Alex Stewart looks at how scouting theory can not just identify player types, but make better use of them
It’s the fifth minute of extra time against Slovakia and England’s hopes of entering the quarter-finals at Euro 2024 are receding at pace, until Jude Bellingham thrashes an audacious overhead kick past Martin Dúbravka, scoring with England’s first shot on target in the game; Bellingham’s side go on to win with a goal from Harry Kane.
It’s the 83rd minute against Senegal in June 2025 and England are again trailing, before Bellingham, who has come off the bench, controls a ball on his thigh and hammers it in; the goal is disallowed, and England lose their first ever game to an African side, with a technically excellent Senegal incisive and slick on the break.
You can’t keep getting bailed out – eventually, the luck runs out.
Jude Bellingham is a superb footballer, doubtless one of the finest in the world at the moment. And this piece is not about his talent, or his ability to come up with clutch goals and assists for England or his club side, Real Madrid. It is, instead, about how we view footballers and what we expect from them, and how a combination of confusion or necessity or perhaps bad decision-making can lead to misidentifying or misusing talent.
Because Jude Bellingham is a ceiling-raiser, but England treat him like, maybe even need him to be, a floor-raiser.
Some scouting theory
First, what do we mean by that? In a recent, excellent piece by Ashwin Raman for SCOUTED, which prompted some of the thinking behind this article, a few helpful concepts are introduced (and see the end of this article for some more suggested reading):
Let’s pause for a bit and take a look at some scouting theory. For this, we require a quick detour to basketball analysis and the concept of ‘scalability’ or ‘portability’. Ben Taylor of Thinking Basketball writes that the concept refers to players “maintaining impact, or scaling up, on better and better offenses”, which is especially crucial to sides chasing NBA championships. Some skills, like off-ball movement and passing, scale up more on better offenses than others, like isolation scoring, which means the front office for a top-level offensive side should prioritise signing players with the former skills over the latter. As a result, we get ‘floor raisers’, who can elevate low-level or middling offenses to a decent level, and ‘ceiling raisers’, who make very good offenses even better.
Ashwin uses these concepts to investigate how Nico Williams raises the floor for Athletic Club, “being the focal point in a middling attack” with high shot volume and usage rate (to which we shall return), while “in a settled possession side like Spain, it’s even more evident how his off-ball intelligence improves an attack” that is already replete with attacking talent – for Spain, he raises the ceiling.
As Ashwin goes on to say:
If you sent one scout to only watch Williams’ domestic season and another scout to stick to Williams’ international appearances, you’d get two different-sounding scouting reports. The high-usage floor raiser that may scare some scouts because of possible ‘inefficiency’ or ‘rawness’ turns into a low-usage ceiling raiser who can thrive without the ball.
And that is a fun thing about international vs domestic football: it’s the closest we get to seeing the same player ported to another situation, despite the myriad differences; the player is, at least, the same age, same talent level, same fitness and so on.
Back to Bellingham
So, back to Bellingham. A quick check of his domestic work shows a versatile player, someone who went from connective 8 and occasional screening 6, to a box-crashing 10, to a advanced playmaker.
We can use TransferLab’s comparison function to see his qualitative outputs over the last four club seasons and it’s clear that he excels at pretty much whatever he’s asked to focus on. There has been some drop-off in a few areas in 2024/25, probably more a result of tactical tweaking at Real Madrid than any inherent decline but by and large, this is an elite footballer whether he plays 6, 8, or 10.
So, we have a world-class midfielder who can do everything at more-or-less elite level. At Dortmund, he was the connective tissue that facilitated the other players. Bellingham is very good on the half-turn, good at retaining possession under pressure and bypassing it. He kept play ticking over, safely and sensibly, while then occasionally breaking the lines or creating something interesting. Back when I worked at Tifo Football, I presented a video on Bellingham that emphasised these abilities: “He’s not the guy who plays the really sexy passes, but he is the guy who makes sure they continue to tick over.” And, as this piece from The Analyst back in 2023 shows, Bellingham’s performances at Dortmund suggested that his eventual role at Real Madrid could be anything and whatever it ended up being, it would be good, especially if he added creativity and goals to his understanding of space, his engine, his technical ability under pressure, and his passing. And he did, bringing greater box movement, more ‘gravity’, and more threat.
