ML models for baseball pitch speed prediction show poor cross-individual generalizability, with R-squared falling from 0.91 within individuals to 0.38 across individuals, and trunk/pivot leg kinematics providing the strongest transfer.
That is, if skilled athletes achieve higher performance than others, despite similar motion inputs, ML models may consistently underestimate expert performance
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.HC 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Cross-individual generalizability of machine learning models for ball speed prediction in baseball pitching
ML models for baseball pitch speed prediction show poor cross-individual generalizability, with R-squared falling from 0.91 within individuals to 0.38 across individuals, and trunk/pivot leg kinematics providing the strongest transfer.