Why he is a ball player and not a no-nonsense

#560. Ross: "You pick a position so centre backs and then we show for that player why we think he is a ball player or a no nonsense... this also shows how close they are to other roles that we might want to use in our tactic."

Two questions, two numbers. Overall is how good he is in that role, base included — it answers "would I pick him". Suits him ignores everything the six roles agree on and scores only what makes each role itself — it answers "which one is he". Twenty-one real centre backs, the seeded roles, and RoleScoring's own arithmetic, computed in the page.

Player

Phase

Roles with demands of their own

Role	Suits him	Against your squad	In squad	Overall	What sets this role apart, and what he posts

The standard options nothing much of their own to ask for

Role	Overall	Against your squad	In squad	Why it is not ranked above

How the two numbers are worked out

Overall is his weighted average attribute as a percentage of 20, over everything the role weights — the base plus its own. 75 means averaging 15; 90 would mean averaging 18, which nobody reaches: across all 126 player-and-role combinations in this squad the highest is 82.5. The usable range is roughly 25 to 85, which is why the bar marks where your squad actually sits rather than pretending the scale is 0 to 100.

Suits him needed three attempts. Scoring only what a role adds beyond the shared base was not enough, because the six roles also overlap with each other — nothing at all is unique to Ball-Playing. What separates it is not which columns it asks for but how hard: Passing and Composure at Major where its siblings ask at Useful or not at all. So for every column, take the average weight across the six roles; a role's own demand is its weight minus that average, kept where positive. Ball-Playing's comes out as Passing +1.5, First Touch +2.0, Technique +1.3, Vision +2.0, Composure +1.0.

And a role can fail to have demands at all. Totalling that extra weight: Overlapping 41.5, Advanced 32.7, Ball-Playing 13.2, Wide 8.0 — then Centre-Back 1.3 and No-Nonsense 1.0. Those last two ask for what every centre-back role asks for, a fraction more. Less than one column's worth of extra weight is not a specialism, so they are listed separately rather than allowed to win a contest on a sliver. Out of possession the same threshold splits the same way: 14.7, 14.7, 12.0, 9.5 against 2.7 and 2.2.

What building it found

Our scoring reproduced FM's own fault, and this fixes it. Ross, from the FM screens: "de Ligt is the better defender but Kim has better passing (+2), composure (+1), vision (+3)... we need to say yes de Ligt is the better allrounder but if you want a ball playing then Kim is better." Scored the old way, de Ligt leads Ball-Playing 75 to 73 — the same answer FM gives, for the same reason: his defending drowns the question. On what actually makes the role, Kim 66, de Ligt 62.
Van Dijk, put through the same arithmetic, comes out top of this squad as a ball player — 81 against Kim's 66, and 1st of 22. His overall score makes him 1st of 22 for all six roles, which is true and useless for picking one. Only the second number separates them.
De Ligt has no specialism at all, and that is the right answer. His four distinctive scores land on 63, 62, 62, 62 while his standard-option scores are 79 and 81. That is what an all-rounder looks like when you measure him this way: excellent at the job every centre back does, no particular flavour.
Two roles out of possession are identical by construction. Covering Centre-Back and Covering Wide Centre-Back carry the same 14.7 of distinctive weight over the same columns, so they tie for everybody, forever (#499). Switch phase and they sit together, always.
Attributes only, deliberately. Metrics need the squad-relative machinery, and thirteen of the twenty-one have no minutes, so the app already scores those thirteen on attributes alone.

Still open

Where this lives; whether the screen may write the sentence "he is a ball player"; how the identical pair is explained; and one property worth saying out loud — the distinctive weights depend on which roles exist in the position, so adding or removing a role shifts every number in the top table. That is defensible, since "what makes him a ball player" is inherently a comparison against the alternatives, but it is not absolute the way the overall score is.