Om Arvind wrote a neat summary of portability in football, which we can use to show that Bellingham is, absolutely and unequivocally, a ceiling-raiser:
Stemming from that logic, here are the key traits and skills that make portable football attackers, ordered by their importance to scalability (this is definitely up for debate):
- Off-ball movement: running beyond the last line, advanced positioning/resisting coming to the ball, and box movement; they all have an effect on opponent defensive structure even if the person making these motions never gets possession (off-ball gravity).
- Positional versatility and diverse spatial occupation, which allows a coach to deploy a player in multiple different formations and attacking set-ups. Aided by two-footedness.
- Good passing, including “connective tissue” link-up and combination play.
- Pressing and defensive work-rate.
- Aerial ability.
Together, this combination of attributes should allow an attacker to make most top-level sides in world football better.
Bellingham has 1-4 without question, and isn’t bad in the air, either. He’s definitely ticked one ceiling-raiser box (well, four).
Which brings us neatly to usage. Usage is another basketball concept, explained by Ashwin here: “A player’s ‘usage rate’ refers to the percentage of possession sequences a player ‘ends’ through shots, turnovers, and sometimes shot assists, in both football and basketball.” Generally, two things hold true: floor-raisers (strong individuals who make weak teams better) are comparatively and proportionally high usage players; and having two or three high usage players together doesn’t work.
Using Martin Hawkes-Teeter’s calculation, explained in this piece, I calculated the usage rate for Bellingham and his respective club mates for the four seasons, 2021/22-2024/25 (using FBref.com data), for all players who played over 20% of available minutes (players with fewer minutes can skew usage stats by ball-hogging in late-stage game-states). What we can see clearly is that Bellingham is not a high usage player, but he is a player with generally very positive outcomes.
| Season | Usage Rate | Ranking | Positive outcome % | Ranking |
| 2021/22 | 10.62 | 10th | 22.96 | 4th |
| 2022/23 | 13.19 | 7th | 22.13 | 11th |
| 2023/24 | 13.40 | 6th | 33.82 | 3rd |
| 2024/25 | 13.11 | 6th | 30.64 | 5th |
One could argue that this stat is most useful for out-and-out attacking players, and I wouldn’t disagree – the point here is that Bellingham again shows himself to be a ceiling-raiser, not a floor-raiser.
The full tables are the end of the article for anyone who is interested.
So as a ceiling-raiser, we would expect Bellingham to scale his output playing alongside better players and add, rather than detract, from his teammates. And he has done exactly that since the move from Dortmund to Real. Playing with world class players, Bellingham has added goals and creative output, improving his positive outcome percentage drastically, while not compromising his all-round excellence – pretty much the definition of ceiling-raising.
His goals and assists output has increased, peaking in 2023/24, as he has moved higher up the pitch. And the journey from 6/8 to 8/10 is pretty clear from visualisations. The Dortmund Bellingham played deeper and more centrally, while the Real Madrid Bellingham plays higher and a lot more in the left half-space and penalty area.
And this is reflected in a tighter shot-map, one more reflective of a striker than a midfielder.
England – floored by the talent
And this is the problem for England.
Because England have tried to shift their talisman forward to play in tandem with Harry Kane, and it’s not clicking. To check this, we can look at usage rates from England’s World Cup 2022 campaign, where Bellingham played as the kind of connective player we saw at Dortmund. He played as a 6 in the first two games and then as a left-sided 8 behind a pretty attacking front three of Foden, Kane, and Saka.
England’s usage rates in that tournament show Bellingham roughly equivalent to his Dortmund outputs (again, excluding players with fewer than 20% of minutes, sorry Jack). Usage hog Rashford excelled off the bench as an impact sub; otherwise there’s a nice split across pitch zones and players.
| Player | Minutes | Possessions Used | Usage Rate | % Positive |
| Rashford | 140 | 32 | 19.44 | 34.38 |
| Shaw | 424 | 77 | 15.45 | 15.58 |
| Foden | 258 | 44 | 14.51 | 31.82 |
| Trippier | 206 | 34 | 14.04 | 11.76 |
| Henderson | 271 | 44 | 13.81 | 13.64 |
| Saka | 288 | 45 | 13.29 | 26.67 |
| Kane | 402 | 58 | 12.27 | 31.03 |
| Maguire | 429 | 61 | 12.10 | 13.11 |
| Bellingham | 414 | 47 | 9.66 | 21.28 |
| Mount | 186 | 18 | 8.23 | 27.78 |
| Sterling | 149 | 14 | 7.99 | 21.43 |
| Walker | 236 | 20 | 7.21 | 0.00 |
| Pickford | 450 | 34 | 6.43 | 0.00 |
| Rice | 417 | 26 | 5.30 | 3.85 |
| Stones | 434 | 24 | 4.70 | 20.83 |
But what of England in 2024 at the European Championships? Excluding players with fewer than 20% of minutes, Bellingham is suddenly third.
