Ok, time for my first ‘proper’ article after my short intro.
One massive benefit of doing my podcast is being able to pick the brains or people working at the very top of the game. Ask questions, get opinions, and of course try to learn as much as I can.
During my recent conversation with Jonny Whitmore, Director of Analytics at Opta, he mentioned a statistic that genuinely caught my attention. Some recent analyst roles that he had hired for, had attracted around 1,500 applications.
On the surface, that number tells you something about how competitive the football analytics industry has become. But I think it also tells us something much more interesting about how the profession itself is changing.
Ten or fifteen years ago, having the ability to code or build statistical models was enough to make you stand out. Football analytics was still a relatively new field, and simply being able to work with data was a significant advantage.
Today, that is no longer the case.
The industry has matured. There is more public data than ever before, there are countless online courses teaching everything from Python and SQL to machine learning, and open source projects have made it easier for people to develop their skills and showcase their work. AI tools such as ChatGPT and Claude have also made it much easier for people to learn new techniques and accelerate their technical development.
That is undoubtedly a positive thing. The analytics community has grown because knowledge has been shared more openly, and the number of people capable of contributing to the field has increased significantly.
However, there is a consequence to that progress.
Technical skills alone are becoming less of a differentiator.
If 1,500 people are applying for the same role, it is unlikely that knowing Python is what separates the successful candidate from everyone else. The ability to manipulate data, create visualisations or build models is increasingly becoming the entry point rather than the thing that makes someone exceptional.
This raises a more interesting question…
What value does an analyst actually create?
For years, aspiring analysts have understandably focused on improving their technical ability. Which programming language should I learn? Which statistical methods should I understand? How can I build a more advanced model? These are all useful questions, and technical development remains incredibly important.
But perhaps they are not the questions that matter most anymore. The more important question is whether the analysis helps someone make a better decision.
One of the things that stood out from my conversation with Jonny was that we spent surprisingly little time discussing algorithms or coding. Instead, we kept returning to the importance of communication, application and understanding the problems that need solving.
That is a subtle but important shift.
The value of analytics has never really come from producing more numbers. It comes from using information to help people make better decisions.
A recruitment analyst does not need another metric simply because another metric can be created. They need information that helps them identify the right player. A coach does not need the most complex dashboard possible. They need insights that improve preparation and decision making. A broadcaster does not need endless statistics. They need context that helps supporters understand the game in a more meaningful way.
The challenge has changed.
For much of the last decade, football analytics was focused on proving what could be measured. The industry wanted richer event data, tracking data, better expected goals models, possession value models and increasingly sophisticated ways of describing what happens on the pitch.
Those developments have transformed our understanding of football. We can now quantify elements of the game that would have seemed impossible to analyse not long ago. We can evaluate chance quality, passing difficulty, pressing intensity, off ball movement and countless other aspects of performance.
The answer to the question, “Can we measure this?” is increasingly yes.
The more interesting question is whether we should.
Measurement itself is no longer the difficult part. The difficult part is judgement. Understanding which problems are worth solving, which questions are worth asking and which insights will genuinely influence decisions.
That is where great analysts separate themselves.
The best analysts are not necessarily the ones who can build the most complicated models. They are the ones who understand why a model matters in the first place. They know how to connect data with football. They understand the context behind the numbers and can explain complex ideas in a way that coaches, recruiters and decision makers can actually use.
Technical ability still matters. Of course it does. Statistics matter. Coding matters. Understanding methodology matters.
But increasingly, those skills are becoming the foundation rather than the differentiator.
The analysts who stand out will be those who combine technical knowledge with curiosity, judgement and communication.
Interestingly, when I asked Jonny whether he thought artificial intelligence would create more or fewer analyst jobs, his answer was more nuanced than simply saying it would replace people.
His view was that AI could potentially lead to fewer analyst roles in the short term as organisations become more efficient and some tasks become automated. However, he also suggested that this may only be a temporary effect. As organisations begin to understand what skilled analysts can achieve when they are supported by AI, rather than replaced by it, demand could increase again.
I think that is an important distinction.
AI is likely to change the role of the analyst, but changing a profession is not the same as removing it. Throughout history, technology has often reduced the time required to complete certain tasks, but it has also raised expectations. The people who benefit most are usually those who know how to use new tools to create more value.
The same is likely to be true with AI.
A person who can use AI to write code faster, explore ideas more efficiently and test hypotheses more quickly will be more valuable than someone who simply performs those tasks manually. But the technology still needs someone who understands the problem, questions the output and knows whether the answer is actually useful.
That is the part that is much harder to automate.
Football analytics has, in many ways, become a victim of its own success. The barriers to entry have never been lower, but that means the challenge of standing out has never been greater.
More people have access to the same tools. More people can produce impressive looking analysis. More people can build models and create visualisations.
That is a good thing for the industry.
But it means the definition of a great analyst has to evolve.
The future does not belong solely to the people who can create the most advanced models. It belongs to the people who understand football problems, ask better questions and communicate ideas clearly enough that others trust their conclusions.
The question aspiring analysts should perhaps be asking is no longer:
“Can I analyse this?”
It is:
“Can this analysis help someone make a better decision?”
Because that is where the real value lies.
Not in the complexity of the model. Not in the number of metrics produced. Not in how technically impressive something appears. The best analysts have always been problem solvers. The difference today is that everyone has access to better tools.
The advantage will belong to those who know how to use them to solve the right problems.
Check out the full conversation with Jonny, right here:
I’d be interested to hear your thoughts. Has football analytics reached a point where technical skills are no longer the biggest differentiator? Or do you think coding and modelling will continue to be the qualities organisations value most when hiring analysts?
Thanks for reading 👌🏼
First and foremost an analyst would need to understand what they are analysing and why. I see a lot of tables and graphs and charts on the internet for example which highlight who's good at something, but it's equally as important to find out what they are bad at, how they would fit in, what their role would be etc, but this goes for the recruiter of the analyst too... Do you need an analyst that can build a prototype or do you need an analyst who can spot patterns in data that others perhaps can't? There is so much data out there, we all potentially read data differently, which is why when recruiting analysts I would be looking for those who look at the data objectively, rather than using the data as their tool to provide proof for something which everyone already knows. As an analyst that has played football I sometimes think we over analyse the game, that metrics aren't as important as hunger, heart, leadership, personality, but data has no personality, so when recruiting a data analyst, is personality even needed? Each role at every business is different, but equally as important is not just picking the right candidate, but that candidate picking the right role.
Well done. Data and numbers can give the text of a story but the why is the plot that is hopefully intriguing for the reader. It definitely makes me think more when writing fantasy football data deep dive articles. Thanks for the article!