How to Win the Premier League: The Inside Story of Football’s Data Revolution

by Ian Graham

THE SUNDAY TIMES BESTSELLER AND A FINANCIAL TIMES, SUNDAY TIMES AND TELEGRAPH BOOK OF THE YEAR 2024

'The best book on football I have ever read' Daniel Finkelstein

'Fascinating and educational. An enjoyable and informative read' Sir Kenny Dalglish

'Deserves a place among the great modern books on football' Sam Wallace, chief football writer, Telegraph
______________________________

The insider account of the data revolution that has swept through the modern football world written by one of its key architects, Ian Graham.

Between 2012 and 2023, Ian Graham worked as Liverpool FC's Director of Research. His tenure coincided with the club’s greatest period of success since the 1980s, including winning the Premier League in 2020 – Liverpool’s first league title after an agonising three decades.

Here for the first time, Graham reveals the fascinating data that informed some of the club’s most pivotal moments of the past decade, from the appointment of Jurgen Klopp as manager in 2015 to the signing of Mohamed Salah in 2017. Along the way, he shares groundbreaking insight into the modern game, including how a season largely played behind closed doors transformed our understanding of home team advantage, or why the GOAT (greatest of all time) might not be who you think. And, in a game that is increasingly dominated by an elite few, Graham charts a path for the future where a data-savvy competitor will always find the edge.
______________________________

'A fascinating, witty and remarkably insightful account of Liverpool's reinvention and a clear-eyed, detailed explanation of the inner workings of modern football. Nobody is better qualified to tell either of those stories than Ian Graham, who stands as one of the most (quietly) significant figures in the recent history of the club and the sport' Rory Smith, author of Expected Goals and Chief Soccer Correspondent, New York Times

'Never before has the data revolution in football been described so well from the inside.' Christoph Biermann, author of Football Hackers

‘How to Win the Premier League gives you three books for the price of a book about all of the silly inefficiencies plaguing European football, a book about what it's like to be empowered to use outside-the-box thinking to help turn your boyhood club into the best team in the world, and a book about how football actually works. If he really wanted to, Ian could easily put me out of a job.’
Ryan O’Hanlon, author of Net Gains and staff writer ESPN

*A Sunday Times bestseller August 2024

Genres: Football, Sports, Nonfiction, Soccer, Audiobook, Business, Technology

296 pages, Kindle Edition

Published August 15, 2024

Reviews

[Matt Bayliss · Rating 5 out of 5]

I knew Keita was good for us and now I have a literal book I can throw at people to prove it

[Tom Holden · Rating 5 out of 5]

Audiobooked in two weeks. A brilliant and fascinating look at the role data can play in football. Makes me want to sesh FM. Brendan Rodgers is a chump

[Courtney Goodridge · Rating 4 out of 5]

Would have given 5 stars but the nerds have ruined football and I’m not happy about it (except when it benefits Liverpool, up the Reds 👊🏽)

[Alialiarya · Rating 4 out of 5]

با ترجمه خوب نشر گلگشت خوندم

[Emma Strawbridge · Rating 5 out of 5]

first of all, what an excellent book. ian graham is a great writer, his thoughts are organized clearly, and he is doing things that are statistically interesting. most importantly he cites his sources!! barring the heinous crime that is endnotes (presumably an editor’s choice rather than an author’s) graham cites his papers and all that good stuff. what i appreciated most was that he explicitly credited his coworkers for inspiration, for pointing out his mistakes, for arguing with him, and for directly helping sign particular players. it’s an excellent description (and seems like a rather ideal scenario especially with klopp as manager) of how much cooperation and compromise goes into turning a team around quickly. he’s also very good about looking back at and mentioning his mistakes, even if they ended up being minimal. he’s specific with what the mistake was and how the team learned as well. he honestly could have skipped the acknowledgments with how thorough he was with credit and recognition of successes and failures. another thing that was really well done in this book was (again thru the credit-giving process) how graham explained the ways data fell short of a holistic player evaluation and the importance of qualitative analysis, or at the very least, what role it played and how the results of its influence had been successful or not. it’s good to hear from both sides, the eye test and the analytics people, that both are useful. “of course, correlation doesn’t imply causation, but as Randall Munroe put it: ‘It does waggle its eyebrows suggestively and gesture furtively while mouthing “look over there”.’ (246) stats is probabilistic and ultimately a person has to decide what odds are good enough.

