Abstract
This paper details a route classification method for American football using a template matching scheme that is quick and does not require manual labeling. Pre-defined routes from a standard receiver route tree are aligned closely with game routes in order to determine the closest match. Based on a test game with manually labeled routes, the method achieves moderate success with an overall accuracy of 72% of the 232 routes labeled correctly.
Appendix
Examples of game routes that were labeled the same in the Denver Broncos versus Los Angeles Chargers game during the 2017 season. The magenta line represents the median route of the group showing this method is able to partition essential qualities of common routes.
References
Chu, D., M. Reyers, J. Thomson, and L. Wu. 2019. “Route Identification in the National Football League.” Journal of Quantitative Analysis in Sports, arXiv preprint arXiv:1908.02423.10.1515/jqas-2019-0047Search in Google Scholar
Hochstedler, J. and P. T. Gagnon. 2017. “American Football Route Identification Using Supervised Machine Learning.” in Proceedings of the 2017 MIT Sloan Sports Analytics Conference.Search in Google Scholar
Intille, S. S. and A. F. Bobick. 1999. “A Framework for Recognizing Multi-Agent Action from Visual Evidence.” Pp. 518–525 in AAAI/IAAI 99.Search in Google Scholar
Sterken, N. 2019. “RouteNet: A Convolutional Nueral Network for Classifying Routes.” in NFL Big Data Bowl. https://operations.nfl.com/media/3671/big-data-bowl-sterken.pdf.Search in Google Scholar
Quionero-Candela, J., M. Sugiyama, A. Schwaighofer, and N. D. Lawrence. 2009. Dataset Shift in Machine Learning. Cambridge, MA, USA: The MIT Press.10.7551/mitpress/9780262170055.001.0001Search in Google Scholar
Vonder Haar, A. 2019. “Exploratoy Data Analysis of Passing Plays using NFL Tracking Data.” in NFL Big Data bowl. https://operations.nfl.com/media/3672/big-data-bowl-vonder-haar.pdf.Search in Google Scholar
© 2020 Walter de Gruyter GmbH, Berlin/Boston