Technology is becoming more influential within football. From VAR to player wearables, the game is changing at an incredible pace. Teams are adopting or pioneering new methods for recruitment, training, fan engagement, opposition analysis, and more. This article will focus on the data side of these technologies and show you how you can get started with a career in football.
Not too long ago, a football team might have hired a scout based solely on their past experience in the footballing world. The average age of this scout may have been 60 years old, and a history of playing professional football could well have been a CV requirement.
Fast forward to 2022, and many football teams are on the front-foot, hiring data scientists for football roles. The brightest minds coming out of universities now have fantastic opportunities to work in football. The BBC reported in 2021 that “data experts are becoming football’s best signing”. If you are interested in football and always had a way with numbers, there really has been no better time to get started with football analytics.
Surround Yourself with Great Content
So, where do you start this journey? If you have not studied data science or do not have a mathematical degree from university, don’t worry. There’s plenty of scope within football analytics that you can focus on and provide value. In this day and age, we can learn anything online. I personally never went to university and every skill I needed to build OddAlerts was acquired thanks to the internet, desire and discipline.
I would recommend surrounding yourself with smart footballing minds. I do this by routinely adding quality content I see on Twitter to a list I have created. I have made this list public, so you can view all of the contents and subscribe to the list. I’m always on the lookout for top quality analysis as I continue my own educational journey into football analytics, so expect the list to grow considerably in the future. It contains a great mix of different perspectives on the game. UEFA Pro coaches, journalists, analysts, students, fans… Following this list on Twitter will let you see what kind of content is being put out by those that do this as a full time job.
Where To Get Football Data
If you have never worked with “data” before, prepare to be overwhelmed. I’m not trying to put you off, but it’s just the reality of the world you are jumping into. The market leader for footballing data is StatsBomb who records over 3,000 unique events per fixture. StatsBomb provides their invaluable data to football teams such as Liverpool, Dortmund, Wolves, Fiorentina, AS Roma, Melbourne Victory… To name a few.
StatsBomb does a lot for the space it occupies and encourages those interested in football analytics to check out their open data on Github. The data is provided as JSON files, so that you can import it into your programs that you a) build or b) use. The data, whilst amazing, is just one side of this coin. Your job is to make sense of that data and bring insights and new ideas to the table.
Another popular source for football data with analysts is FBRef. The level of detail is fantastic and they offer easy enough ways to export the data to the format of your choice, ready for you to play around with.
Apps for Manipulating Football Data
During my journey into working with football data, I have come across the following apps that are commonly used:
Tablaeu is a paid product and is used by many in all aspects of business. It’s a powerful data visualisation tool for any use case. Here is Ninad Barbadikar (from the OddAlerts Twitter list) showing you how to create scatter charts, bar charts, and player dashboards. The work he has put in here is fantastic and much appreciated by the community. You can view a live example dashboard that uses World Cup 2018 data. This was put together by the amazing Edd Webster, who has also put together a Github Gist for Football Analytics projects. Bookmark that page!
R Studio is an open-source data science tool that lets you code with ‘R’ and see the results of your code in the way of data visualisations. As with the data itself, learning to code can be quite overwhelming at first but is very liberating once you get over the initial learning curve. Mark Wilkins, who is a tennis coach and a sports science graduate, has the perfect introduction into this software. His tutorial takes freely available StatsBomb data and has you creating pass map graphics. Learning R or Python would be a smart move if you are serious about a career in deep football analytics, so I will include a section on that below.
If you learn better with videos, then Friends of Tracking on YouTube has a 1h 32m tutorial on getting started with R and StatsBomb data.
As well as these products, I encourage you to check out the following:
Learning Python Football Analytics
If you are looking to learn Python, I must start by recommending the YouTube channel of McKay Jones. The educational content on there is amazing. His tutorials are easy to follow and can have you creating all types of charts, dashboards, heat maps, and more in no time.
Python is a programming language that is used for both scraping and manipulating data. Before jumping into any detailed tutorials that might scare you off, you should learn the fundamentals of Python. There are many websites and guides I could recommend, but seeing as this is aimed at those looking to get started with football analytics, then I would suggest you check out FC Python, a website dedicated to taking you through the educational journey step by step. You will first learn the basics of Python, and then move onto data analysis and visualisations.
Learning ‘R’ for Football Analytics
Again, I must lean on the incredible work being done by the community and suggest that you go and check out the work of Nicolo Figiani, who runs Inverted Winger. He has a two-part series on setting up R Studio, grabbing some data from FBRef and creating visualisations with that data. I really appreciated his writing style and his tutorials are easy to follow along with. Nicolo is also included in our Twitter List.
As you explore this space, you will soon see how amazing the community is and just like FC Python, there is a dedicated website to teaching R to football fans and aspiring analysts. It’s called FCRStats. It follows a similar format, taking you through the basics before diving into data visualisations and then scraping data from other sources.
Another superb resource for learning R is ‘R for Data Science’, a web port of a popular book by O’Reilly of the same title. If you are already familiar with R, you might want to check out Advanced R by Hadley Wickham, who has 10 years experience in coding with R.
As mentioned above, you will want to download R Studio. If you are running into any issues whilst coding, or with the app itself, don’t worry. The chances that someone else had the same issues during their learning curve is high, and there is a R Studio Community website that will help you.
Summary
These single resources and tutorials will not complete your knowledge and make you job ready, but I hope they will whet your appetite and put you in digital spaces where the amount you are able to learn is only limited by your desire. The YouTube accounts of Friends of Tracking and McKay Jones alone can have you spending hours every night learning something completely new with each day. If you are at the start of your analytics journey, I wish you the best of luck!
Written by Joe (Founder @ OddAlerts)