i read most of this book on the train from london and edinburgh and had a list of questions about football because there are SO many things that make it different from american sports that i wanted to learn more about. fortunately, a nice couple with an adorable baby were sat across from me and the dad was happy to answer my questions so shoutout to him. in no particular order, i asked about: transfers&the reporting of their fees, trades&few of them, loans (SO interesting), relegation, and what his favorite team was (chelsea). i did not get to ask about leagues being allowed to use different balls, how it was possible that man city could be accused of 115 breaches of league rules in 2023???, or why there are so many biting incidents going on across the league (i was particularly interested in this one). all in all great book and hiiii little baby sitting with me you are sooooo cute!!!

[Mohit Rauniyar · Rating 3 out of 5]

I really enjoyed the first part of this book , which felt rich with stories, insights, and well-grounded research. It pulled you in and made the whole data side of football feel alive and meaningful. The second half, though, dipped quite a bit. It leaned too heavily into numbers and lost some of that narrative energy that made the beginning so engaging.

It is also very Liverpool-centric, which is expected, but I did find myself wishing for a broader range of stories and perspectives.

A decent read, but not one that fully sustains its early promise.

[Tyler · Rating 3 out of 5]

It's rare that we get this sort of contemporary insight from someone deeply rooted in the game. I feel as if the book tries a bit too hard to be Moneyball instead of leaning into what Graham is best at which is first hand experience and behind the scenes evaluations.

Graham also falls into the common make myself look good trap where he spends inordinately more time on what they got right instead of the failures in process. There's a couple moments from the early years but as the narrative gets closer to present day Graham is loath to criticize anyone involved.

The last portion of the book had a few chapters that were wholly unnecessary (Messi v Ronaldo is a prime example) but there were nuggets throughout that gave good context to what decision making at high level football club is like.

[Elliott Charles · Rating 3 out of 5]

A nice read for anyone who's got an interest in the Premier League and appreciates anything data-driven. That said, given its reviews I did feel a tad underwhelmed. As someone who has read Michael Calvin's manager/scout books and Rory Smith's "Expected Goals", there was a lot of content in this book that I knew already.

Secondly, a lot of Graham's narrative is around the history of data analytics in football (rather than its insights or impact), a story I found quite dry, especially when told through just text and no visuals.

Finally, I thought the way the book was put together was also a bit peculiar. It started with a big section about the Klopp Liverpool era, but then seemingly went into little random chapters of subjects the author found interesting, including a weird chapter where he takes on the book The Numbers Game by trying to refute its analysis. Not least, the book never really answered its title...

That said, there were plenty of great tales to tell from the Klopp era which as a Liverpool fan I enjoyed, accompanied by other interesting titbits. I found most interested how the usage of data means clubs are more thorough and diligent in their transfers nowadays in particular (compared to other parts of their football departments), and just how far football has come in using data as a route to become a really professional industry. All the dinosaur pundits might bang on about how ridiculous xG is and how "we'd never use that in my day", but the fact this type of insight and data has clearly become so engrained in clubs shows who's right!

[Quincy Hopper · Rating 5 out of 5]

Absolutely excellent. Any book that references Thinking, Fast and Slow at least half as much as this book is automatically promoted to 4 stars. It’s just the way things work, I don’t make the rules

[Harry Osborne · Rating 3 out of 5]

Did not win the premier league after reading this book, do not recommend.

Nah actually quite interesting stuff about data and how anti-stats the biggest sport on earth has been, even today (when it’s quite obvious that it can help). Also interesting hearing about motivations behind certain player moves from and to Liverpool over the last decade or so.

Loses points (lol) on being slightly too self-led when it comes to setting expectations - e.g. “our data model was really good because it did X”, then next chapter “well X is actually dumb because we didn’t consider Y, therefore our new model Z is superior” and repeat.

[Aaron Perrin · Rating 4 out of 5]

Exceptional insight to data usage in modern football from 2010s to present day.

