Strength, speed, and anthropometric predictors of in-game batting performance in baseball

ABSTRACT

A key focus of sports science research is the identification of quantitative assessments that can predict players’ on-field performance and developmental potential. Despite efforts to establish predictive models, there are few validated measures that show reliable associations and large gaps in understanding. Here, we test a multidimensional battery of assessments developed through the USA Baseball, Prospect Development Pipeline that capture strength and functional movement abilities, and anthropometric characteristics, in a two-year cohort of collegiate baseball players from the Appalachian League. Swing propensity metrics for Zone Contact Percentage (ZCP: proportion pitches in strike zone swung at and hit) and Hard-Hit Percentage (HHP: proportion in-play balls with exit velocity ≥ 95 mph) were calculated on 189 players. Models testing hierarchical combinations of anthropometric and anthropometric plus assessment data were implemented using nested cross-validation with random forest and elastic net regression. Results indicate that anthropometric features account for 29% of variance in ZCP and 50–55% of HHP, while the addition of assessment contributed an additional 1–3% to ZCP and 5–12% to HHP, with top predictors coming from PDP strength and power assessments. These findings delineate contributions of andromorphic and physical abilities to in-game baseball performance using a validated assessment battery and advanced game statistics.

KEYWORDS:

• Batting performance
• machine learning
• assessments
• prediction
• athlete

Acknowledgments

The authors would like to thank USA Baseball and the Prospect Development Pipeline for partnership in this research. Special thanks to Russell Hartford, Jake Barns, Drew Pomeroy, Maria Vargas, Jade Williams, and Manny Portugal for their efforts to collect, manage, share, and discuss these data and their meaning.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The data that support the findings of this study are available from USAB and MLB, but restrictions apply to the availability of these data, which were used under licence for the current study, and so are not publicly available. Data are however available from the authors upon reasonable request with permission of USAB and MLB.

Supplementary material

Supplemental data for this article can be accessed online https://doi.org/10.1080/02640414.2024.2363679.

Additional information

Funding

This research was funded by support to L.G.A. through the Wu-Tsai Human Performance Alliance.