Player Load Equation Comparison - Analysis Pipeline

MATLAB code supporting: Not All Player Load Equations Are Equivalent: Formula Choice Alters Load Magnitude and Interpretation in Elite Football (Callaway, Ellis & Williams). This repository contains 1) the complete analysis pipeline used to process optical positional tracking data from the 2022 FIFA World Cup and compare eleven player load equations across four signal-processing conditions. and 2) code for generating figures from those data. Starting from raw PFF Sports JSONL tracking files, the pipeline extracts XY coordinates, derives acceleration using a central finite-difference scheme, applies Butterworth and Kalman smoothing conditions, accumulates all eleven equations per player-match observation, and produces the descriptive statistics, Bland-Altman agreement analyses, conversion coefficients, and rank-order comparisons reported in the manuscript. Requirements: MATLAB R2020b or later, Signal Processing Toolbox. Input: PFF Sports JSONL tracking files and event JSON files (not included; available from PFF Sports: https://www.blog.fc.pff.com/blog/pff-fc-release-2022-world-cup-data ). Output: Per-player-match load values for all eleven equations, Bland-Altman statistics, pairwise conversion coefficients, and all manuscript figures.