Player Tracking Technology and Data for Injury Prevention in the National Football League

AbstractContinuous-time assessments of game outcomes in sports have become increasingly common in the last decade. In American football, only discrete-time estimates of play value were possible, since the most advanced public football datasets were recorded at the play-by-play level. While measures such as expected points and win probability are useful for evaluating football plays and game situations, there has been no research into how these values change throughout the course of a play. In this work, we make two main contributions: First, we introduce a general framework for continuous-time within-play valuation in the National Football League using player-tracking data. Our modular framework incorporates several modular sub-models, to easily incorporate recent work involving player tracking data in football. Second, we use a long short-term memory recurrent neural network to construct a ball-carrier model to estimate how many yards the ball-carrier is expected to gain from their current position, conditional on the locations and trajectories of the ball-carrier, their teammates and opponents. Additionally, we demonstrate an extension with conditional density estimation so that the expectation of any measure of play value can be calculated in continuous-time, which was never before possible at such a granular level.

Download Full-text

Player-Tracking Technology: Half-Full or Half-Empty Glass?

International Journal of Sports Physiology and Performance ◽

10.1123/ijspp.2016-0499 ◽

2017 ◽

Vol 12 (s2) ◽

pp. S2-35-S2-41 ◽

Cited By ~ 48

Author(s):

Martin Buchheit ◽

Ben Michael Simpson

Keyword(s):

Team Sports ◽

Cost Benefit ◽

Ease Of Use ◽

Return To Play ◽

Improve Performance ◽

Multiple Systems ◽

Player Tracking ◽

Load Monitoring ◽

Ongoing Development ◽

Tracking Technology

With the ongoing development of microtechnology, player tracking has become one of the most important components of load monitoring in team sports. The 3 main objectives of player tracking are better understanding of practice (provide an objective, a posteriori evaluation of external load and locomotor demands of any given session or match), optimization of training-load patterns at the team level, and decision making on individual players’ training programs to improve performance and prevent injuries (eg, top-up training vs unloading sequences, return to play progression). This paper discusses the basics of a simple tracking approach and the need to integrate multiple systems. The limitations of some of the most used variables in the field (including metabolic-power measures) are debated, and innovative and potentially new powerful variables are presented. The foundations of a successful player-monitoring system are probably laid on the pitch first, in the way practitioners collect their own tracking data, given the limitations of each variable, and how they report and use all this information, rather than in the technology and the variables per se. Overall, the decision to use any tracking technology or new variable should always be considered with a cost/benefit approach (ie, cost, ease of use, portability, manpower/ability to affect the training program).

Download Full-text

Video Analysis of Reported Concussion Events in the National Football League During the 2015-2016 and 2016-2017 Seasons

The American Journal of Sports Medicine ◽

10.1177/0363546518804498 ◽

2018 ◽

Vol 46 (14) ◽

pp. 3502-3510 ◽

Cited By ~ 20

Author(s):

David J. Lessley ◽

Richard W. Kent ◽

James R. Funk ◽

Christopher P. Sherwood ◽

Joseph M. Cormier ◽

...

Keyword(s):

