What needs to be available right away for the coach and what can wait for a day? This also requires knowledge of the domain itself and even more of the user-cases of the coaches/staff.
If you could start over what skill would you build on first?
Domain knowledge! It’s really important to know a lot about the domain to be able to build proper software. You need to be able to challenge your end-users when you want to build proper software. If you cannot challenge them you are just building the things they ask for and not what they actually need. I like the “tree swing” cartoon ( https://nl.pinterest.com/pin/
What is more important player development or recruitment? Why?
To be honest I don’t have a lot experience so I will answer this question based on what I see happening. I believe they go hand-in-hand. Player development is really important. Right now most players (also on amateur level) don’t have good coaches. So the first thing we need to start with is trainer-development. We need to make sure all players get better. Pro clubs have the obligation to help clubs in the region to improve the quality of the trainers. The result will be that players at all clubs in a region will become better at the game (let’s call it player development). A good way to monitor the development of all players in a region should be in place (with data and video, and human reports). The pro clubs then can use that “breeding pond” to do recruitment.
What is your favorite sports moment? Why?
Dennis Bergkamp's goal during world cup 1998. Probably everyone knows it. And why. Mostly because of the commentary of Jack van Gelder.
What coach/player/team inspires you? Why?
Not a single coach/player/team I think.. I love to see what’s happening in coach-land were more and more coaches work at the top-leagues without a history as pro player. This means people start to look at what someone can do instead of what they did (in a different role!), and that there is a group of coaches that keep on going even when they get a lot of resistance from the established order.
And Joe Mulberry.. I can ask him a bunch of stupid questions and he will always answer them. Great mentor!
What advice would you give to someone wanting to get into this space?
Find something you believe in and make sure you go for it. Don’t get frighten when people don’t like your idea and keep on going. Expect that your idea needs several iterations with coach/other users: share your work and improve it again and again. Keep challenging yourself if you are learning or only doing. A mentor can help with reflective question to keep you in learning mode.
Who is your favorite athlete? Why?
Not sure if I have a favorite athlete. But when I have to pick one it would probably be Nathan Rutjes. He is such a positive player! And does a lot with kids, always to trying to help others.
What is your favorite quote or saying?
There are a lot of quotes I like. Mostly from philosophers. For now I’d pick “The more you know, the more you realize you don’t know” by Aristotle. When you learn more about a subject you’ll notice all the pieces of the puzzle you are still missing. When you are still unexperienced it will look easy.. for the same reason.
No comments:
Post a Comment