| Player | Minutes | Possessions Used | Usage Rate | % Positive |
| Palmer | 149 | 35 | 24.01 | 17.14 |
| Foden | 617 | 85 | 14.08 | 22.35 |
| Bellingham | 670 | 85 | 12.97 | 15.29 |
| Trippier | 450 | 56 | 12.72 | 17.86 |
| Saka | 632 | 75 | 12.13 | 21.33 |
| Pickford | 690 | 78 | 11.56 | 0.00 |
| Walker | 690 | 69 | 10.22 | 1.45 |
| Kane | 601 | 58 | 9.87 | 37.93 |
| Rice | 690 | 62 | 9.19 | 22.58 |
| Mainoo | 368 | 31 | 8.61 | 25.81 |
| Shaw | 178 | 13 | 7.47 | 7.69 |
| Gallagher | 120 | 8 | 6.81 | 0.00 |
| Guehi | 570 | 34 | 6.10 | 11.76 |
| Stones | 690 | 37 | 5.48 | 5.41 |
And if you exclude Palmer, the usage king super sub who only just sneaks over the 20% threshold, equivalent to 138 minutes, it looks even worse. Bellingham has more possessions than anyone else save Foden, and is only second because Foden played fewer minutes by a small margin.
Crucially, 2022’s Bellingham is roughly equal to his Dortmund percentage for positive outcomes for possession, whereas in 2024, it’s less than half what he managed for Real Madrid, averaged across the two seasons. Bellingham is suddently doing proportionally far more and he’s also doing it far less well.
| Player | Minutes | Possessions Used | Usage Rate | % Positive |
| Foden | 617 | 85 | 14.78 | 22.35 |
| Bellingham | 670 | 85 | 13.61 | 15.29 |
| Trippier | 450 | 56 | 13.35 | 17.86 |
| Saka | 632 | 75 | 12.73 | 21.33 |
| Pickford | 690 | 78 | 12.13 | 0.00 |
| Walker | 690 | 69 | 10.73 | 1.45 |
| Kane | 601 | 58 | 10.36 | 37.93 |
| Rice | 690 | 62 | 9.64 | 22.58 |
| Mainoo | 368 | 31 | 9.04 | 25.81 |
| Shaw | 178 | 13 | 7.84 | 7.69 |
| Gallagher | 120 | 8 | 7.15 | 0.00 |
| Guehi | 570 | 34 | 6.40 | 11.76 |
| Stones | 690 | 37 | 5.75 | 5.41 |
(It’s also worth noting that the next best players for usage rates after Palmer from the full, unfiltered squad, were Alexander-Arnold and Eze, both subs with fewer than 140 mins, which suggests a degree of desperation in the use of subs).
Banging one’s head against the ceiling
So why is this happening? Why is one of the best players in the world nose-diving in positive possession outcome percentage even as he becomes England’s go-to guy? Well, scouting theory suggests it’s precisely because he’s become England’s go-to guy.
But let’s first look at a reason why that might have happened. For one, it’s pretty simple to see an England player scoring a hatful of goals for one of the best teams in the world and think, hey, he should do that for us! Carlo Ancelotti moved him to the 10 spot, effectively as a shadow striker too, rather than a playmaker, so again, playing behind Harry Kane and with quick wide men either side, surely it should work?
Jonathan Wilson wrote a great piece recently on his Wilson’s World (Of Football) newsletter explaining both symptom and cause for why not. Firstly, the symptom: tactical clutter.
There is a clear problem trying to play Jude Bellingham and Harry Kane together, now that Bellingham has ceased to be a deeper-lying midfielder. Kane likes to drop back into that space. Bukayo Saka or Cole Palmer like to drift from the right into that space. Eberechi Eze, if used on the left rather than centrally as he was on Saturday, likes to cut into that space. As does Phil Foden, whose form has fallen off a cliff. Rice likes to drive forward into that space. That’s five players all jockeying to operate in a central attacking midfield area. Adding to the issues are the lack of an attacking left-sided full-back who could overlap a wide forward tucking in, which means England inevitably skew to the right.
