National Football League: 2018–2019 Season

The interdisciplinary nature of sports and the presence of various systemic and non-systemic factors introduce challenges in predicting sports match outcomes using a single disciplinary approach. In contrast to previous studies that use sports performance metrics and statistical models, this study is the first to apply a deep learning approach in financial time series modeling to predict sports match outcomes. The proposed approach has two main components: a convolutional neural network (CNN) classifier for implicit pattern recognition and a logistic regression model for match outcome judgment. First, the raw data used in the prediction are derived from the betting market odds and actual scores of each game, which are transformed into sports candlesticks. Second, CNN is used to classify the candlesticks time series on a graphical basis. To this end, the original 1D time series are encoded into 2D matrix images using Gramian angular field and are then fed into the CNN classifier. In this way, the winning probability of each matchup team can be derived based on historically implied behavioral patterns. Third, to further consider the differences between strong and weak teams, the CNN classifier adjusts the probability of winning the match by using the logistic regression model and then makes a final judgment regarding the match outcome. We empirically test this approach using 18,944 National Football League game data spanning 32 years and find that using the individual historical data of each team in the CNN classifier for pattern recognition is better than using the data of all teams. The CNN in conjunction with the logistic regression judgment model outperforms the CNN in conjunction with SVM, Naïve Bayes, Adaboost, J48, and random forest, and its accuracy surpasses that of betting market prediction.

Download Full-text

Concussion in National Football League Athletes Is Not Associated With Increased Risk of Acute, Noncontact Lower Extremity Musculoskeletal Injury

Orthopaedic Journal of Sports Medicine ◽

10.1177/23259671211003491 ◽

2021 ◽

Vol 9 (5) ◽

pp. 232596712110034

Author(s):

Toufic R. Jildeh ◽

Fabien Meta ◽

Jacob Young ◽

Brendan Page ◽

Kelechi R. Okoroha

Keyword(s):

Lower Extremity ◽

National Football League ◽

Field Performance ◽

Return To Play ◽

Lower Extremity Injury ◽

Extremity Injury ◽

Lower Extremity Injuries ◽

Increased Risk ◽

Extremity Injuries ◽

And Control

Background: Impaired neuromuscular function after concussion has recently been linked to increased risk of lower extremity injuries in athletes. Purpose: To determine if National Football League (NFL) athletes have an increased risk of sustaining an acute, noncontact lower extremity injury in the 90-day period after return to play (RTP) and whether on-field performance differs pre- and postconcussion. Study Design: Cohort study, Level of evidence, 3. Methods: NFL concussions in offensive players from the 2012-2013 to the 2016-2017 seasons were studied. Age, position, injury location/type, RTP, and athlete factors were noted. A 90-day RTP postconcussive period was analyzed for lower extremity injuries. Concussion and injury data were obtained from publicly available sources. Nonconcussed, offensive skill position NFL athletes from the same period were used as a control cohort, with the 2014 season as the reference season. Power rating performance metrics were calculated for ±1, ±2, and ±3 seasons pre- and postconcussion. Conditional logistic regression was used to determine associations between concussion and lower extremity injury as well as the relationship of concussions to on-field performance. Results: In total, 116 concussions were recorded in 108 NFL athletes during the study period. There was no statistically significant difference in the incidence of an acute, noncontact lower extremity injury between concussed and control athletes (8.5% vs 12.8%; P = .143), which correlates with an odds ratio of 0.573 (95% CI, 0.270-1.217). Days (66.4 ± 81.9 days vs 45.1 ± 69.2 days; P = .423) and games missed (3.67 ± 3.0 vs 2.9 ± 2.7 games; P = .470) were similar in concussed athletes and control athletes after a lower extremity injury. No significant changes in power ratings were noted in concussed athletes in the acute period (±1 season to injury) when comparing pre- and postconcussion. Conclusion: Concussed, NFL offensive athletes did not demonstrate increased odds of acute, noncontact, lower extremity injury in a 90-day RTP period when compared with nonconcussed controls. Immediate on-field performance of skill position players did not appear to be affected by concussion.

Download Full-text

Editorial Commentary: Operative Versus Nonoperative Management of Shoulder Instability in the National Football League Athlete: Do What Needs to Be Done—Treatment Choice Does Not Affect Future Performance or Games Played

Arthroscopy The Journal of Arthroscopic and Related Surgery ◽

10.1016/j.arthro.2021.01.053 ◽

2021 ◽

Vol 37 (5) ◽

pp. 1446-1448

Author(s):

Timothy R. McAdams

Keyword(s):

Shoulder Instability ◽

National Football League ◽

Nonoperative Management ◽

Treatment Choice ◽

Editorial Commentary ◽

Future Performance

Download Full-text

CAREER LENGTH AND LEFT VENTRICULAR DIASTOLIC FUNCTION AMONG RETIRED NATIONAL FOOTBALL LEAGUE PLAYERS

Journal of the American College of Cardiology ◽

10.1016/s0735-1097(21)04566-6 ◽

2021 ◽

Vol 77 (18) ◽

pp. 3211

Author(s):

Camilo Fernandez - Alonso ◽

G. Elizabeth Smith ◽

Alexander Razavi ◽

Gregory Stewart

Keyword(s):

Diastolic Function ◽

National Football League ◽

Left Ventricular Diastolic Function ◽

Left Ventricular ◽

Ventricular Diastolic Function ◽

Career Length

Download Full-text

Prediction of sports attendance: A comparative analysis

Proceedings of the Institution of Mechanical Engineers Part P Journal of Sports Engineering and Technology ◽

10.1177/1754337120983135 ◽

2020 ◽

pp. 175433712098313

Author(s):

Mehmet Şahin ◽

Murat Uçar

Keyword(s):

Neural Network ◽

Machine Learning ◽

Comparative Analysis ◽

National Football League ◽

Relevant Literature ◽

Performance Evaluations ◽

Influential Factor ◽

Gradient Boosting ◽

Home Team ◽

Machine Learning Methods

In this study, a comparative analysis for predicting sports attendance demand is presented based on econometric, artificial intelligence, and machine learning methodologies. Data from more than 20,000 games from three major leagues, namely the National Basketball Association (NBA), National Football League (NFL), and Major League Baseball (MLB), were used for training and testing the approaches. The relevant literature was examined to determine the most useful variables as potential regressors in forecasting. To reveal the most effective approach, three scenarios containing seven cases were constructed. In the first scenario, each league was evaluated separately. In the second scenario, the three possible combinations of league pairings were evaluated, while in the third scenario, all three leagues were evaluated together. The performance evaluations of the results suggest that one of the machine learning methods, Gradient Boosting, outperformed the other methods used. However, the Artificial Neural Network, deep Convolutional Neural Network, and Decision Trees also provided productive and competitive predictions for sports games. Based on the results, the predictions for the NBA and NFL leagues are more satisfactory than the predictions of the MLB, which may be caused by the structure of the MLB. The results of the sensitivity analysis indicate that the performance of the home team is the most influential factor for all three leagues.

Download Full-text