I now fully appreciate xG as a stat amongst some other metrics mentioned in this. The other key metric stat ‘possession value model’, explains how and why Ian Graham and his team have utilised this to determine successful club transfers prospect and true value player ratings.

Pains me to say as a united fan but Liverpool (although due to chance luck, at not signing tom ince) have dominated the premier league pound for pound in the last 8years, unfortunately for them, coincided in a pep man city era.

Some of the more technical aspects of data explanations in this, went a little over my head, with particular reference to book analysis of the ‘numbers game’ and the explanation of why goals are more important than clean sheets. But there are lots of gems in this. Data clearly and sensibly explain, data really is king.

If you rate football would highly recommend this as a read. Just wish it was united that had adopted this approach in the post fergie era

[Lily · Rating 4 out of 5]

Given that football is such a secretive business, I think Ian’s book is one of the best insights into the “data revolution” that you’re going to find. He’s exceptionally clear in explaining the analysis he undertook and the impact it had. There are some nice little anecdotes in there too which I appreciated as a Liverpool fan. At times reading part three felt more like reading a blog, but certainly a very good one.

[Flo · Rating 4 out of 5]

4,5 sehr viel gelernt

[T.M. Fairman · Rating 4 out of 5]

Felt like a sliding doors book

[Jonny Ord · Rating 4 out of 5]

Good one to get me out of my slump

[Jackie Hwang · Rating 3 out of 5]

Overall, good insights into the evolution of football data analytics, data in general, and evolution of the sport. Perhaps the most prominent theme Graham presents is that data is only useful in context: a player may look highly skilled in one area -- but is it because he plays for an easy league, gets a lot of chances, or is truly skilled? To combat this, Graham presents his own statistical models for integrating other data elements in his calculations when ranking players. Also, while it may seem 'obvious', Graham also asserts that it is critical for all parties of the club to be aligned: the data department, the manager, and the scouting team/transfer team. When these parts are at odds or have different philosophies, there become decision inefficiencies that impact results.
Graham also asserts that its important to not overestimate results over underlying performance -- but how can you when sport is so filled with emotion? Data is supposed to present us with the objective truth (when used correctly), but given the wealth of data now, it becomes easy to cherry pick the insights that you want to warrant your angle.
The evolution and development of football data is also interesting; previously, data teams were limited by lack of developed technology, inaccessibillity to data, and lack of sharing. Over time, as technology developed and enabled better collection, teams were able to analyse data and share it with other clubs. Seems like the league also decided it was in their best interest to develop the sport and placed in regulations to streamline and enforce data sharing.

[Caspian Taylor · Rating 4 out of 5]

Very interesting book, written by one of the pioneers of the data revolution in football. Made me want to work for a football team doing some sort of data analysis. Great combination of maths and football, so if you’re into those two things then would defo recommend this book

[Nicholas Bestbier · Rating 3 out of 5]

Often in sport books — I find there are so many business lessons. This book is no different. A phenomenal account of how Liverpool used data science to help give them an edge that culminated in winning the Premier League. I loved the descriptions of the development of the various algorithms they used. From assessing the influence of potential managers to controlling parts of the pitch. Really enjoyable read - inspired me to think more deeply about how to leverage data science.

[Atish]

Statistics is a corruption of mathematics and we need to return to theology and geometry

[Matthew Nash Knight · Rating 5 out of 5]

Very good book which brings in fascinating statistical ideas to football and makes you think about it in a completely different way. Well written and has piqued my interest further in data analytics.

Anecdotes from Ian’s time at Liverpool also well worth the read. 5/5 good book to start the year.

[Daniel McElvenny · Rating 4 out of 5]

Really enjoyed it, although a little too Liverpool-heavy for my liking

[Jackson Evans · Rating 3 out of 5]

Easy read - doesn’t really dive into any models in depth.
Love some confirmation bias towards Klopp and the kids.

