Open play is chaos shared by eleven players; a corner is a designed routine you can point at. Set pieces are a large and growing share of goals, and because a dead ball is separable from open play, the set-piece coach produces the cleanest, most isolable fingerprint of any actor on the staff. The engine rates him on both ends, the goals he wins from dead balls and the ones he stops the team conceding, pulls it out of the manager's credit, and feeds it straight into the simulation. It is the cleanest signal in the building, and also the noisiest, so it is trusted carefully.
Most of what happens on a pitch is tangled: a goal in open play is the striker, the passer, the system, the space, the moment. A set piece is a rehearsed situation from a stopped ball, so its output is measurable and largely separable from open play. That separability is why the set-piece coach's contribution isolates more cleanly than any other actor's.
This is the actor the rest of the market underrates precisely because it is now separable enough to measure. A good set-piece coach is worth real points a season, on a channel that used to be dismissed as luck, and the engine treats those points as the coached, attributable thing they are. The dead ball is the one place where you can point at a coach and say exactly what he added.
Illustrative engine read on the real set-piece signal (output measurable and largely separable from open play; the cleanest portable fingerprint in the staff). Composite reads, demonstration figures.
The set-piece coach is rated on two residuals, not one. Attacking: set-piece expected goals and conversion above the baseline for the squad's attacking quality. Defending: set-piece expected goals conceded below the baseline for its defensive quality. Both are isolated from open play, and the read is taken at the team-season level, then sliced to the players who most drive it.
A coach who wins you a goal every few matches from corners and one who quietly stops you conceding them are doing different things, and a single number would blur them. The engine keeps the two ends apart, the same way it keeps a player's attacking and defending apart, so a set-piece coach who is a menace going forward but leaky at the back reads exactly as that. Read the dead ball at both ends, because a match can be won at one and lost at the other.
Illustrative engine read on the real set-piece output residual (attacking xG above baseline and defensive xG conceded below baseline, isolated from open play, at team-season level, sliced to delivery and target players). Composite reads, demonstration figures.
There is an honest tension here. Set-piece output is the cleanest contribution to isolate, but it is also the most volatile season to season, a handful of goals either way swings the raw number. So despite being the easiest to measure, this actor is confidence-gated harder than the others: the engine wants a fingerprint that holds across clubs and across seasons before it trusts it.
The engine refuses to crown a set-piece coach on one hot season, because it knows the channel is noisy, and it refuses to bury a good one on one cold one. It waits for the record, and once the fingerprint is real it flows straight into how the simulation prices a match, because a team that scores and defends dead balls above its quality is a genuinely harder opponent. The clearest signal still has to survive the noise before the engine will bet on it.
Illustrative engine read on the real confidence discipline (set-piece output volatile season to season, gated harder on sample), the Simulation set-piece feed, and the manager credit-split. Composite reads, demonstration figures.
Open play is chaos that belongs to everyone at once, but a dead ball is a designed routine with an author, and that makes the set-piece coach the cleanest, most isolable fingerprint on the staff. The engine reads him at both ends, the goals he manufactures from corners and free-kicks and the ones he keeps out, isolates both from open play, and slices the team residual to the taker and the target who drive it. It pulls the set-piece share out of the manager's credit so nothing is counted twice, and it flows the validated fingerprint straight into how the simulation prices a match. But it holds the clearest signal to the strictest standard, because set-piece output is noisy season to season, so it waits for a record that holds across seasons and clubs before it trusts the number, and it never lets any of it touch a player's rating. The cleanest signal in football still has to earn the bet.
The Set-Piece Coach read isolates the cleanest, most separable contribution in football, reads both ends of the dead ball, gates it hard for noise, and feeds the validated fingerprint into the match engine.