The intelligence layer
for football.

TotalFootball.ai extracts 17 skeletal keypoints per player from standard footage at 25fps. A football-trained language model interprets those pose sequences and delivers real-time coaching intelligence for match analysis, player recruitment, training, and live commentary.

13M keypoints Per 90 minutes
100K hours LLM training
<1s latency End-to-end
Total Football LLM: Live Feed
System Analysis
Pressing intensity — second half. Blue #7 initiating press from 63% of defensive transitions. Distance per press: 8.2m avg. Wide forwards not engaging on trigger.
Coaching Insight
#7 is the primary press trigger but the wide forwards are not engaging on his cue — creating a coordination gap. Recommend a dedicated pressing shape session to align the trigger timing.
System Analysis
Pose Estimation: Live Demo
Wide-angle pitch detection
Full-Pitch Tracking

All players on the pitch are detected simultaneously in every frame, across the full playing area.

Kickoff team classification
Player & Team Classification

Automatic team assignment from kits. Players and officials identified instantly, no configuration required.

Body orientation skeleton overlay
Skeleton & Orientation

17-keypoint skeleton overlay per player. Shoulder angle, hip line, and body orientation extracted on every detected frame.

The Limits of Event Data

What Event Data Cannot Capture

Most football analytics platforms are built on tagged event data: discrete actions extracted from video at specific moments. These systems record that a pass happened, not how the passer's body was oriented, nor the spatial context that made it possible or prevented something better.

Standard Event Data

2D Event Data

  • Freeze frame snapshots of player positions
  • Manually tagged events (passes, shots, duels)
  • Aggregate statistics with no body context
  • Insights delivered hours after the match
  • No visibility on body posture, orientation, intent
TotalFootball.ai

3D Body Intelligence

  • Every body position on the pitch, every frame
  • 17 keypoints per player: nose to ankle
  • Shoulder orientation, body angle, open/closed stance
  • Processed in real time, no lag
  • LLM layer turns data into coaching language

How It Works

Three-Stage Processing Pipeline

Step 01

Vision Model Ingestion

Broadcast or training footage is processed by a convolutional vision model that detects every player on the pitch in each frame, at up to 4K resolution. No specialist cameras or pitch-side hardware required.

Step 02

Skeleton & Pose Mapping

17 keypoints are detected per player (nose, shoulders, elbows, left hip, right hip, knees, ankles) and connected to form a real-time skeleton representing exact body posture and orientation.

Step 03

Total Football LLM

Structured pose data is passed to the Total Football LLM, a domain-specific model trained on 100,000 hours of annotated match footage, coaching sessions, and tactical analysis. It translates keypoint sequences and orientation patterns into natural-language observations a coaching team can act on.

The Language Model

100,000 Hours of
Football Training Data

The Total Football LLM is a domain-specific language model trained on annotated match footage, training sessions, and coaching commentary, totalling approximately 100,000 hours of material.

It is fine-tuned to interpret sequences of skeletal keypoints and translate spatial and orientation patterns into natural-language observations, framed in the vocabulary coaches and analysts already use.

Total Football LLM: Live Feed
System Analysis
Shoulder orientation data, frames 115–265. Blue #9 showing consistently closed body posture when receiving in the left half. Open body % this half: 34%.
Coaching Insight
#9 is shielding possession rather than progressing it. His orientation is consistently facing away from goal at the point of receipt, limiting forward option selection. Suggest rotating the receiving body shape trigger in the next drill.
System Analysis

Applications

How the Data Is Used

Match Analysis

Body orientation, spatial positioning, and movement data are extracted continuously throughout the match. The LLM layer translates these into coaching observations, delivered in real time.

Recruitment

Skeletal keypoint data extracted from existing footage provides objective measurements of a target player's body mechanics, positional tendencies, and movement signature, independent of team or match context.

Training

Real-time pose data drives immediate audio feedback via Bluetooth earphones, delivered in the player's preferred language. Corrections and movement cues are issued during the session based on live skeletal data, not reviewed from video afterwards.

Commentary

Live positional and pose data feeds a real-time commentary engine that generates spoken narration and tactical analysis from objective spatial data, in multiple languages, suitable for television, radio, or digital broadcast.

Analysis

What the Data Captures

Pose estimation and LLM analysis combine to produce a broad set of outputs across fitness, tactics, individual technique, and team shape.

Physical

Fitness & Load

Distance, speed, and acceleration data captured at 25fps and segmented by intensity threshold. Covers every player across match and training footage.

Sprint Profiles

Sprint count, peak speed, distance, and recovery time per player, segmented by position and game state. Identifies fatigue patterns across a session or match.

Off-Ball Runs

Automatically detects and categorises attacking and defensive runs away from the ball — overlapping runs, third-man runs, channel runs, and diagonal movements.

