SSMT Sports Science Meets Tech

What's real in sports tech right now

Signals — what's emerging, with the one question we're watching.   Grades — what we've actually assessed against the evidence.

Draft — a validation surface, nothing here is published. Signals are forward-looking and lighter than grades; a signal is never a verdict.
🔬 Lab-stage Computer Vision
A single phone camera is catching up to a motion-capture lab
Markerless pose from one ordinary camera hit 5.5° joint-angle accuracy — statistically tied with a two-camera rig.
The watch Whether it survives outside a lab floor, in real patients and normal clothing. Rotations are still unusable — off by 10–20°.
We graded this → SSMT-2026-0006
The claim on the box Wearables & Recovery
“Accurate heart rate, right from your wrist.”
Partially supported drift 3/7
Accurate for steady cardio in healthy, light-skinned adults — not accurate, full stop.
Full evidence → SSMT-2026-0001
📈 Heating up Computer Vision
Ordinary match footage is turning into 3D tactical data
AI is extracting 3D player tracking from normal broadcast video — no stadium hardware required.
The watch Accuracy against stadium tracking systems has not been independently established. Treat the 3D as an estimate, not ground truth.
Not yet graded
The claim on the box Wearables & Recovery
“See your deep, REM and light sleep every night.”
Partially supported drift 5/7
Directionally useful in aggregate, unreliable for any single night — and the accuracy that is well established is a different measurement from the one being sold.
Full evidence → SSMT-2026-0002
🌱 Just appeared AI Athletes
AI is learning to run a simulated human body, muscle by muscle
Open benchmarks now train agents to control a musculoskeletal model on athletic tasks.
The watch It's a simulation milestone, not a training tool. There is zero demonstrated transfer to real athletes yet.
MyoSuite / MyoChallenge ↗
The claim on the box Wearables & Recovery
“Train when your body is ready.”
Partially supported drift 4/7
The supported finding is an efficiency one, not a superiority one.
Full evidence → SSMT-2026-0003
📈 Heating up Wearables & Recovery
Your sweat is becoming a live readout
Skin-worn microfluidic patches read hydration and electrolyte signals from sweat during exercise.
The watch The sensing is real. Whether acting on the numbers beats drinking to a sensible plan is untested — that's the gap between a measurement and a benefit.
Microfluidic epidermal sweat sensing — Nature Biotechnology review literature ↗
The claim on the box Wearables & Recovery
“Your readiness score tells you when to train hard and when to back off.”
Unevaluated drift 7/7
Never tested, in any design. Currently unevaluable by independent parties.
Full evidence → SSMT-2026-0004
🔬 Lab-stage Wearables & Recovery
Clothes that read your muscles
Strain-sensing textiles can track muscle activation and posture as you move — clothing as a sensor network.
The watch Lab demonstrations only. Durability, washing, and validity in real training are open questions.
npj Flexible Electronics — textile strain-sensing literature ↗
The claim on the box Computer Vision
“Lab-grade biomechanics from your phone.”
Partially supported drift 4/7
Genuinely good in the sagittal plane; not usable for rotations; and 'from your phone' is the least-evidenced configuration.
Full evidence → SSMT-2026-0006
👀 On our grading radar Wearables & Recovery
Cold plunges are everywhere — and the evidence is genuinely split
Cold-water immersion clearly helps you feel recovered, but may blunt some strength and muscle adaptations depending on timing.
The watch We're grading this next. The open question is whether feeling better costs you adaptation — and when in the week it matters.
Not yet graded
The claim on the box Biomechanics
“Know your injury risk before it happens. (measurement component)”
Partially supported drift 3/7
Force magnitude yes; the asymmetry number itself does not reproduce.
Full evidence → SSMT-2026-0007
🌱 Just appeared Biomechanics
A basketball that grades your jump shot
Sensor-embedded balls now report shot arc, spin and release consistency in real time.
The watch The measurement is a genuine novelty. Whether the feedback actually improves shooting has not been studied.
Not yet graded
The claim on the box Injury Science
“Know your injury risk before it happens.”
Contradicted drift 5/7
The best available evidence points the other way.
Full evidence → SSMT-2026-0008
👀 On our grading radar Wearables & Recovery
Glucose monitors jumped from diabetes to endurance sport
CGMs are now marketed to healthy athletes as real-time fuelling tools.
The watch We already graded this — Unevaluated. During exercise they read 14–34% off, and no trial shows that steering fuelling by the sensor makes you faster.
We graded this → SSMT-2026-0009
The claim on the box Wearables & Recovery
“Fuel smarter. See how your body responds in real time.”
Unevaluated drift 4/7
Bordering Partially supported for the narrower claim that CGM tracks glucose trends during exercise.
Full evidence → SSMT-2026-0009
The claim on the box Biomechanics
“Know exactly how fast you lifted.”
Supported drift 3/7
Supported for linear transducers and camera devices. Contradicted for IMU/accelerometer devices.
Full evidence → SSMT-2026-0010
The claim on the box Biomechanics
“Autoregulate every set.”
Partially supported drift 3/7
Equal on strength; small real gains on jump, sprint and change of direction, often at lower volume.
Full evidence → SSMT-2026-0011
The claim on the box Wearables & Recovery
“Recover faster. Train harder tomorrow.”
Partially supported drift 4/7
Accurate about how you feel; unsupported about what you can then do.
Full evidence → SSMT-2026-0012