Research notes
What race analysis adds to a finishing time
A closer look at our 2025 study of men’s 100 m freestyle.
Looking within the race
Our study analysed 204 male swimmers in long-course competition from 2019 to 2024. Race video provided measures of stroke characteristics, underwater phases and velocity across short race segments.
The random forest model classified performance levels with 89.5% accuracy. Velocities from 65–70 m and 70–75 m were the strongest differentiating features.
Interpreting the finding
This is an association within the analysed sample. It does not establish that changing one race segment will cause a swimmer to reach a higher performance level. Nor should the result automatically be applied to other strokes, distances or female swimmers.
Read the source
Postiglione and colleagues (2025). International Journal of Sports Physiology and Performance. DOI: 10.1123/ijspp.2024-0486.