AI Running Coaching FAQ

Running AI coaching from a smartphone camera measures: cadence (steps per minute), stride length, foot strike pattern (heel, midfoot, or forefoot), vertical oscillation (how much you bounce), hip drop (Trendelenburg sign), and forward lean angle. The most common running fault AI detects is overstriding — landing with the foot in front of the center of mass, which increases braking force and injury risk.

The five running patterns most associated with injury that AI can detect are: overstriding (foot landing in front of center of mass), hip drop greater than 10 degrees (indicates hip abductor weakness), asymmetrical cadence between left and right legs (>5% difference), excessive forward trunk lean (>15 degrees), and knee valgus during stance phase. Each of these patterns is measurable from a side-on or rear-facing smartphone camera.

Running cadence AI analysis counts foot strikes per minute from video and calculates the left-right symmetry of cadence. A cadence below 160 spm typically indicates overstriding. AI coaching provides a target cadence range and tracks improvement over time. The most effective method for increasing cadence is using a metronome during runs at the target cadence while AI verifies the improvement in form.

Optimal running arm swing has: elbows at approximately 90 degrees, arms swinging forward and back (not across the body), hands relaxed (not clenched), and symmetrical swing amplitude between left and right arms. Cross-body arm swing is the most common fault and is often caused by shoulder tightness. AI measures arm swing direction angle and flags any swing that crosses the body's midline.

Hill running AI analysis requires a side-facing camera to capture lean angle and stride mechanics. For uphill running, AI measures: forward lean angle (should increase proportionally with gradient), stride length (should shorten on steeper grades), foot strike position (should shift toward midfoot), and arm drive intensity (should increase to maintain momentum). For downhill, AI measures: braking mechanics, foot strike position (midfoot preferred over heel strike), and trunk lean.

Marathon AI coaching is most valuable for identifying the specific form breakdown pattern that occurs at the runner's individual fatigue threshold. Some runners lose forward lean first; others develop cross-body arm swing; others drop cadence. AI coaching identifies the individual pattern and prescribes targeted strength and drill work to delay that breakdown point in race conditions.

SportsReflector's AI uses advanced computer vision to turn your phone into a portable running lab. After you upload a video, our proprietary pose estimation algorithm, trained on extensive running data, tracks 17 key body points in each frame. This enables precise measurement of dynamic biomechanical parameters. The AI calculates crucial joint angles, such as knee flexion at footstrike (ideally 15-20 degrees) and hip extension at toe-off (around 10-15 degrees), to assess movement efficiency. It also analyzes temporal metrics with millisecond accuracy, including ground contact time (elite runners often under 200ms) and flight time. By comparing your data against an ideal biomechanical model and a vast database, the AI provides a comprehensive form evaluation, highlighting areas for improvement. This detailed analysis, previously exclusive to expensive labs, is now accessible to all runners.

SportsReflector's AI, using sophisticated computer vision and machine learning, pinpoints prevalent running form errors that lead to inefficiency and injury. A frequent issue is **overstriding**, where the foot lands too far forward, acting as a braking mechanism. This often correlates with a **low cadence**, typically below 160 steps per minute, increasing impact and joint stress. Another common fault is **excessive vertical oscillation**, meaning too much 'bounce,' wasting energy; optimal vertical oscillation is often 6-10 cm. The AI also identifies **improper foot strike patterns**, such as a heavy heel strike causing shockwaves, or an overly aggressive forefoot strike leading to calf strain. Additionally, **poor trunk posture** (e.g., leaning too far forward/backward) and **cross-over gait** (feet landing too close to the midline) are flagged, as they disrupt balance and increase injury risk. SportsReflector provides visual and numerical feedback to correct these biomechanical deviations.

SportsReflector's AI coaching offers distinct advantages, primarily in providing **unbiased, quantitative biomechanical analysis** with precision. While human coaches offer qualitative feedback, AI measures joint angles, ground contact time to the millisecond, and vertical oscillation with exact centimeter values. This data-driven approach gives highly specific feedback, identifying subtle form inefficiencies that even experienced human eyes might miss. For instance, AI can detect a 5% increase in pronation or a 10ms change in stride duration, crucial for injury prevention and performance. Human coaches, however, excel in **emotional support, motivational strategies, and adapting to complex, non-quantifiable factors** like stress or sleep. They build rapport, offer empathy, and provide a holistic training experience. SportsReflector's AI acts as a powerful tool augmenting human coaching, supplying objective data for deep mechanical understanding, allowing human coaches to focus on psychological and broader physiological aspects. It democratizes access to high-level biomechanical analysis, making it widely available.

SportsReflector's AI uses advanced computer vision to track a wide array of metrics for in-depth running form analysis. Key **temporal metrics** include **cadence** (steps per minute), typically 170-180 SPM for optimal efficiency, and **ground contact time** (duration your foot spends on the ground), with elite runners often under 200 milliseconds. **Flight time** (both feet off ground) indicates elasticity and propulsion. **Spatial metrics** encompass **stride length**, **vertical oscillation** (vertical movement of center of mass, ideally 6-10 cm), and **stride ratio** (stride length divided by vertical oscillation, a key efficiency indicator). The AI also precisely measures **joint angles** at critical gait phases, such as knee flexion at initial contact (optimal 15-20 degrees) and hip extension at toe-off (around 10-15 degrees). It monitors **pronation/supination** of the foot and **pelvic drop**, vital for stability and injury risk. These metrics are continuously analyzed against personalized baselines and optimal ranges, providing actionable feedback to fine-tune technique for enhanced performance and reduced injury susceptibility.