Injury Prevention ◽

National Football League ◽

Sports Injury ◽

The Body ◽

American Football ◽

Video Footage ◽

Impact Location ◽

High Prevalence ◽

Common Activity ◽

Rule Changes

Background: Concussions in American football remain a high priority of sports injury prevention programs. Detailed video review provides important information on causation, the outcomes of rule changes, and guidance on future injury prevention strategies. Purpose: Documentation of concussions sustained in National Football League games played during the 2015-2016 and 2016-2017 seasons, including consideration of video views unavailable to the public. Study Design: Descriptive epidemiology study. Methods: All reported concussions were reviewed with all available video footage. Standardized terminology and associated definitions were developed to describe and categorize the details of each concussion. Results: Cornerbacks sustained the most concussions, followed by wide receivers, then linebackers and offensive linemen. Half (50%) of concussions occurred during a passing play, 28% during a rushing play, and 21% on a punt or kickoff. Tackling was found to be the most common activity of concussed players, with the side of the helmet the most common helmet impact location. The distribution of helmet impact source—the object that contacted the concussed player’s helmet—differed from studies of earlier seasons, with a higher proportion of helmet-to-body impacts (particularly shoulder) and helmet-to-ground impacts and with a lower proportion of helmet-to-helmet impacts. Helmet-to-ground concussive impacts were notable for the high prevalence of impacts to the back of the helmet and their frequency during passing plays. Conclusion: Concussion causation scenarios in the National Football League have changed over time. Clinical Relevance: The results of this study suggest the need for expanded evaluation of concussion countermeasures beyond solely helmet-to-helmet test systems, including consideration of impacts with the ground and with the body of the opposing player. It also suggests the possibility of position-specific countermeasures as part of an ongoing effort to improve safety.

Download Full-text

Bigger data, better questions, and a return to fourth down behavior: an introduction to a special issue on tracking datain the National football League

Journal of Quantitative Analysis in Sports ◽

10.1515/jqas-2020-0057 ◽

2020 ◽

Vol 16 (2) ◽

pp. 73-79

Author(s):

Michael J. Lopez

Keyword(s):

Quantitative Analysis ◽

Best Practices ◽

National Football League ◽

Tracking Data ◽

Omitted Variables ◽

Special Issue ◽

Player Tracking ◽

Level Data ◽

Game Situation ◽

To Come

AbstractMost historical National Football League (NFL) analysis, both mainstream and academic, has relied on public, play-level data to generate team and player comparisons. Given the number of oft omitted variables that impact on-field results, such as play call, game situation, and opponent strength, findings tend to be more anecdotal than actionable. With the release of player tracking data, however, analysts can better ask and answer questions to isolate skill and strategy. In this article, we highlight the limitations of traditional analyses, and use a decades-old punching bag for analysts, fourth-down strategy, as a microcosm for why tracking data is needed. Specifically, we assert that, in absence of using the precise yardage needed for a first down, past findings supporting an aggressive fourth down strategy may have been overstated. Next, we synthesize recent work that comprises this special Journal of Quantitative Analysis in Sports issue into player tracking data in football. Finally, we conclude with some best practices and limitations regarding usage of this data. The release of player tracking data marks a transition for the league and its’ analysts, and we hope this issue helps guide innovation in football analytics for years to come.

Download Full-text

Expected hypothetical completion probability

Journal of Quantitative Analysis in Sports ◽

10.1515/jqas-2019-0050 ◽

2020 ◽

Vol 16 (2) ◽

pp. 85-94

Author(s):

Sameer K. Deshpande ◽

Katherine Evans

Keyword(s):

Big Data ◽

High Resolution ◽

National Football League ◽

Probability Model ◽

Tracking Data ◽

Complex Interactions ◽

Player Tracking ◽

Model Predictions ◽

Fundamental Difficulty ◽

Non Parametric

AbstractUsing high-resolution player tracking data made available by the National Football League (NFL) for their 2019 Big Data Bowl competition, we introduce the Expected Hypothetical Completion Probability (EHCP), a objective framework for evaluating plays. At the heart of EHCP is the question “on a given passing play, did the quarterback throw the pass to the receiver who was most likely to catch it?” To answer this question, we first built a Bayesian non-parametric catch probability model that automatically accounts for complex interactions between inputs like the receiver’s speed and distances to the ball and nearest defender. While building such a model is, in principle, straightforward, using it to reason about a hypothetical pass is challenging because many of the model inputs corresponding to a hypothetical are necessarily unobserved. To wit, it is impossible to observe how close an un-targeted receiver would be to his nearest defender had the pass been thrown to him instead of the receiver who was actually targeted. To overcome this fundamental difficulty, we propose imputing the unobservable inputs and averaging our model predictions across these imputations to derive EHCP. In this way, EHCP can track how the completion probability evolves for each receiver over the course of a play in a way that accounts for the uncertainty about missing inputs.

Download Full-text