Your running form, decoded.
Turn raw Apple Watch telemetry into personalized, actionable coaching. Real-time conversational AI coach, daily structured workouts, pre-run primers, ACWR readiness, and indoor treadmill intelligence—100% on your device.
Built for runners who value form and privacy.
No generic 180 SPM rules. No mandatory account signups. Every insight is generated locally on your iPhone.
30-Day Relative Baseline
Your cadence, pace, and cardiac efficiency are benchmarked strictly against your personal 30-day rolling history—never arbitrary textbook numbers.
Interactive AI Running Coach
On-device Apple FoundationModels with pure native tool calling. Ask questions about your form, fatigue, or splits with zero math hallucinations.
Structured Runs & Pre-Run Primers
Daily prescribed sessions tailored to your active goal (5K, 10K, Half, Marathon, or Base Building), with pre-run primers and instant Apple Watch WorkoutKit export.
ACWR Readiness & Biological Safety
Acute:Chronic Workload Ratio tracking protects you against overtraining. Automatically adapts workout volume when fatigue is detected.
Take a closer look at the experience.
Intelligent Running Biometrics
Ditch static vanity metrics. Runalyst extracts live cadence deltas, vertical oscillation, and ground contact balance from Apple Watch and benchmarks them against your rolling 30-day relative baseline.
- Zero arbitrary targets: evaluated strictly against your personal physiology
- Sub-second cadence delta tracking against your 30-day average
- Working stats engine that automatically trims dead-stop traffic pauses
The science, math, and logic under the hood.
Most fitness apps compress hours of running into misleading single numbers. Here is how Runalyst analyzes your continuous movement, isolates real-world form, and protects your health.
Automatic 11-Type Workout Classification
Standard running apps flatten entire workouts into simple averages, missing the critical nuance between warmups, hard intervals, and recovery. Runalyst tracks your pace shifts and cadence rhythm in real time to categorize every segment automatically.
- Multi-Resolution Signal Analysis: Evaluates acceleration gradients, peak cadence clusters, elevation correlation, and stop-to-run transition dynamics.
- 11 Distinct Archetypes: Accurately identifies Pyramid Workouts, Hill Repeats, Urban Traffic Runs, Track Intervals, Tempo Surges, Strides, Steady Effort, Long Run, Recovery Run, Progression Run, and Fartlek.
- 98.26% Machine Learning Accuracy: Validated on authentic outdoor continuous telemetry with strict safeguards preventing classification conflicts.
Visual Workout Phases & Apple Watch Export
Executing complex interval drills requires mental preparation and clear pacing geometry before stepping onto the track or road.
- Proportional Phase Timelines: Warmup, surge reps, recovery blocks, and cooldown segments are rendered with exact duration weights.
- 5-Point Mental Preparation Modal: Presents Target Cues, Biomechanical Focus, Phase Breakdown, Strategy, and Effort bands before initiating.
- Dynamic Duration Scaling: Instantly scales active work and recovery intervals across 5m, 10m, and 15m presets, ready for one-tap export to Apple Watch WorkoutKit.
Cardiovascular Guardrails Prevent Overtraining
AI coaching should never compromise runner safety or recommend intense workouts during physical fatigue. Runalyst continuously monitors heart-rate-to-pace decoupling before prescribing drills.
When heart rate climbs while pace slows down (cardiac drift), Runalyst instantly recognizes internal fatigue:
Your health and long-term adaptation always override race goals. You get adaptive, motivating coaching with zero risk of overtraining.
30-Day Time-Locked Longitudinal Baselines
Generic coaching claiming every runner must hold 180 SPM is physiologically incorrect. Optimal cadence and vertical oscillation depend on leg length, height, running economy, and aerobic history.
- Strict Rolling Window: Every biometric split is benchmarked against your personal rolling 30-day baseline on the day you ran.
- No Hindsight Leakage: Historical workouts are evaluated strictly against what your baseline was on that day—never retroactively biased by runs that happened weeks later.
