02ChallengesDone

Initial Challenge

Who has the ball? Every frame of a full match labelled home possession, away possession or dead ball, from automatic tracking and tactical tags, with no training on the match.

  • 199,319frames labelled H / A / D
  • < 10 sfull match on a laptop CPU
  • 0hand labels, anywhere
  • 15behaviour tests

AIn motion

The idea, animated

Animated synthetic play labelled frame by frame: control within 1.5 m, a touch that does not change possession, an interception that does, the ball lost by the tracker while possession is held, the ball going out, and the throw-in restart.
MotionThe animation will appear here once it is added./assets/hero/hero.svg
Evidence in, one coherent possession sequence out.

BWhat was asked

One label per frame

For each of the 199,319 frames of one match's tactical video, say who has the ball. Two official metrics grade the answer: spellF1, the ranking metric, scores whole possessions; macroF1 scores every frame.

Possession is control, and control persists: a pass in flight belongs to the passer, and a touch is not a turnover. Every label depends on the frames around it.

The catch: the ball is untracked on 41% of graded frames, so a missing ball has to count as no evidence, not as a stoppage.

  • Home team in possession
  • Away team in possession
  • Ball dead

CHow we read the problem

How we read the game

12 reads in four legs, from the question to the proof.

Leg 1

The problem

What is being asked

  1. 01Possession Control, not proximity. What possession means, frame by frame
  2. 02Decoding Decode the whole half. Why possession is a decoding problem

Leg 2

The scoring

How the answer is judged

  1. 03Metrics Count possessions, not frames. spellF1 and macroF1 pull apart
  2. 04Pareto No single best setting. The Pareto lens

Leg 3

The data

What the files can say

  1. 05Data map One clock that counts. What each data asset contributes
  2. 06Tracking Slots are not players. Tactical tracking: the master clock
  3. 07Ball gaps No ball, no evidence. The ball goes missing where possession is decided
  4. 08Tags Context, not events. Tags are context, not events
  5. 09Homography Pixels to pitch. Homography: the bridge between pixels and pitch
  6. 10Broadcast Two cameras, two clocks. Two cameras, two clocks

Leg 4

The rules

What is allowed, and proven

  1. 11Rules Rules are the architecture. The rules are the architecture
  2. 12Glossary Seven false equivalences. Glossary and failure map

DMethod

Three moves, no training

  1. Evidence

    Per frame: a player within 1.5 m of the ball and closer than any opponent, ball out of play, collective stillness, restart windows from Smart Tagging, and calibrated team signals from every tag code.

  2. Decode

    An exact Viterbi pass per half over three states (H, A, D), with a cost for every change of state. One noisy frame cannot flip possession; a reception and a touch can.

  3. Choose

    Parameters come from a Pareto front of spellF1 and macroF1 on simulated matches with known answers. The submitted profile was picked on a real, externally labelled Bundesliga match.

EExternal validation

Tested on labels we did not make

Official metric on open DFL match J03WOH. “Degraded” hides the ball in bursts on 41% of frames, the competition rate, and adds 0.15 m noise, to resemble the competition tracking.
Method spellF1clean macroF1clean spellF1degraded macroF1degraded Possessionsdegraded, truth 402
Organiser baseline method (smooth50) 0.301 0.385 0.235 0.383 570
Ours, hand-reasoned 0.442 0.835 0.249 0.661 166
Ours, balanced 0.448 0.855 0.286 0.686 186

Open DFL/IDSSE data (Bassek et al. 2025, CC-BY 4.0), scored with the official, unchanged metric. There is no Smart Tagging for DFL, so this measures the tracking part of the pipeline.

Read the last column too. The baseline method reports 570 possessions where the match has 402: it invents turnovers. We report 249: turnovers hidden in long tracking gaps arrive late, and that gap is what the roadmap goes after.

FGo deeper

Code and report

Reproduce the submission with Docker in under ten seconds, read the full reasoning, trials and limitations in the technical report, or walk the design call by call in the architecture study.