But the cause is more subtly diagnosed by Wilson:
This is what Scott Murray described in The Blizzard as the Roy of the Rovers syndrome that has held England back for years, the belief that one dynamic driving player from midfield would do something brilliant at the last to save the day (as Bryan Robson and Steven Gerrard often did), that faith obviating the need for a balanced structure that could control games.
Ken Early speaks often of something similar on the Second Captains podcast, how there is a debilitating obsession among the English pundit class with the idea that midfielders should score goals when Paris Saint-Germain just won the Champions League with a brilliant midfield trio that got two goals between them from open play in the knockout stage. It’s why English football culture struggles to produce deep-lying ball-playing midfielders. And when it does occasionally produce them, it often fails to appreciate them.
To each point: now there is no clear usage-dominant player, and too many middling-usage players vying for the same field position, things are cluttered; and, more importantly in terms of this piece, England have turned a world class ceiling-raiser into a floor-raiser, in large part because a) shifting to being a more attacking player worked for Bellingham elsewhere and b) England loves a box-crashing, do-or-die, last minute saviour from midfield. The narrative has shaped a tactical decision – it’s genuinely hard to see it any other way – while the context determined the first shift far more than England might have expected.
Because, and here is the strange thing, there is no reason why England should not excel and let Jude raise the ceiling. The squad is blessed with a surfeit of attacking talent outside the Real Madrid man. Bellingham could take his portable talents and use them to elevate the squad. Instead, England have focussed on the least portable, isolated attacking, and (and here is the thing which data can’t really measure) created an expectation around the kind of output that Bellingham should produce.
England do not seem blessed with many ceiling-raisers, at least when playing for England: players who knit things together, who get better as the quality of their teammates gets better, who use gravity and movement and guile to maximise opportunity rather than grabbing the ball and trying to do it all themselves (Saka is the closest, I would say, but needs far more help from the full back and right-sided midfielder). And they definitely are not using them in midfield.
So let Bellingham do what he was born to. Shuffle him back to 8 alongside Rice and a genuinely good 6 (Wharton?) and you have a proper midfield that can elevate those playing ahead.
But making Jude the man, the blood-and-thunder last-ditch floor raiser, dragging a mediocre England kicking and screaming into the next round is a misidentification of what he does well. It might feed into the narrative of what an English midfielder ought to be, but it is a waste of Bellingham’s enormous talent.
Further reading:
Mohamed Mohamed – Golden Boys of Yesteryear: Wayne Rooney (although the whole series is a must-read)
Alex Stewart for SCOUTED – Why the best relegated players make the worst signings
Ben Taylor – Backpicks GOAT: Philosophy and Science of Player Evaluation
Ben Taylor for FANSIDED – Nylon Calculus: The Thunder are who we thought they were
