Player tracking data

Player tracking data records the x, y, and sometimes z coordinates of every player and the ball or puck many times per second, using optical cameras, radar, or wearable sensors, throughout a full match rather than only at the moment of a shot or pass. This continuous stream is distinct from event data, which logs discrete actions like a goal, a foul, or a completed pass with a timestamp but nothing in between.

From raw coordinates, analysts derive speed, distance covered, acceleration, spacing between players, and pressure on the ball carrier, which in turn feed shot-quality models such as expected goals and player-workload metrics used for load management. This differs from box-score statistics, which summarize outcomes but say nothing about positioning or effort that did not directly produce a recorded event.

Player tracking data matters because it lets analysts evaluate contributions that never show up in a box score, defensive positioning, off-ball movement, and fatigue, extending the tradition of sabermetrics into aspects of the game earlier metrics could not see. The main pitfalls are the sheer volume and noise of the data, which requires careful cleaning before use, and the fact that tracking systems and their underlying models differ between vendors, so raw feeds are not always directly comparable across leagues or providers.