Most people searching for this are really asking two different questions at once. What is football analytics, and how do I get paid to do it. The answer depends on which one you care about more.
I came at this with no football contacts, no stats degree, and no industry hook. I had a gambling background, a self-taught patch of Python and Tableau, and a habit of publishing rough work online. It took me two years to go from "I want in" to paid full-time.
Everyone's journey is different. But the people I see stall are the ones who treat football analytics as one thing and look for one path. There isn't one. There are at least four.
What football analytics actually is (and what it is not)
Football analytics is a category, not a job title.
When a club or agency says they're hiring "an analyst", that could mean four very different roles, reporting to different departments, with different skill demands. The advice for someone chasing a recruitment analyst role is not the advice for someone chasing a data scientist role at a vendor. If you miss that distinction, you'll waste six months learning the wrong things.
There are also two broad sides of the industry. Video analysis, which sits close to the coaching staff. And data analysis, which sits close to recruitment, the sporting director, or a sports science department. They overlap more than they used to. They're still not the same job.
Here are the four doors:
Recruitment/scouting analyst. Lives inside a club's recruitment department. Uses data and video together to build shortlists, profile targets, and feed recommendations to the sporting director or head of recruitment. The output is a decision someone can act on.
Performance analyst. Sits with the coaching staff. Codes match footage, builds opposition reports, presents clips to players and coaches. More video-heavy. More presentation-heavy. Less about databases.
Data scientist at a club. Rarer than the internet makes it sound. Builds models, player ratings, and in some cases event-level physical or tactical analysis. Reports into a head of analytics or sporting director. Python-heavy, SQL-heavy, statistics-heavy.
Analyst at a data vendor. StatsBomb, Wyscout, Opta, SkillCorner, Football Radar. Not working for a club, working for the company that sells data to clubs. Could be in research, product, customer success, or consulting. Often the most accessible entry point for people coming from outside football.
| Role | Main output | Main tools | Reports to |
|---|---|---|---|
| Recruitment/scouting analyst | Shortlists, player profiles, transfer recommendations | Wyscout, StatsBomb, Excel, sometimes Python | Head of recruitment / sporting director |
| Performance analyst | Opposition reports, post-match clips, pre-match presentations | Hudl Sportscode, InStat, Wyscout | Head coach / analysis lead |
| Data scientist at a club | Models, ratings, dashboards, research | Python/R, SQL, cloud tools | Head of analytics / sporting director |
| Analyst at a data vendor | Research, product work, client analysis | Python, SQL, the vendor's own platform | Research lead / product lead |
Pick the door before you pick the tools. The rest of this guide assumes you know which one you're walking through.
Who actually gets hired (and why the passion discount is real)
Entry-level football analytics in the UK pays roughly £18,000 to £28,000. Some first roles are voluntary. Some are part-time alongside another job. Nobody is breaking in on £45k.
Why so low. Supply exceeds demand, especially at Premier League level. Every club gets stacks of applications for any entry role. For every full-time job posted publicly, there are unpaid candidates already doing similar work for free to get in the door. That's the passion discount and it's real.
This is not meant to put you off. It's meant to protect you from the wrong expectation. If you come in thinking you'll be on £40k in a year, you'll quit before the career even starts.
The filter is not "who is most qualified". Data science skills on their own won't get you in. You'll be up against people with ten years of watching and writing about football. Football knowledge on its own won't get you in either. You'll be up against people with portfolios full of Python models.
The real filter is: can you do the work, and can the person hiring you actually see the work.
That second half is what people miss. You can be capable and invisible. The industry rewards visible capability.
The skills that matter, ranked by what gets you hired
If I had to rank them for someone starting today.
1. Football knowledge. You're competing against people who've watched 2,000 games and can tell you why Brighton's build-up shape shifted after the 2023 manager change. You cannot fake this. Watch, read, form opinions, defend them. The good news: this is free and you can start today.
2. Python or R. Python is more common now. You need pandas for data manipulation, matplotlib or mplsoccer for visualisation, and enough scikit-learn to run a simple model without breaking it. You do not need to be a full-stack engineer.
3. SQL. Clubs and vendors run databases. The ability to pull and join data cleanly is quietly one of the most valuable skills in the industry. It also gets ignored on most "how to get into football analytics" lists.
4. Data visualisation. Tableau, Power BI, or code-based (Python/R). A well-designed chart moves a decision. A messy one wastes a meeting.