[Juan A. · Rating 2 out of 5]

Despite its title, How to Win the Premier League doesn’t provide a formula for victory but explores how data can inform critical decisions in football. For fans, football is deeply emotional, driven by skilful players who surprise opponents and moments of triumph against adversity. Yet, translating these emotions into data is challenging, as not all goals carry equal weight when viewed through the lens of passion and struggle. Reading this book, I highlight the importance of a club’s football philosophy as the foundation for long-term success. Building a team where players embody the club’s identity and grow together fosters a strong connection between the team, supporters, and the club itself. However, money remains a double-edged sword, capable of both creating and destabilising teams. Ultimately, football is a blend of art and emotion, where some elements defy measurement. While data can guide strategy, the heart of the game lies in its unpredictability and human spirit.

[Ben Ell · Rating 2 out of 5]

Found the first quarter really engaging, could see how different managers worked with the data team and lots of interesting insights. After that though... I think he struggled to turn his data experience into something that engaging for a reader (certainly for me).

[Liam Joyce · Rating 4 out of 5]

Mostly fascinating, but did almost slip into “norse in pub who knows everything” a couple of times.

[Jamie · Rating 4 out of 5]

Surprised he didn't marry Michael Edwards at the end.

[Michael Checchia · Rating 5 out of 5]

No one at Everton has read this book

[Joshua · Rating 5 out of 5]

Really liked the depth of descriptions into the stats and how they are calculated and used. I think some clubs this season accidentally sorted their list of transfer targets bottom to top.

[Pete · Rating 4 out of 5]

How to Win the Premier League: The Inside Story of Football’s Data Revolution by Ian Graham is a really interesting inside view of how data can help football teams play better. Graham has has a PhD in physics. He worked for Liverpool in their analytics department when they won the Champions League and the Premier League.

The book is really interesting because of the perspective it gives on football analytics. Graham first worked for a firm that was contracted to Tottenham. There he and the data analytics team had to prove their worth. They faced coaches who distrusted the data and had to slowly make inroads.

Graham worked with Michael Edwards at Tottenham. Edwards later became the director of football at Liverpool. There Edwards hired Graham. Initially Brendan Rogers was the coach there and Rogers was not keen on analytics. Despite an excellent result and nearly winner the Premier League Rogers did not work out. After Rogers Liverpool hired Jurgen Klopp, who valued analytics from his time at Dortmund.

Graham does an excellent job of describing how hard football is to analyse. He points out that possession and pass completion data is insufficient for good analysis. He talks about how Expected Goals (xG) works and also its limitations.

The book has lots on how players are critical for a team and how to evaluate them. Graham makes the very startling point that most transfers don’t work. He establishes a reasonable criteria to evaluate them. He suggests starting in the team after 12 months as a good mark. He also points out that only 2 in 5 transfers meet this mark. Given the huge costs paid in transfers this represents a very poor return.

Graham goes on to describe how good metrics can help teams to get better players. But he also describes their limitations. He describes how a football director, the manager, the scouts and the analytics team need to work together to get the best players. He provides really interesting examples of where the analytics didn’t show the value of some players. Luka Modric is one.

There are also some interesting discussions of where Graham thinks statistics have been incorrectly used. He looks at how the value of clean sheets has been overstated by some authors. He also writes about how money and analytics can work together. There he cites Manchester City as an example.

How to Win the Premier League is a really interesting book. Anyone who is interested in football and analytics will really enjoy it.

[Samuel Smith · Rating 3 out of 5]

Interesting-ish. Nothing hugely insightful. Was an hoping to be surprised by the quality of the analysis but that wasn’t really the case.

The author does the thing that all good analysts do of making their models sound more impressive than they really are by using jargon. Not a knock on them, that’s just what happens and is necessary in a book like this.

I guess it emphasises to me that the big advantage Liverpool had was the data itself and the decision-making processes. What matters is not the complexity of the model but the ability to collect accurate data (something which economic policymakers in this country are grappling with) and interpret it in a sensible way.

Liverpool benefited from being able to do this. It’s sort of shocking that other clubs (I.e. United) didn’t have processes in place to do this simple analysis. It sounds like they could have benefited a lot from an even just a few smart undergrads.

Having praised Liverpool, I would just say that if your model says Naby Keita is a good player, then your model is wrong. The guy’s first touch and decision-making were so inconsistent.

This is the second football book I’ve read this year which makes Brendan Rodgers look completely incompetent.