Player Load in Football: How the Catapult Metric Works

Player Load is a sum of accelerometer-derived vector magnitudes:

• Tri-axial accelerometer. The wearable (a small pod between the shoulder blades) measures acceleration on three axes: forward-backward (X), side-to-side (Y), and up-down (Z).
• Vector magnitude calculation. Each frame (100 Hz typical), Catapult computes √((X_now − X_prev)² + (Y_now − Y_prev)² + (Z_now − Z_prev)²) — the change-in-acceleration vector.
• Summed across the session. All vector magnitudes are summed across the training session or match. The total is reported in arbitrary units (AU).
• Typical match Player Load values. ~700-1,100 AU for a Premier League midfielder over 90 minutes.
• Position-specific norms. CBs ~600-850 AU; CMs ~800-1,100 AU; wingers ~900-1,100 AU; FBs ~850-1,050 AU.

Distance-only GPS metrics underestimate intensity in three scenarios where Player Load excels:

• High-intensity short bouts. A 5-meter sprint with deceleration generates more accelerometer load than walking 50m. Distance treats them similarly; Player Load doesn't.
• Changes of direction. A 180° turn generates lateral acceleration with no distance gained — invisible to GPS, but high cost on the body.
• Jumps and aerial duels. Vertical accelerations during jumping for headers; contributes to load but produces zero horizontal distance.
• Tackles and contact events. Impact-driven accelerations that fall outside any speed-based metric.

Three primary applications:

• Daily load monitoring. Coaches plan training intensity using Player Load targets — e.g., 70% of match load on a moderate day, 100% on a high day.
• Acute-chronic workload ratio (ACWR). Player Load is one of the most-used inputs to ACWR calculations (acute = 7-day rolling Player Load, chronic = 28-day average).
• Return-to-play protocols. Post-injury, players progress through graded Player Load targets (50%, 70%, 90%, 100% of pre-injury baseline) before being cleared.
• Position-specific benchmarks. Each player has individual baseline Player Load by training-day type and match-day type — anomalies trigger review.

Four honest constraints:

• Vendor lock-in. Player Load is Catapult-specific. STATSports has a similar but not identical metric (Dynamic Stress Load); GPSports has its own. Inter-vendor comparison is unreliable.
• Doesn't directly predict injury. Despite ACWR research, Player Load alone is a weak injury-risk predictor. Combined with sleep, RPE, and prior injury history, it's informative — alone, it's not.
• Position-context dependent. A high Player Load reading is normal for a winger but anomalous for a CB. Coaches must interpret per position.
• Internal load missing. Player Load is purely external (movement). It doesn't capture internal load (cardiovascular, perceived effort) — must be paired with HR data and session RPE for the full picture.

Where Player Load fits in the load-monitoring stack:

• Total Distance (TD). Simple but undercounts intensity. Use alongside Player Load for full picture.
• High-Speed Running (HSR). Distance covered above a threshold (typically 5.5 m/s or 19.8 km/h). Captures a specific intensity bin.
• Sprint distance. Distance above 7.0 m/s (25.2 km/h). Highest-intensity bin.
• Player Load. Captures full-body movement load — including elements (jumps, COD, decelerations) that HSR and sprint distance miss.
• Session RPE. Subjective perceived effort × duration. Internal-load proxy. Combine with Player Load for internal + external view.
• Heart rate / TRIMP. Direct internal-load measure. Pair with Player Load.

Three recommendations for clubs:

• Standardise on one vendor. Catapult, STATSports, or other — pick one. Cross-vendor comparisons are unreliable.
• Build position-specific baselines. Track Player Load per position over a full season; build the team's individual norms before benchmarking.
• Combine with internal load. Pair Player Load with session RPE and (where available) HR data. External-only is half the picture.