- Coupled Form Diagnostics: Vertical oscillation (>10 cm) is mathematically linked to cadence to detect forward bounding vs. compact turnover.
Smart Traffic Pause Trimming & Field Calibration
Waiting at traffic lights and street crossings corrupts pace and biomechanical averages with resting zeros. Runalyst filters out stationary stops to report your true Working Stats while running.
- Smart Pause Detection: Intelligently detects and trims stationary pauses at red lights to calculate true Working Stats. When running continuously without stops, it preserves pure Apple HealthKit data without distortion.
- Tested on Authentic Outdoor Roads: Calibrated and validated across real-world outdoor road workouts with steep hills, traffic pauses, and variable weather—not sterile synthetic simulations.
- Sensor Noise Resilience: Rigorously stress-tested against real-world GPS jitter, cadence anomalies, and heart rate monitor lag across 137 automated test suites to guarantee zero false alerts.
Why We Focus on Those Metrics
Most fitness trackers report cadence, stride length, and ground contact time as isolated, disconnected metrics. Runalyst connects all three to pinpoint running form: overstride is not a single number—it is the direct interaction of tempo, reach, and impact time.
Low Cadence (Δ < Baseline)
When turnover drops below your personal 30-day baseline, each stride requires higher vertical flight time to sustain pace. This allows the swinging foot to drift and strike significantly ahead of your center of mass.
High Stride Length / Speed
A long stride is only efficient when driven by powerful rear hip extension at race speeds. Artificially reaching forward with a straight knee during aerobic paces generates sharp deceleration shockwaves through the tibia and knee.
Elevated Ground Contact Time
Spring-like running relies on rapid Achilles tendon elastic recoil (<230 ms). Landing in front of the center of mass spikes GCT (>260 ms) as the body stalls in a braking phase rather than immediately rolling forward.
-
No Protractor Needed — How It Feels: You feel weightless and light, like stepping onto a low step right beneath your pelvis. Your knee stays soft like a shock absorber, immediately rolling into effortless forward propulsion.
-
How Apple Watch Confirms It: Rapid Achilles tendon elastic recoil (<230 ms Ground Contact Time), controlled vertical bounce (6.0–9.0 cm), and fluid cadence (170–182 SPM).
-
Forward Energy Transfer: Because ground force directs horizontally through the hip, zero energy is lost fighting deceleration.
-
No Protractor Needed — How It Feels: You feel a sharp deceleration jolt through your heel, knee joint, and lower back with every step. You feel like you are continually stomping the brakes and fighting your own momentum.
-
How Apple Watch Detects It: Prolonged pavement stall (>260 ms Ground Contact Time), excessive vertical bouncing (>10.0 cm), and sluggish turnover (<160 SPM).
-
Braking Collision Shockwave: Landing ahead of the hip converts forward speed into severe joint collision forces, spiking shin splint and patellar tendinitis risk.
Architectural Comparison: Runalyst vs. Standard Fitness Apps
| Feature Dimension | Standard Fitness Apps | Runalyst Native Engine |
|---|---|---|
| Run Classification | Flat averages (misses interval structure and rests) | 11 Automatic Workout Profiles • Dynamic Pace Intervals • Visual Phase Timelines |
| AI Coaching Logic | Generic advice or cloud LLM guesses without physiological safety | Personalized Safety Guardrails • 100% On-Device AI |
| Baseline Standards | Static universal charts (e.g. 180 SPM for all body types) | Personalized 30-Day Moving Window (time-locked to run date) |
| Sensor Noise & Pauses | Traffic stops dilute stride length and cadence | Smart Dead-Stop Trimming with precision HealthKit sync |
| Privacy & Data Pipeline | Health telemetry transmitted to external cloud servers | 100% On-Device • SwiftData SQLite • Apple Silicon local inference |
Ready to run smarter?
Experience private, on-device biometric telemetry and intelligent AI form coaching on your iPhone and Apple Watch.