Australian Football League Season Research Articles

The effect of assessor expertise on reliability of analysis of video signs associated with concussion in Australian football

ObjectivesTo determine whether spotters with medical training and experience in managing concussion have higher inter-rater reliability and accuracy than non-medical personnel when identifying video signs associated with concussion in Australian football. DesignRetrospective cohort study. MethodsVideo clips were collected of all impacts potentially resulting in concussion during 2012 and 2013 Australian Football League (AFL) seasons. Raters were divided into medical doctors and a non-medical group comprising allied health practitioners (physiotherapists) and non-medical/non-allied health personnel (performance analysts). Raters assessed 102 randomly selected videos for signs of concussion. The inter-rater reliability was calculated. Sensitivity, specificity, positive and negative predictive values were calculated by comparing the rater responses to the consensus opinion from two highly experienced clinicians with expertise in concussion. ResultsNo statistically significant difference in inter-rater reliability was observed between the medical doctors and the non-medical group. Both groups demonstrated good to excellent agreement for slow to get up, clutching at head/face and facial injury. Both groups displayed intra-class coefficient >0.55 for no protective action-floppy, loss of responsiveness, and motor incoordination, and displayed lowest agreement for no protective action-tonic posturing, impact seizure and blank/vacant look. No statistically significant difference was found between the groups for sensitivity, specificity, positive and negative predictive values for correctly classifying video signs compared to the expert consensus opinion. ConclusionsAfter completing sufficient standardised training and testing, medical and non-medical personnel demonstrate comparable reliability in identifying video signs of concussion in professional Australian football and may be suitable for the role of video spotter.

The Eye in the Sky—A Method to Obtain On-Field Locations of Australian Rules Football Athletes

The ability to overcome an opposition in team sports is reliant upon an understanding of the tactical behaviour of the opposing team members. Recent research is limited to a performance analysts’ own playing team members, as the required opposing team athletes’ geolocation (GPS) data are unavailable. However, in professional Australian rules Football (AF), animations of athlete GPS data from all teams are commercially available. The purpose of this technical study was to obtain the on-field location of AF athletes from animations of the 2019 Australian Football League season to enable the examination of the tactical behaviour of any team. The pre-trained object detection model YOLOv4 was fine-tuned to detect players, and a custom convolutional neural network was trained to track numbers in the animations. The object detection and the athlete tracking achieved an accuracy of 0.94 and 0.98, respectively. Subsequent scaling and translation coefficients were determined through solving an optimisation problem to transform the pixel coordinate positions of a tracked player number to field-relative Cartesian coordinates. The derived equations achieved an average Euclidean distance from the athletes’ raw GPS data of 2.63 m. The proposed athlete detection and tracking approach is a novel methodology to obtain the on-field positions of AF athletes in the absence of direct measures, which may be used for the analysis of opposition collective team behaviour and in the development of interactive play sketching AF tools.

Discovery and Characterisation of Forward Line Formations at Centre Bounces in the Australian Football League.

The extent of player formation usage and the characteristics of player arrangements are not well understood in Australian football, unlike other team-based invasion sports. Using player location data from all centre bounces in the 2021 Australian Football League season; this study described the spatial characteristics and roles of players in the forward line. Summary metrics indicated that teams differed in how spread out their forward players were (deviation away from the goal-to-goal axis and convex hull area) but were similar with regard to the centroid of player locations. Cluster analysis, along with visual inspection of player densities, clearly showed the presence of different repeated structures or formations used by teams. Teams also differed in their choice of player role combinations in forward lines at centre bounces. New terminology was proposed to describe the characteristics of forward line formations used in professional Australian Football.

Relationships between contract status and player performance in the Australian Football League

ABSTRACT This study analysed the extent to which player performance differs within the Australian Football League (AFL) with respect to the status of a player’s contract. AFL Player Ratings (AFLPR) and contract data were obtained during the 2013–2020 AFL seasons for all 827 players listed by an AFL club at the beginning of the 2020 season. A model of “expected performance” was created allowing for an exploration into the differential with actual performance as a function of contract status. Paired t-tests indicated that there was a difference in performance pre- and post-signing their contract for players who signed mid-season (mean change and 95% confidence interval of −1.48 ± 0.93 and −0.49 ± 0.48 AFLPR, at ten match intervals for those in- and out-of-contract at the conclusion of that year’s season, respectively). Further differences existed between the groups of players who signed mid-season, as compared to those who signed during the off-season. Correlation analyses indicated that more consistent performers are somewhat less likely to see a reduction in performance post signing as compared to less consistent performers. The applications of these findings have the potential to support organisational decisions relating to the timing and nature of player contracting.

