Mohamed Salah's Radar Love

When data paints a surprising picture

There’s a fantastic browser extension called FBCHARTS. Once installed, it will automatically produce a radar of stats for any player on their FBRef page. You can also use it to overlay one radar over another, enabling a comparison of players which will prove definitively and without question who is better than whom.

This is the obvious starting point for using this tool. Who doesn’t love arguing about sport on the internet? Don’t ask me; I posted the above radar on X without a comment then muted the conversation immediately. Thanks, but no thanks.

There are other, far more intriguing, applications for radar comparisons. When a club signs a player, it does so in the hope that it will improve their squad, and ideally their starting XI. In most instances, the idea will be for the footballer to deliver what they gave to their previous club as a baseline, and then improve from there.

So, who has done this? By using scouting reports from different seasons for one player, we can see how their performance changed, if it did. But while this might be a good idea in theory, experimenting with the radars revealed a key issue, namely that many players who change clubs aren’t expected to fulfil identical roles within both teams.

Liverpool’s transfer business from the summer of 2023 is Exhibit A. Let’s start with Alexis Mac Allister, and do a little generalising. In his final Brighton season, he either played in the holding pair or as the number 10 in a 4-2-3-1 system. At Liverpool, he was the sole six or an eight in a 4-3-3 formation. Trying to compare all of that under the umbrella term of midfielder is pointless.

Similarly, Dominik Szoboszlai was previously a wide forward or the right-sided attacking midfielder in a different set up to that deployed by Jürgen Klopp. Wataru Endō probably had the most similar role for his two clubs in the last two seasons, but one battled relegation while the other challenged for the league title.

Playing around with radars did highlight something of interest, though. If you compare Liverpool’s Ryan Gravenberch with the version from his final season at Ajax, you get a neat visualisation of what might be called the Eredivisie tax on multiple metrics.

There’s another thought-provoking angle for using comparison radars for a single player. If the time has come to renegotiate their contract, are they still delivering output at the same level that they once were?

My initial intent with Mohamed Salah was to look if his final Roma season radar foretold in any way the explosion that took place once he joined the Reds. As FBRef’s detailed data stretches back to 2017/18, that wasn’t possible.

“But what the heck,” I thought, “let’s contrast the present day Salah with the one that delivered 32 goals and 10 assists in his first Premier League campaign with Liverpool.” Below is a comparison between a 31-year-old who suffered his longest injury absence for at least a decade and his former self, a freshly turned 25-year-old who delivered the second most non-penalty goal contributions in a 38-game season in the modern era of English football.

Between the end of 2017/18 and the start of last season, Salah added 251 Liverpool games to his tally. Add on his exertions with Egypt and it’s a total of 23,592 minutes, an additional 262 ‘per 90s’ in his legs. Yet his underlying performance barely dipped at all.

The only sizeable fall occurred with his success rate for take-ons, and that was inevitable. As far back as 2016, Colin Trainor was writing for Statsbomb that:

“The median winger will have 1.1 successful dribbles when they are 26 years old compared to 1.6 when they are 20 or 21 but their decreased ball carrying does not seem to have an adverse impact on ultimately how creative they are.”

The Salah of 2017/18 was 25 and completing 2.5 take-ons per 90 minutes, and only saw that drop to 2.2 the following season. In other words, he was doubling ‘the median winger’ in 2018/19 and wasn’t far adrift of the average 26-year-old in his age 31 season, with 1.0.

The primary FBRef scouting report contains 19 metrics. It details the player’s per 90 minute average for each, as well as their percentile standing against positional peers from the competition in question.

If you compare the average percentile changes for the player examples above, the findings are fascinating. Endō’s dropped by 18.6, largely as he wasn’t attacking much for the Reds like he had for Stuttgart, while Gravenberch went down by 13.9 thanks to the aforementioned tax.

The trend was not down for everyone. Szoboszlai (1.5) and Mac Allister (4.5) saw their averages rise despite role changes, which augers well. As for Salah, he was matched across campaigns which had five years in-between and his average percentile standing went up by 0.4. Consistency personified.

The Egyptian appeared to play at center-forward in Arne Slot’s first friendly with Liverpool. The make-up of the squad which has flown to the United States for the three-game tour suggests he will likely do so in the matches with Real Betis, Arsenal and Manchester United too.

Dribbling isn’t as important in that role so with everything else largely the same, should Salah play centrally from now on? His radars for 2017/18 and 2023/24 make a compelling case that perhaps he should.