01
Wearable Telemetry Units
Body-worn sensor pods with 1000 Hz inertial capture, dual-band GNSS and an indoor local positioning fallback.
Module detail — Wearable Telemetry UnitsThe platform
One measurement chain, six modules you can buy separately
Most performance departments do not need everything at once. They need one part of the chain fixed — usually the sensor layer or the analysis layer — without inheriting a decade of vendor lock-in on the rest. Every module below shares an athlete index, a time base and an export format, and every one of them can be the only thing you buy.
Modules
What each part does
Hardware first, then the models that consume it, then the places the output has to travel to. Follow any link for the engineering detail, the limitations we know about, and what a deployment involves.
02
Training Load & Readiness
External and internal load in one model, with RMSSD-based readiness and honest uncertainty bands.
Module detail — Training Load & Readiness
03
Markerless Motion Capture
Multi-camera pose estimation for joint kinematics without suits, markers or a laboratory booking.
Module detail — Markerless Motion Capture
04
Injury Risk Modelling
Survival models over exposure history that surface changing risk profiles instead of a single traffic light.
Module detail — Injury Risk Modelling
05
Team Analytics Suite
The browser workspace where squad load, availability and session planning live in one reviewable place.
Module detail — Team Analytics Suite
06
Live Broadcast Data Feeds
Sub-second, rights-controlled telemetry out to broadcast graphics, second-screen apps and league feeds.
Module detail — Live Broadcast Data FeedsPrinciples
Four commitments that shape the product
These are not marketing lines; they are the constraints we design against, and the reason some obvious features are missing.
Reproducible metrics
Every derived value carries the algorithm version that produced it. When we change a model, historical sessions are not silently rewritten — you can recompute deliberately and compare the two.
Uncertainty stays visible
Where a figure has meaningful measurement error, the interface shows a range. A single tidy number that hides its own noise is the fastest route to a bad selection decision.
Open export, always
CSV, FIT, JSON and documented raw binary. If you cancel, you keep the archive in formats that other tools can read. Data portability is contractual, not discretionary.
Athlete-first governance
Role-scoped access, regional data residency, per-field consent for anything that leaves the club, and an audit log the athlete's representative can request.
Getting started
A first deployment, week by week
Four to six weeks from signature to handover for a single senior squad. Multi-squad academies and multi-venue clubs take longer, mostly because of the indoor positioning survey.
- Scoping. Squad size, venues, existing athlete management system, who owns data governance, and which of your current metrics have to survive the transition unchanged. We would rather find the awkward requirement now than in week five.
- Hardware fit. Harness sizing across the squad, pod assignment, dock placement and network configuration. If you need indoor positioning we run an ultra-wideband site survey and specify anchor positions per hall.
- Parallel running. Two to three weeks of capturing identical sessions on both your current system and ours, so your staff can see where the numbers agree, where they diverge, and why. Divergence is normal — unexplained divergence is not.
- Integration. Connecting the athlete management system, agreeing the export schedule, configuring role-based access for coaching, S&C and medical staff separately.
- Handover. Staff training on the workspace, a written data-governance pack you can give to athletes and their representatives, and a named support contact for the season.
What we will tell you before you sign
If your primary use case is something our models handle weakly — for example single-sensor estimation of joint loading, or sleep staging accurate enough for a clinical decision — we would rather say so at the scoping call than defend it in month four. There is a list of these in each module page under “What this does not do”.
Background reading
The thinking behind the models
What HRV Actually Tells You About Readiness
RMSSD is a genuinely useful signal and a badly abused one. How to read heart-rate variability without over-claiming.
Read the note
Detecting Sprint Asymmetry From IMU Data
What a single trunk-mounted inertial unit can and cannot say about left-right differences in maximal running.
Read the note
Why Sensor Placement Decides Your Data Quality
Soft-tissue artefact, harness tension and pod orientation cost you more accuracy than any algorithm gains back.
Read the noteTell us which part of the chain is broken.
Most conversations start with one specific frustration — an export that will not reconcile, an indoor venue with no positioning, a readiness score nobody trusts. Start there.