Post-game recovery of isometric hamstring strength in a high-risk hamstring strain injury group throughout an Australian Football League season

ABSTRACT This study sought to investigate post-game hamstring strength recovery of 26 Australian Football League (AFL) players with a previous hamstring strain injury (HSI) across an AFL season. Maximal unilateral isometric knee flexion strength was assessed using an externally fixed dynamometer, and inter-session reliability was measured during the pre-season period. Linear mixed effects models investigated the influence of numerous variables on post-game hamstring strength decrement (relative change between initial weekly test and individual baseline) and individual within-week strength change following gameplay. The test demonstrated good inter-tester reliability (ICC = 0.81–0.88; CV = 6.73–7.33), and an acceptable level of error (MAE = 5.77–7.14%). Player as a random effect strongly influenced post-game strength decrement and within-week strength change (marginal R2 = 0.185–0.407; conditional R2 = 0.455–0.654). Within-week hamstring strength change was strongly determined by post-game strength decrement alone (estimate = 0.51, 95% CI = -0.66- -0.36 ; η2 = 0.32; P=<0.001) and in interaction with number of days post-game (estimate = 0.43, 95% CI = 0.20–0.66; η2 = 0.096; P=<0.001). This study shows the importance of early individual assessment of post-game hamstring strength in players with prior hamstring injury and could be valuable to inform post-game hamstring recovery in future applications.

Covid-19 abruptly halted the 2020 afl season. what was the impact on athlete performance?

Introduction: The coronavirus pandemic (COVID-19) lead to the implementation of strict lockdown measures in 2020 for all residents. Self-isolation rules meant that sporting seasons, such as the typical 23-week Australian Football League (AFL) season were suspended for 12 weeks soon after the season commenced. While popular opinion suggested training limitations associated with lockdownd would have negative effects on athletes’ performance, medal tallys in the Olympics and Paralympics indicated otherwise.

Effects of concussion on technical performance in professional Australian football

ABSTRACT Introduction There is limited research on the on-field performance of previously concussed athletes. The aim of this exploratory study was to investigate athlete technical performance pre- and post-concussion in Australian Football. Methods Using publicly available data, male athletes who sustained a concussion during the 2016–19 professional Australian Football League seasons were analysed across five games pre- and post-concussion (concussion events n = 41, age 25.4 ± 3.5 years; control n = 39, age 25.2 ± 3.6 years). Mean technical performance metrics (goals, time-on-ground percentage, kicks, ground ball and disposal efficiency, contested marks) over the five games pre- and post-concussion, as well as within-athlete performance variability measures (standard deviation and coefficient of variation) were analysed. Results Results showed no significant group-by-time interactions, nor effect of time (pre-post) for any technical performance metric. Similarly, the within-athlete standard deviation and coefficient of variation of technical performance metrics showed no group-by-time interaction, nor effect for time. Conclusion This retrospective study has shown that athlete performance averaged over five games is not affected post-concussion in elite men’s Australian Football. Further prospective studies controlling for contextual match factors based on opposition and environmental conditions may be required to identify potential in-game technical performance changes following return-to-play from concussion.

Team numerical advantage in Australian rules football: A missing piece of the scoring puzzle?

The primary aim of this study was to determine the relationship between a team numerical advantage during structured phases of play and match event outcomes in professional Australian football. The secondary aim was to quantify how players occupy different sub-areas of the playing field in match play, while accounting for match phase and ball location. Spatiotemporal player tracking data and play-by-play event data from professional players and teams were collected from the 2019 Australian Football League season played at a single stadium. Logistic regression analysed the relationship between total players and team numerical advantage during clearances and inside 50’s. Total players and team numerical advantage were also quantified continuously throughout a match, which were separated into three match phases (offence, defence, and stoppage) and four field positions (defensive 50, defensive midfield, attacking midfield, and forward 50). Results identified an increased team numerical advantage produced a greater likelihood of gaining possession from clearances or generating a score from inside 50’s. Although, an increased number of total players inside 50 was likely associated with a concomitant decrease in the probability of scoring, irrespective of a team numerical advantage. Teams were largely outnumbered when the ball was in their forward 50 but attained a numerical advantage when the ball was in the defensive 50.