Header image copyright IMAGO / Fabio Ferrari / LaPresse
Usage tables
2021/22 – Borussia Dortmund
| Player | Minutes | Possessions Used | Usage Rate | % Positive |
| Meunier | 1402 | 300 | 16.17 | 12.33 |
| Hazard | 1337 | 286 | 16.16 | 15.73 |
| Brandt | 2078 | 415 | 15.09 | 20.72 |
| Dahoud | 1658 | 298 | 13.58 | 17.79 |
| Guerreiro | 1795 | 316 | 13.30 | 19.62 |
| Wolf | 1321 | 218 | 12.47 | 14.22 |
| Schulz | 836 | 134 | 12.11 | 8.96 |
| Reus | 2503 | 393 | 11.86 | 29.52 |
| Malen | 1684 | 239 | 10.72 | 29.71 |
| Bellingham | 2790 | 392 | 10.62 | 22.96 |
| Can | 1690 | 236 | 10.55 | 5.93 |
| Haaland | 1911 | 262 | 10.36 | 40.08 |
| Hummels | 1765 | 216 | 9.25 | 7.87 |
| Zagadou | 952 | 116 | 9.21 | 3.45 |
| Pongracic | 827 | 93 | 8.50 | 1.08 |
| Akanji | 2261 | 210 | 7.02 | 7.14 |
| Kobel | 2565 | 184 | 5.42 | 0.00 |
| Witsel | 1941 | 104 | 4.05 | 13.46 |
2022/23 – Borussia Dortmund
| Player | Minutes | Possessions Used | Usage Rate | % Positive |
| Brandt | 2397 | 478 | 15.06 | 25.73 |
| Wolf | 1570 | 301 | 14.47 | 15.28 |
| Guerreiro | 2290 | 439 | 14.47 | 23.46 |
| Malen | 1706 | 325 | 14.38 | 30.46 |
| Reyna | 625 | 119 | 14.38 | 31.93 |
| Adeyemi | 1386 | 252 | 13.73 | 27.78 |
| Bellingham | 2691 | 470 | 13.19 | 22.13 |
| Ryerson | 1410 | 235 | 12.58 | 10.21 |
| Reus | 1353 | 194 | 10.83 | 30.93 |
| Schlotterbeck | 2252 | 307 | 10.29 | 12.05 |
| Haller | 1225 | 152 | 9.37 | 34.21 |
| Moukoko | 1141 | 138 | 9.13 | 36.23 |
| Hummels | 2018 | 242 | 9.05 | 11.98 |
| Modeste | 813 | 89 | 8.27 | 34.83 |
| Sule | 2118 | 228 | 8.13 | 23.25 |
| Can | 1900 | 183 | 7.27 | 19.67 |
| Ozcan | 1576 | 136 | 6.52 | 9.56 |
| Kobel | 2430 | 178 | 5.53 | 0.00 |
| Meyer | 630 | 43 | 5.15 | 0.00 |
2023/24 – Real Madrid
| Player | Minutes | Possessions Used | Usage Rate | % Positive |
| Vinicius Junior | 1864 | 414 | 20.09 | 27.05 |
| Modric | 1688 | 327 | 17.52 | 26.30 |
| Vazquez | 1419 | 226 | 14.41 | 21.24 |
| Diaz | 1545 | 244 | 14.29 | 30.33 |
| Kroos | 2124 | 322 | 13.71 | 32.61 |
| Bellingham | 2315 | 343 | 13.40 | 33.82 |
| Rodrygo | 2380 | 348 | 13.23 | 39.37 |
| Garcia | 1398 | 196 | 12.68 | 15.31 |
| Carvajal | 2171 | 265 | 11.04 | 15.85 |
| Valverde | 2904 | 350 | 10.90 | 32.00 |
| Joselu | 1687 | 183 | 9.81 | 40.98 |
| Camavinga | 1817 | 195 | 9.71 | 20.00 |
| Alaba | 1154 | 119 | 9.33 | 21.01 |
| Tchouameni | 1987 | 189 | 8.60 | 20.63 |
| Rudiger | 2709 | 211 | 7.05 | 11.37 |
| Nacho | 1746 | 113 | 5.85 | 9.73 |
| Mendy | 1720 | 105 | 5.52 | 15.24 |
| Kepa | 1198 | 70 | 5.29 | 0.00 |
| Lunin | 1890 | 87 | 4.16 | 1.15 |
2024/25 – Real Madrid
| Player | Minutes | Possessions Used | Usage Rate | % Positive |
| Vinicius Junior | 2253 | 481 | 19.40 | 29.94 |
| Modric | 1827 | 349 | 17.36 | 26.93 |
| Guler | 1250 | 231 | 16.80 | 35.93 |
| Mbappe | 2907 | 506 | 15.82 | 40.71 |
| Rodrygo | 1928 | 324 | 15.27 | 33.33 |
| Bellingham | 2488 | 359 | 13.11 | 30.64 |
| Vazquez | 2189 | 279 | 11.58 | 19.00 |
| Garcia | 2171 | 262 | 10.97 | 15.27 |
| Camavinga | 1102 | 130 | 10.72 | 16.15 |
| Diaz | 1385 | 163 | 10.70 | 32.52 |
| Valverde | 3032 | 339 | 10.16 | 29.50 |
| Militao | 938 | 96 | 9.30 | 17.71 |
| Ceballos | 1215 | 113 | 8.45 | 25.66 |
| Rudiger | 2291 | 187 | 7.42 | 16.58 |
| Tchouameni | 2689 | 186 | 6.29 | 12.90 |
| Courtois | 2700 | 174 | 5.86 | 0.57 |
| Mendy | 1005 | 48 | 4.34 | 8.33 |
| Asencio | 1670 | 74 | 4.03 | 13.51 |