Transition Speed

Measures how quickly teams and individuals reorganise following a change of possession, in both attacking and defensive directions, from first reaction to shape recovery.

Tactical

Shape & Formation

Identifies each team's true positional shape in and out of possession. Tracks defensive line height, width, structural compactness, and how shape shifts through transitions.

Phases of Play

Breaks each possession into build-up, progression, and creation phases, with full positional data for all 22 players throughout each phase.

Pressing Triggers

Identifies the positional and postural cues that initiate a team's coordinated press. Measures the speed, compactness, and effectiveness of each trigger event.

Defensive Coverage

Tracks ground covered by each player during out-of-possession phases, identifying defensive workload distribution and structural gaps in the team's defensive shape.

Technical

Body Orientation

Shoulder angle and hip line extracted per frame. Captures whether each player is open or closed relative to goal, teammates, and the defensive shape at the moment of receipt.

Receive & Turn

Body posture at the moment of receipt determines a player's ability to turn and progress. Every receiving situation is tracked, categorised, and scored.

Duel Positioning

Body orientation, balance, and weight distribution in 1v1 situations. Identifies whether a player is set to win possession or being manipulated out of position.

Pressure Intensity

Detects and quantifies when defenders close down a player in possession — speed, angle, and body position of the press captured for every pressing situation.

Passing & Creation

Line Breaking Passes

Identifies passes that successfully break a defensive or midfield line, using real-time player position data to define where each line sits in every frame.

Pass Predictions

Player orientation and positioning at the moment of passing, combined with teammate locations, to quantify pass difficulty and quality of option selection.

Set Piece Positioning

Player locations, movement patterns, and body orientation on every set piece, from delivery through to second ball. Covers corners, free kicks, and throw-ins.

Carry Detection

Identifies all ball-carrying runs, tracking duration, distance, direction relative to goal, and the defensive shape encountered during each carry.

Deployment

Local or Cloud

The Total Football LLM is available in two deployment configurations. Both run the same underlying model and produce identical output. The difference is where inference happens.

On-Device

No connectivity required

Inference runs entirely on local hardware at the venue. No data leaves the facility and no network connection is required, making it suitable for training pitches, pre-season camps, or away fixtures where connectivity cannot be guaranteed.

  • No internet connection required
  • Sub-second inference latency
  • Full data sovereignty: nothing leaves the site
  • Runs on standard hardware

Cloud

Always current, always scalable

Inference is handled by our hosted infrastructure. The model receives continuous updates as new training data is incorporated, and capacity scales automatically with session volume. Connects directly to existing analytics pipelines via REST or WebSocket.

  • Continuous model updates
  • Scales with session volume automatically
  • MCP Server, REST API & WebSocket access
  • Long-term data storage and retrieval

Integration

Output Formats &
System Integration

MCP Server
REST API
JSON
XML
WebSocket

Structured Data Output

TotalFootball.ai outputs structured data in real time. A REST API handles live streaming pipelines; standard formats (JSON and XML) support existing analytics tools; a native MCP server enables integration with AI-native toolchains.

Headless by Design

TotalFootball.ai has no proprietary interface. It is a pure intelligence layer: all output is consumed through the API, leaving your team free to build on top of it using any visualisation tool, analytics platform, or custom application.

Native MCP Server

TotalFootball.ai exposes a native Model Context Protocol (MCP) server. MCP is an open standard that allows AI models and agents to query live data sources directly. This means external LLMs and AI assistants can query pose data, player orientation, and LLM analysis in real time, without custom adapters or middleware.

TotalFootball.ai: Live Data Stream
// Live stream payload — pose + LLM analysis, 25fps { "frame": 1847, "timestamp_sec": 73.88, "match_period": 2, "players": [ { "id": 9, "team": "blue", "pitch_pos": [52.3, 31.4], "velocity_ms": 4.2, "orientation_deg": 142.5, "body_status": "closed", "open_to_goal": false, "keypoints": { "nose": [52.3, 31.2, 1.74], "left_shoulder": [52.4, 31.1, 1.42], "right_shoulder": [52.2, 31.7, 1.41], "left_elbow": [52.5, 30.9, 1.18], "right_elbow": [52.1, 31.9, 1.17], "left_hip": [52.4, 31.1, 0.96], "right_hip": [52.2, 31.7, 0.95], "left_knee": [52.4, 31.2, 0.52], "right_knee": [52.2, 31.6, 0.51], "left_ankle": [52.3, 31.1, 0.08], "right_ankle": [52.2, 31.7, 0.08] }, "llm_analysis": "Closed body on receipt. Shoulder line 142° — facing away from goal. Consistently limiting forward option selection in this half." } ] }

Request a Demo

We're onboarding a limited number of clubs in the initial access phase. Submit your details and we'll be in touch to arrange a technical demonstration.

No commitment. Response within 48 hours.