Technical involvements and pressure applied influence movement demands in elite Australian Football Match-play

ABSTRACT Match activity analyses in Australian Football (AF) have provided information on the loads of match-play and identified contextual factors influencing physical activity. Match physical activity has been shown to be influenced by technical and tactical factors. This study examined the association of in-game contextual factors and technical involvements on the activity demands of elite AF players. Global positioning system data were recorded from 35 players in 13 matches throughout the 2019 Australian Football League (AFL) season. Technical involvements – player possession, and defensive pressure applied – were attained from Champion Data (AFL statistics provider), while possession phases and the commencing possession chain event were manually coded. Mixed models examined the influence possession chain contextual factors had on the total distance (TD) and high-speed running (HSR). During attacking phases, physical activity increased when a player had a possession, when a greater number of opposition players applied pressure, and when play was initiated from a turnover (p ≤ 0.001). During defensive phases, physical activity was greatest when an individual player applied pressure, however, also increased when the total number of players applying pressure increased and when play was initiated from a turnover or kick-in (p ≤ 0.001). Overall, this study combines the three constructs of AF performance (physical, technical and tactical) and demonstrates that TD and HSR demands are greatest in a chain when a player was directly involved in the play or applying pressure on the opposition.

Lower limb joint position sense and prospective hamstring injury

BackgroundThe hamstrings remain the most injured muscle group within the Australian Football League (AFL). ObjectiveTo investigate preseason measures of hip and knee joint position sense (JPS) and prospective hamstring injury in AFL players. DesignProspective cohort study. MethodsA total of 116 AFL players were recruited for this study. JPS was assessed with 3-D sensors using mono-articular hip (45° flexion and 0°) and knee (90° and 45° flexion) joint reproduction tests conducted in the preseason. Hamstring injury data were collected prospectively in the following AFL season. Wilcoxon-signed rank tests were used to assess between the subsequently injured and uninjured limbs. Mann-Whitney U tests were used to assess between group differences and odds ratio (OR) were used to predict players at risk of hamstring injury. ResultsEight players with JPS data sustained a season hamstring injury and 108 players did not. JPS was not significantly different between the subsequently injured and uninjured limbs (all P values > 0.05). No significant differences for any JPS measure were found between the subsequently injured and uninjured players (all p's > 0.05). ORs did not achieve significance for AE (2.7, p = 0.21) or for RMSE (OR = 1.9, p = 0.44). ConclusionLower limb JPS measures were not predictive of hamstring injury in AFL players.

Can Elite Australian Football Player’s Game Performance Be Predicted?

Abstract In elite Australian football (AF) many studies have investigated individual player performance using a variety of outcomes (e.g. team selection, game running, game rating etc.), however, none have attempted to predict a player’s performance using combinations of pre-game factors. Therefore, our aim was to investigate the ability of commonly reported individual player and team characteristics to predict individual Australian Football League (AFL) player performance, as measured through the official AFL player rating (AFLPR) (Champion Data). A total of 158 variables were derived for players (n = 64) from one AFL team using data collected during the 2014-2019 AFL seasons. Various machine learning models were trained (cross-validation) on the 2014-2018 seasons, with the 2019 season used as an independent test set. Model performance, assessed using root mean square error (RMSE), varied (4.69-5.03 test set RMSE) but was generally poor when compared to a singular variable prediction (AFLPR pre-game rating: 4.72 test set RMSE). Variation in model performance (range RMSE: 0.14 excusing worst model) was low, indicating different approaches produced similar results, however, glmnet models were marginally superior (4.69 RMSE test set). This research highlights the limited utility of currently collected pre-game variables to predict week-to-week game performance more accurately than simple singular variable baseline models.

A League-Wide Evaluation of Factors Influencing Match Activity Profile in Elite Australian Football.

Introduction: Accurate interpretation of activity profile data requires an understanding of the variables influencing player movement during matches.Methods: Over 65,000 stints (player rotations) from all 207 matches of the 2018 Australian Football League (AFL) season were evaluated. The relative activity profile including total distance per minute (TD), high-speed running distance per minute (HSR) and Player LoadTM per minute (PL) was determined for each stint and analysed against a range of match-related, player-related and environment-related predictor variables using multivariate linear mixed modelling. Effect size statistics along with the uncertainty in the estimates (95% confidence interval) were used to interpret the findings.Results: The smallest important effects on TD, HSR, and PL were calculated as 1.5%, 5.5%, and 2.4%, respectively. Stint duration had small to moderate negative effects on TD (−6%), PL (−7.7%), and HSR (−13%), while recovery duration between stints had a small positive effect on HSR (+7%). There were moderate reductions in TD (−8%), HSR (−23%), and PL (−9.6%) in the last quarter compared to the first quarter of matches, while similar reductions existed in subsequent stints compared to the first stint in each quarter. Moderate to large differences of up to 9% in TD, 48% in HSR and 12% in PL existed between positions. The TD of less experienced players was slightly higher than their more experienced counterparts (2–3%). A 5% increase in body mass was associated with a small reduction in HSR (−5.5%). There were small reductions in TD (−2%), HSR (−10%), and PL (−3%) during the Finals Series compared to the Premiership Season. Moderate levels of rainfall during matches and higher apparent temperatures had small negative effects on TD (−2%) and HSR (−6 and −9%). The number of days break between matches, score margin, match outcome, ground hardness, ground size, and traveling for the current or the previous match had trivial effects on the activity profile.Conclusion: Player position and stage of the match (quarter) had the largest effects on match activity profile while stint duration, recovery duration, stint timing, professional experience, body mass, stage of the season, and weather conditions also had substantial effects.

