Offside decisions are now routinely settled by video review with a graphic showing lines across the pitch. The process involves more judgment than the finished image suggests, and the uncertainty sits in two specific choices.
The frame must be chosen first
Offside is assessed at the moment the ball is played, so the review must identify the exact frame in which the passer's foot makes contact.
Cameras capture a limited number of frames per second, and the true contact may fall between two of them, which introduces a small error before any line is drawn.
Because attackers are usually moving, a fraction of a second of error corresponds to a real distance on the pitch, often comparable to the margins being judged.
The body point is a second judgment
The rule concerns the parts of the body that can legally score, so the line is drawn to a shoulder, knee or foot rather than to the whole player.
Selecting that point on a moving figure is done by an operator, and different operators may place it slightly differently on the same image.
Limb-tracking systems automate this by fitting a skeleton to the player, which improves consistency but still depends on the model's accuracy.
The lines are a projection, not a photograph
A camera views the pitch at an angle, so a line across the field must be calculated using the pitch geometry and the camera's calibrated position.
The resulting graphic is a reconstruction, and its precision depends on how well the camera has been calibrated for that particular stadium.
Presenting it as a clean image conveys a certainty the underlying measurement does not have, which is a communication problem rather than a technical one.
Automation reduces delay but not disagreement
Semi-automated systems track the ball and player limbs continuously and can flag a probable offside within seconds, removing most of the manual work.
The output still requires confirmation by an official, because the system reports a geometric fact rather than deciding whether a player was interfering with play.
Faster decisions have reduced the interruptions considerably, though they have not settled arguments about whether a margin measured in centimetres should decide a goal.
The margin question is a rule question
Proposals to award marginal calls to the attacker, by requiring clear daylight before an offside is given, address the outcome rather than the measurement.
Any such tolerance simply relocates the boundary, and cases will then cluster at the new edge exactly as they cluster at the current one.
The measurement can be improved indefinitely without resolving the underlying disagreement, which is about how a binary rule should treat a continuous quantity.