5. Statistics. Nothing exotic. Sample size. Regression to the mean. Per-90 beats totals. Why a 15-game hot streak doesn't mean a player is now that good. This is where most amateur analysis falls apart.
6. Communication. The most underrated skill in the industry. Turning a dataset into a decision a coach, scout, or sporting director can act on. If your output needs a footnote every time, the output is broken.
On tools, Wyscout is the single industry platform you'll see everywhere. StatsBomb for data. Hudl Sportscode and InStat for video. SkillCorner if you're in tracking data territory. Do not try to learn all of them at once. Pick the ones that match your door.
Where to learn, paid vs free
You can go a long way on free resources before you spend anything.
Free. StatsBomb's open data library on GitHub for real event data to play with. FC Python and McKay Johns on YouTube for walkthroughs. Edd Webster's GitHub for a curated list of resources that is probably better than anything you'll pay for. LinkedIn and X for the community itself.
One important update. In January 2026, Opta terminated its data agreement with FBref, so the advanced stats that made FBref the default free data source are gone. Basic historic data is still there. For the xG, progressive passes, and advanced metrics most people used it for, you'll need to look elsewhere. StatsBomb open data is the strongest free replacement for learning. Some community-built alternatives are filling the gap. Assume the free-data landscape keeps moving.
Paid, in priority order. A Wyscout subscription is the single best paid spend in this industry. Individual scout tiers currently sit around €299 to €399 a year depending on the level of video access you need, and availability can vary by market. If you're serious about recruitment or performance work, it unlocks the video and data most of the industry actually uses. Nothing else comes close for the money.
After that, structured courses. StatsBomb's intro to analytics, CIES Football Observatory, Sports Data Campus, David Sumpter's Soccermatics. All solid. None of them hire you. They give you a frame and some practice.
Books, if you like reading. Soccernomics by Simon Kuper and Stefan Szymanski. Football Hackers by Christoph Biermann. The Expected Goals Philosophy by James Tippett. Each gives you a different lens on how the industry thinks.
Can I get into football analytics without a degree. Yes. I did. Most of the people I've helped through The Recruitment Room did. A data science or sports analytics MSc helps if you're 18 and choosing a degree anyway. It is rarely worth £10,000 plus as a career change move in your late twenties. You can replicate most of that syllabus for free and spend the money on tools and time.
Certificates don't hurt. They don't hire you either. The portfolio does.
Your first project: pick one, finish it, publish it
The biggest mistake I see is people trying to learn everything before they build anything. Six months in, they've read ten books, half a Python course, and produced zero public work. They're no closer.
Ship rough. Get feedback. Iterate. That's the whole game.
Pick one of these based on your door:
If you're chasing a recruitment or scouting role. Build a scout report on a single player using whatever data you can get (StatsBomb open data, community sources, Wyscout if you have it) and video you can find online. Pick somebody playing outside the top five leagues. Give it structure: role, strengths, weaknesses, statistical profile, comparable players, a clear recommendation (sign, pass, monitor). Two or three pages. Publish it on LinkedIn.
If you're chasing a performance analysis role. Do a set-piece breakdown using public video. Pick a team, pick a phase (attacking corners, short free kicks, whatever). Clip the examples, show the pattern, explain what the team is trying to create, and where it's breaking down. A short video or a slide deck is better than a long essay here.
If you're chasing a data science or data vendor role. Build a simple player rating model in Python using StatsBomb open data. Keep it simple. Per-90 metrics. Role-adjusted comparisons. A notebook someone can actually read. Publish it to GitHub with a clear README and write a LinkedIn post explaining the thinking.
One project. Finished. Published. That is further than 80% of the people in your cohort will ever get.
The rule: rough and shared beats polished and hidden. Every single time.
How to actually get noticed
Visibility is the part of the industry nobody teaches properly.
Post the work publicly. In 2026, LinkedIn is where football analysts and hiring managers actually are. X still has a community but it's thinner than it was. Default to LinkedIn. Write a short post to go with the project. Explain what you did, what you found, what you'd do next.
Tag thoughtfully. Not spam-tag fifty analysts hoping one replies. Tag the two or three people whose work your piece is genuinely responding to. I tagged en masse once when I started. It didn't work. People noticed in the wrong way.
Comment with substance. The algorithm rewards engagement but hiring managers notice the person whose comments make them think, not the person who left "great post" on thirty updates a week.
