In football, the adoption of data teams is comparatively recent and has been slow to gain momentum compared to corporate data teams. Stopforth explains that his focus is essentially split into two sections: tech and data. The tech side is constantly assessing what each department at the training ground does and their processes to see if there are gaps in what is being done, what can become more efficient and to see if further tech investments and integrations can provide the answer to any problems. This involves a lot of innovation and Stopforth is constantly in conversation with up-and-coming tech innovators that are keen to get into football data and examining how these new solutions can be utilised for Brentford FC.
Regarding the data side, Stopforth states that data is used “day-to-day on all levels, training analysis, de-briefing environments and competitor knowledge”. Despite the constant use of data, there is a mindset that is moving away from immediate-term attitudes of what data can provide into a broader image and larger, long-term projects across all aspects of the organisation. However, connecting different types of data for different areas (medical, recruitment, etc) is difficult because it is siloed, but this is changing. “It is my job to bring all this together, but it is a large long-term project,” said Stopforth. “I have been speaking to companies about data lake integration, which outside of football is quite normalised, but inside football is slower to take hold.”
Data usage
Stopforth describes himself as a research department discovering the new technologies that can be the next best thing for Brentford’s development. “We have a scientific approach to new technologies that includes setting up experiments and picking apart the results to see if this will truly help the players on the pitch, which is the ultimate aim of a football club,” said Stopforth.
As football clubs embrace data and data teams, there are new challenges for clubs like Brentford that have been using data for many years. These challenges include having more people and new technologies involved in football data to contend with. One prime example is the use of player tracking data technologies which can be bought and integrated simply by teams without any prior data knowledge. “Compared to other clubs, we do not have the financial resources. We cannot outspend our competitors, but we can outthink them.”
One year, Brentford had a comfortable safety net in the league and realised that they would not be able to push to get higher up the league, but they could prepare for next year. Using data, the team noted that the transfer window in January often had inflated player prices, so they sold players for increased amounts and then reinvested the funds into the following season where a push within the league was more likely to see results.
In another example, Stopforth explains, “we are not as strong as other teams at counter attacking, so what can be done about it? I will use data to look at teams across the globe and their counter attacking metrics, form a list of suitable examples and then analyse footage and other data to better understand what they are doing well and how this relates to Brentford FC as a team.” Stopforth will examine if a team has a similar style of play and a similar player makeup as Brentford to then deep dive into how they achieve their results and what can Brentford do to enhance its own areas of weakness over multiple years. “It is hard to pinpoint the turning point of data showing its worth in the moment, it is always a case of looking back to see the moment. You recognise progression within the team and the longer-term strategies start to become more detailed.”
After this, Stopforth says it is important to have the senior board explaining why strategies are being used, such as selling high-performing players mid-season, and what the end goal of a certain decision is expected to be.
Owning Your Data
Rather than focusing on standardisation of data as a primary concern, Stopforth and the data team at Brentford are working to ensure the security and ownership of their data. “At the moment, a lot of our data is with outside companies which has limitations,” said Stopforth. “For example, if the club was to be sold, we cannot say ‘here is our data for the last ten years’ to show a potential buyer as it is stored on other platforms.”
The other concern about external businesses operating your data is what happens if a business relationship breaks down and Brentford no longer wants to work with a specific company, or what if one of the outside partners goes under? “We use data really well in our processes, but we still have a long way to go in terms of storage and linking it together,” said Stopforth, who is working on the development of datalakes for Brentford, but this investment is new territory for many football teams. “When I joined, although it was a data-friendly environment, my desk was in the canteen – resources were minimalistic.”
Embracing Differences
It is an often-repeated question for those in data: do you need new hires to be data literate? Stopforth explained that Brentford FC “does not expect people to be data literate, but there are important questions around whether someone will be a good fit for the team and embrace the culture during the interview processes.”
With the advent of social media, football clubs have been able to expand their reach, and some have created competitions for data researchers where the winner is then hired off the back of their contest submission. “I always recruit with the team goals in mind and then find skillsets that are needed to achieve those goals. Can they be found within the existing team or another department? If it needs to be visual, I will try to find someone with computer design skills, or if it needs to be data heavy then I would find people with the skills and experience to deliver.”
As well as hiring new data staff, football has a notorious tendency to change managers throughout a season. Although there is a framework for data to be disseminated throughout the club, new managers tend to work differently, and this frequently changing environment needs to be catered for. “One previous manager was very conscious about feedback methods, so he would mix up his styles where different people would present to the players, or sometimes they would be split into groups with a laptop each and watch a specific part of the game where one of the players would then present,” said Stopforth. It is therefore imperative that the data teams create a system that can accept this level of flexibility whilst maintaining enough of a structure that inter-department communication and data standardisation can be implemented.
