Mohamed Salah: The transfer that changed football

AFC BournemouthArsenalAston VillaBrentfordBrighton & Hove AlbionBurnleyChelseaCrystal PalaceEvertonFulhamLeeds UnitedLiverpoolManchester CityManchester UnitedNewcastle UnitedNottingham ForestSunderlandTottenham HotspurWest Ham UnitedWolverhampton WanderersScores & ScheduleStandingsFantasyThe Athletic FC NewsletterPodcastsInside Arsenal’s Title WinMohamed Salah's arrival at Liverpool had an impact well beyond Merseyside Getty Images Share articleMohamed Salah’s extraordinary Liverpool career is drawing to an end. Since arriving at the club in the summer of 2017, the Egyptian has amassed 257 goals in 441 games — a record only bettered by two players in the club’s history. His time at Liverpool has not been without controversy — as events in the last week have underlined — but his legend is secure. His legacy, however, spreads far beyond Merseyside. This week, The Athletic is publishing a special three-part series examining Salah’s time at Anfield, including his playing legacy and his wider impact as a social and cultural icon. Today, we examine how his move from Roma transformed football’s transfer market, proving what could be done with data and why the smartest signings are not always the most obvious. This is not an ode to Mohamed Salah. The goals, trophies and countless broken records already speak for themselves, with column inches rightly filled this week with tributes to one of the all-time greats of Liverpool and the Premier League. Instead, Salah’s departure is also an opportunity to recognise the impact of his signing — not just for Liverpool, but the role it played in waking up the football world to the value of data and analytics. Ironically, it is impossible to quantify the ripple effect that Salah’s transfer caused across the rest of European football. But how was a “failed Chelsea winger” — one who made just six Premier League starts at the London club — plucked from Italian side Roma for such a cost-effective fee without other clubs queuing up for his signature? The story goes that Jurgen Klopp preferred to sign a young Julian Brandt at Bayer Leverkusen in the summer of 2017, but the wider recruitment committee at Liverpool — which included Klopp, Michael Edwards (then technical director), Dave Fallows (head of recruitment), Barry Hunter (chief scout) and Ian Graham (director of research) — convinced their manager that Salah was head and shoulders above any other option on the market. Klopp pushed back at that idea on his most recent visit to Merseyside but, whatever the precise details, there was no question that it was only by digging through the data that the case to sign Salah became unanswerable. “From a complicated data point of view, he ticked all the boxes,” Graham told The Athletic in 2024. “He came out as the best wide forward in Europe aged 24 or under. Mo came with the baggage of having failed in the Premier League, but our data analysis helped us to understand that we could ignore that.” Parting with £37million — a bargain even by 2017 standards — Salah, by then 25 years old, joined Liverpool from Roma with relatively little fanfare from the wider football sphere. Salah might now be viewed as the poster boy for Liverpool’s recruitment strategy, but this was a process instilled from the very top of the club years before, when Fenway Sports Group (FSG) bought Liverpool in 2010. Led by U.S. private equity tycoons John W. Henry and Tom Werner, the investment came with an agreement that the club would lean into a smart, data-led approach that brought them similar success in their baseball ventures. Henry’s Boston Red Sox won their first World Series championship in 86 years in 2004 with the help of statistician Bill James, who was hired as senior advisor on baseball operations. Liverpool were not the only club showing innovation in their use of data in the 2010s. Sarah Rudd was head of analytics for Arsenal after they bought the company she worked for, StatDNA, in 2012. Brought in by then-Arsenal chief executive Ivan Gazidis, Rudd’s department worked closely with the recruitment team to support Arsene Wenger on potential signings. “The fact that you had massive clubs like Liverpool and Arsenal being successful (using data) was great, but you also had smaller clubs like Brighton and Brentford succeeding,” Rudd says. “So it showed that this (method) works, it’s not just a club having a lot of money. Data can help to make really good decisions and evaluate players better than a scout in isolation.” Brighton and Brentford are two clubs that have consistently maximised the transfer market thanks to the professional gambling backgrounds of their respective owners, Tony Bloom and Matthew Benham. Crucially, their rise from the English Football League to the Premier League has come from a data-led strategy that is woven into the fabric of the club.  Their stories are packed with their own intrigue, but neither has Liverpool’s media profile. The Anfield club’s rise back to the pinnacle of the English game — capped by winning a first title since 1990 in 2020 — was a scenario that was comparable to Henry’s previous impact in American sport. “If you look at what caused analytics to take off in baseball, it was actually only once the big teams started to combine that level of intelligence with money,” said Luke Bornn, the co-founder of Zelus Analytics (now Teamworks Intelligence), and who has worked with multiple European clubs, including Toulouse, AZ and AC Milan. “In the early days, it was the Oakland Athletics winning way above their payroll — which is impressive, but they didn’t win the championship. It was only when the (Boston) Red Sox, the (New York) Yankees and the (Los Angeles) Dodgers started to combine deep pockets with intelligence that it became a lethal combination.  “There’s an analogy with football here. Liverpool were not the first team to use data, but going in with Ian Graham and Michael Edwards at the highest level, I think that’s really what raised the attention.  “How many articles were written about AZ in the Netherlands over the last decade? Very few. How many are written about Liverpool’s analytics? Hundreds, and Salah took that to the forefront.”   Bornn was working as a professor at Harvard University when he answered Roma’s call to be their head of analytics in 2015. Salah was playing at Fiorentina at the time, but even the most basic statistical model revealed that the Egypt international’s profile was worth investing in. “His data was so overwhelmingly strong that it was pretty clear that the €15 million option (to buy) was an absolute no-brainer,” Bornn says. “It was just a case of looking at this guy’s ability to progress the ball. Early analytics work quantified how players who move the ball towards the opposition’s goal are extremely valuable — and Salah did that in spades, both creating for himself and others.” Two productive seasons at Roma were all it took before Salah returned to the Premier League, after that unsuccessful stint at Chelsea that yielded just two goals in 19 appearances. Salah was Liverpool’s chosen man, but knowing that Brandt was an alternative option provides a parallel universe of what could have been. It is an example that highlights another crucial role of analytics within clubs: steering a manager away from players, as well as towards them. “There were definitely a lot of signings that didn’t happen because we differed from the narrative of the scouts,” says Rudd, the co-founder of analytics company ‘src ftbl’. “It wasn’t always like this, but we’re really good at identifying the good players across the industry now. What’s a lot harder is finding the right player for the team at the right moment, at the right price. “Using data and analytics can help answer those questions a lot more. But we can also say this is not the right player for us. He might be wonderful for another team, but what we need this player to do is very different from how he’s operating. I think we can better quantify those risks going from one environment to another.”  Bornn agrees with the sentiment. “My biggest contribution is the things that didn’t actually happen,” he says. “There’s an alternative world here where a bunch of other transactions were made instead of these ones. Often, the value of analytics is the things that you don’t see — it’s stopping the really poor decisions.” Salah will always be the icon of Liverpool’s success, but the arrivals of Roberto Firmino, Sadio Mane, Georginio Wijnaldum, Andrew Robertson, Fabinho, and Diogo Jota are further examples of their fruitful data-led recruitment strategy over the last decade. Eyebrows were occasionally raised by rival clubs at the time, but Liverpool’s unwavering trust in their process was crucial to the advantages they have benefited from in the long-term. It is no coincidence that their shrewdest signings came at a time when Graham’s research team were at their most prominent. “Using data consistently is really hard, because you’re consistently overriding your human intuition,” Bornn says. “Data tends to agree with the human eye about 80 per cent of the time, but when you’re truly driving decisions with data, it’s that 20 per cent — when there’s a real disconnect with what you see with your eyes — where you get the benefits. “We’re not wired to acknowledge our flaws or biases, and override them by using these statistical models. If you really want to make a difference, it’s how you act when the data doesn’t match your eye — that’s when you get really interesting conversations and potentially a real competitive advantage.” Today, nearly every Premier League club has some form of dedicated data department. While the value of analytics within recruitment is well-established, surprisingly few clubs are using it to its full potential. “After the original Moneyball book (published in 2003), you had the movie (released in 2011), which started to take this story into the public consciousness,” Bornn says. “Then later, once Liverpool started to get media attention, that really caused teams to say, ‘We have to do this too’. “Whether the subsequent teams have used data correctly or not is another question. I think there’s a bunch of factors at play, but I would say that (Liverpool’s approach) was a big step change.” The support that FSG has given to Liverpool’s research and data science team has been crucial in their recruitment success, but having such deeply-rooted support from those at the top of the club is not always the case. “I would say the number of clubs still doing this work in the optimal way is quite small,” Rudd says. “What we often see is a lot of involvement and a lot of good departments, but they don’t have that organisational structure to really maximise the impact with it.” At the very least, most clubs know that data can provide you with the tools to make smart recruitment decisions, but there is no guarantee that a player will succeed once they walk through the door.  Salah’s impact was immediate, with Liverpool unlikely to make a more influential signing for a very long time. The process to find him was relatively simple: smart minds, a data-led approach and the bravery to ignore the outside noise. Other clubs have been searching for their own version ever since. Spot the pattern. Connect the terms Find the hidden link between sports terms