Background: The risk of sustaining a subsequent injury is elevated in the weeks after return to play (RTP) from an index injury. However, little is known about the magnitude, duration, and nature by which subsequent injury risk is increased. Purpose: To quantify and describe the risk of injury in a 12-week period after RTP from an index injury in Australian football players. Study Design: Cohort study; Level of evidence, 2. Methods: Injury data were collected from 79 players over 5 years at 1 Australian Football League club. Injuries were classified with the Orchard Sports Injury Classification System and by side of the body. Furthermore, injury severity was classified as time loss (resulting in ≥1 matches being missed) or non–time loss (no matches missed). Subsequent injury was categorized with the SIC-2.0 model and applied to the data set via an automated script. The probability of a time loss subsequent injury was calculated for in-season index injuries for each week of a 12-week period after RTP via a mixed effect logistic regression model. Results: Subsequent injury risk was found to be highest in the week of RTP for both time loss injuries (9.4%) and non–time loss injuries (6.9%). Risk decreased with each week survived after RTP; however, it did not return to baseline risk of participation (3.6%). Conclusion: These findings demonstrate that athletes returning to play are at an increased risk of injury for a number of weeks, thus indicating the requirement for tertiary prevention strategies to ensure that they survive this period.

In Australian football (AF), few studies have assessed combinations of pre- game factors and their relation to game outcomes (win/loss) in multivariable analyses. Further, previous research has mostly been confined to association-based linear approaches and post-game prediction, with limited assessment of predictive machine learning (ML) models in a pre-game setting. Therefore, our aim was to use ML techniques to predict game outcomes and produce a hierarchy of important (win/loss) variables. A total of 152 variables (79 absolute and 73 differentials) were used from the 2013–2018 Australian Football League (AFL) seasons. Various ML models were trained (cross-validation) on the 2013–2017 seasons with the–2018 season used as an independent test set. Model performance varied (66.5-73.3% test set accuracy), although the best model (glmnet – 73.3%) rivalled bookmaker predictions in the same period (70.9%). The glmnet model revealed measures of team quality (a player-based rating and a team-based) in their relative form as the most important variables for prediction. Models that contained in-built feature selection or could model non-linear relationships generally performed better. These findings show that AFL game outcomes can be predicted using ML methods and provide a hierarchy of predictors that maximize the chance of winning.

No abstract

Watts, SP, Binnie, MJ, Goods, PSR, Hewlett, J, Fahey-Gilmour, J, and Peeling, P. Demarcation of intensity from 3 to 5 zones aids in understanding physiological performance progression in highly trained under-23 rowing athletes. J Strength Cond Res XX(X): 000–000, 2023—The purpose of this investigation was to compare 2 training intensity distribution models (3 and 5 zone) in 15 highly trained rowing athletes (n = 8 male; n = 7 female; 19.4 ± 1.1 years) to determine the impact on primary (2,000-m single-scull race) and secondary (2,000-m ergometer time trial, peak oxygen consumption [V̇O2peak], lactate threshold 2 [LT2 power]) performance variables. Performance was assessed before and after 4 months training, which was monitored through a smart watch (Garmin Ltd, Olathe, KS) and chest-strap heart rate (HR) monitor (Wahoo Fitness, Atlanta, GA). Two training intensity distribution models were quantified and compared: a 3-zone model (Z1: between 50% V̇O2peak and lactate threshold 1 (LT1); Z2: between LT1 and 95% LT2; Z3: >95% LT2) and a 5-zone model (T1–T5), where Z1 and Z3 were split into 2 additional zones. There was significant improvement in LT2 power for both male (4.08% ± 1.83, p < 0.01) and female (3.52% ± 3.38, p = 0.02) athletes, with male athletes also demonstrating significant improvement in 2,000-m ergometer time trial (2.3% ± 1.92, p = 0.01). Changes in V̇O2peak significantly correlated with high-quality aerobic training (percent time in T2 zone; r = 0.602, p = 0.02), whereas changes in LT2 power significantly correlated with “threshold” training (percent time in T4 zone; r = 0.529, p = 0.04). These correlations were not evident when examining intensity distribution through the 3-zone model. Accordingly, a 5-zone intensity model may aid in understanding the progression of secondary performance metrics in rowing athletes; however, primary (on-water) performance remains complex to quantify.