Methodology
Our scientific approach
Recommendations grounded in science
Pipeline d'analyse
How it works
Where others settle for a questionnaire, we measure. Our analysis pipeline combines your training data, your morphology and biomechanics for truly personalized recommendations.
01
Profile and training data analysis
Connect your Strava account and we analyze your last 30 runs: pace, terrain, elevation, heart rate. Combined with your quiz (weight, shoe size, goals), we build your complete runner profile.
02
Foot scan
Using our 3D scan technology, we measure your plantar arch, your foot width and your pressure zones to determine the shoe shape that suits you best.
03
Video biomechanical analysis
Film your stride with a smartphone. Our AI detects your anatomical points, analyzes your cadence, pronation, oscillation and symmetry for a complete biomechanical profile.
04
Personalized matching
Your profile is matched against our database of 500+ shoe models. A multi-criteria compatibility score guides you to the models suited to your gait, your terrain and your goals.
Biomécanique
What our algorithm measures
Machine learning, biomechanical data and human expertise combined
Cadence & Rhythm
Step frequency and regularity of the stride cycle.
Foot strike type
Classification of heel, midfoot or forefoot strike.
Pronation & Support
Analysis of support dynamics and foot rotation.
Core stability
Oscillation and tilt of the torso while running.
Vertical Oscillation
Amplitude of the vertical bounce with each stride.
Left/right symmetry
Biomechanical balance between the two sides of the body.
Recherche
The expertise of our team
Specialists in biomechanics, podiatry and sports science
Maxime combines expertise in sports biomechanics and data science to develop StrideMatch's recommendation algorithms.
“Every runner is unique. Our mission is to translate that uniqueness into precise shoe recommendations.”
1,885+
Runners in datasets
+500
Shoe models
4
Academic datasets
15+
Biomechanical metrics
Datasets
Scientific datasets used
Running Injury Clinic Kinematic Dataset
2024Ferber R, Osis ST, Hettinga BA, Noehren B
University of Calgary · n=1 798 coureurs
CadenceBlessuresCinématique
Fukuchi Running Biomechanics Dataset
2017Fukuchi RK, Fukuchi CA, Duarte M
Universidade Federal do ABC · n=39 coureurs
GRFCinématique 3DTapis
Van Hooren Running Dataset
2023Van Hooren B, Fuller JT, Miller J, et al.
Maastricht University · n=19 coureurs
TerrainGRFIMU
Publications
Referenced biomechanical research
Increased vertical impact forces and altered running mechanics with softer midsole shoes
2015Baltich J, Maurer C, Nigg BM
Journal of Biomechanics
Biomechanical risk factors associated with running-related injuries: a systematic review
2019Ceyssens L, Vanelderen R, Barton C, et al.
Sports Medicine
Foot strike patterns and collision forces in habitually barefoot versus shod runners
2010Lieberman DE, Venkadesan M, et al.
Nature
A public dataset of running biomechanics and the effects of running speed on lower extremity kinematics and kinetics
2017Fukuchi RK, Fukuchi CA, Duarte M
PeerJ
The effect of running shoes on lower extremity joint torques
2009Kerrigan DC, Franz JR, et al.
PM&R
Machine Learning
Transparency & methodology
We publish our methodology so you can understand how we calculate our recommendations
Open Methodology
We believe in complete transparency. Our matching algorithms are based on peer-reviewed biomechanical research, not marketing claims.
Collaborative Filtering
Our system learns from runners with similar profiles to yours to refine its recommendations.
Feedback Loop
Every user feedback and new Strava run helps improve the overall accuracy of the algorithm.