DM people one step ahead of you. Not heads of recruitment at Premier League clubs. The analyst who got their first club role six months ago. They remember the path. They'll reply.
Volunteer at a semi-pro or academy club if you can. It's not glamorous. It's real experience and a line on your CV that says somebody trusted you with actual decisions.
Track what you send out. Who you DMed, what project you linked, what came back. A simple spreadsheet. Without it you'll repeat yourself and miss follow-ups.
What to do in your first 90 days
A clear plan for the first three months. Adjust the dates, not the shape.
Week 1-2. Pick your door. Pick your first project based on that door. Don't debate this for a month. Pick one and move.
Week 3-6. Build it. Ship it rough. Better a v1 you published than a v3 living in your drafts.
Week 7-10. Publish. Send it directly to three analysts one step ahead of you and ask for honest feedback. Iterate based on what they say, not based on what feels comfortable.
Week 11-13. Start the second project while the first is in review. Keep the first one live and circulating. Do not wait for permission before starting the next piece.
Four finished, published pieces a year is more than most people in this industry produce in public. That's the compounding bit nobody mentions. One a quarter for a year and you look completely different to someone who started at the same time and is still "learning Python".
The time horizon on all this is usually six months to two years. Some people break in faster. Some take longer. The ones who get there almost always have one thing in common: they kept going when the signal was quiet.
Football analytics rewards visible work, not credentials. That doesn't change once you're inside. The same habit that gets you in keeps you moving.
If you want a clearer breakdown of the role differences, I've written more on the two types of football analyst, on how I went from gambling to full-time football scout, and on why your online portfolio matters more than your CV. If you're still collecting qualifications and wondering why nothing's happening, read stop collecting certificates, start building proof. For the Python path specifically, building player ratings from scratch with Python walks through a first model.
If you want honest, specific updates from inside the industry, including what's working for people breaking in right now, join the newsletter. It's free and it's where I share the stuff that doesn't fit in articles.
And if you've decided this is the career you want and you want structure, mentorship, and people one step ahead of you to show you what they did, that's what The Recruitment Room is for. Visibility, portfolio, network, all in one place. Not for everyone. Right for the people who are done trying to figure it out alone.
Frequently asked questions
What qualifications do you need to work in football analytics?
None are mandatory. Most job specs list a degree as "preferred" rather than "required", and plenty of people working in analytics right now don't have one in a related field. What hiring managers actually look at is your portfolio, your visible thinking, and whether you can do the work. Certificates and courses don't hurt, but they aren't a substitute for shipped projects.
Can you get into football analytics without a degree?
Yes. I did. Most of the people I've helped break in did too. A data science or sports analytics MSc can help if you're choosing a degree anyway, but spending £10k plus as a career change in your late twenties rarely makes sense. Your time is better spent building a project, publishing it, and getting feedback.
How much do football analysts earn in the UK?
Entry-level roles sit roughly between £18,000 and £28,000. Mid-level roles at established clubs and vendors usually land in the £30,000 to £45,000 range. Senior analytics leads at Premier League clubs or major vendors go higher, but those roles are rare and almost always filled internally or through the network. Some first roles are voluntary or part-time and that's still normal at the entry point.
Do I need to know Python to work in football analytics?
It depends on your door. For data scientist or data vendor roles, yes, Python (or R) is non-negotiable. For recruitment and scouting analyst roles, you can get a long way with Wyscout, Excel, and strong football knowledge. Python helps everywhere but it's not the only route.
How long does it take to break into football analytics?
Six months to two years for most people who get there. Some make it faster, some take longer. The common factor in people who get in is simple: they kept producing visible work when the signal was quiet and nobody was replying.
What is the difference between a football scout and a football analyst?
A scout traditionally watches games, builds an opinion on players, and feeds that into recruitment. An analyst usually works with data, video, and structured processes to support those decisions. The roles have converged. A modern recruitment analyst does both, and a modern scout at a serious club uses data tools daily. The title matters less than what the role actually involves day to day.
What should my first football analytics project be?
Match the project to the door you're walking through. Chasing recruitment, build a scout report on one player with a clear recommendation. Chasing performance analysis, do a set-piece breakdown with clipped video. Chasing data science or a vendor role, build a simple player rating model in Python using StatsBomb open data and publish it to GitHub. One project, finished, published. That's further than most people get.