Data Reduction Approaches to Athlete Monitoring in Professional Australian Football.

To apply data reduction methods to athlete-monitoring measures to address the issue of data overload for practitioners of professional Australian football teams. Data were collected from 45 professional Australian footballers from 1 club during the 2018 Australian Football League season. External load was measured in training and matches by 10-Hz OptimEye S5 and ClearSky T6 GPS units. Internal load was measured via the session rate of perceived exertion method. Perceptual wellness was measured via questionnaires completed before training sessions with players providing a rating (1-5 Likert scale) of muscle soreness, sleep quality, fatigue, stress, and motivation. Percentage of maximum speed was calculated relative to individual maximum velocity recorded during preseason testing. Derivative external training load measures (total daily, weekly, and monthly) were calculated. Principal-component analyses (PCAs) were conducted for Daily and Chronic measures, and components were identified via scree plot inspection (eigenvalue > 1). Components underwent orthogonal rotation with a factor loading redundancy threshold of 0.70. The Daily PCA identified components representing external load, perceived wellness, and internal load. The Chronic PCA identified components representing 28-d speed exposure, 28-d external load, 7-d external load, and 28-d internal load. Perceived soreness did not meet the redundancy threshold. Monitoring player exposure to maximum speed is more appropriate over chronic than short time frames to capture variations in between-matches training-cycle duration. Perceived soreness represents a distinct element of a player's perception of wellness. Summed-variable and single-variable approaches are novel methods of data reduction following PCA of athlete monitoring data.

Possession chain factors influence movement demands in elite Australian football match-play

ABSTRACT Contemporary analysis of physical activity in Australian Football (AF) is typically presented as a total measure and independent of game context, which is not representative of how the game is played and/or assessed by coaches. This study examines the activity profile of individual possession chains and determines the influence that field position, initial chain state, and possession phase have on these activity characteristics in men’s AF. Global positioning system data were attained from 35 players in 13 matches across the 2019 Australian Football League season. Matches were coded into different possession phases, initial field location of the ball, and initial chain state. Mixed models were built to observe the influence of field position and initial chain state for each possession phase. Less TD and HSR distance were covered during attacking chains in the forward 50 and attacking midfield, while defensive chains covered less TD and HSR in the defensive 50 and defensive midfield (p < 0.001). Significant main effects for possession phase and initial chain state were observed for TD and HSR. TD and HSR were higher during attacking chains, while chains beginning from a stoppage were lower than intercept and kick-ins (p < 0.001). Overall, the most intense moments of the game appear similar across all possession phases when field location is accounted for and that transitioning the ball quickly from the defensive end of the field results in greater physical activity. These findings can be used for prescription and monitoring of training drills specific to AF requirements.

Understanding the relative contribution of technical and tactical performance to match outcome in Australian Football

ABSTRACT The aim of this study was to assess if tactical and technical performance indicators (PIs) could be used in combination to model match outcomes in Australian Football (AF). A database of 101 technical PIs and 14 tactical PIs from every match in the 2009–2016 Australian Football League (AFL) seasons was merged. Two outcome measures Win-loss and Score margin were used as dependent variables. The top 45 ranked technical and tactical PIs from a feature selection process were used to model match outcome using decision tree and Generalised Linear Models (GLMs). Of the top 45 selected features, this included seven tactical PIs. The Win-loss-based Decision tree model achieved a classification accuracy of 89.0% and GLM 93.2%. A Score margin-based GLM achieved a root mean squared error (RMSE) of 6.9 points. A combined approach to the classification of match outcomes provided no improvement in model accuracy compared with previous literature. However, this study has established the relative importance of technical and tactical measures of performance in relation to successful team performance in AF.

Pubic apophysitis in elite Australian Rules football players: MRI findings and the utility of VIBE sequences in evaluating athletes with groin pain

To confirm that pubic apophysitis is common in Australian Rules footballers with groin pain. Thirteen male Australian Football League (AFL) players with groin pain were assessed with volumetric interpolated breath-hold examination (VIBE) MRI over the 2017, 2018, and 2019 AFL seasons. Images were reviewed for pubic maturation, the presence of pubic apophysitis, and associated bone pathology and correlated with side of groin pain. Pubic apophysitis was seen in 92% of AFL players with groin pain. Delayed maturation of the pubic apophyses was observed in 85%. Pubic bone erosions and cyst-like changes were common (100% and 46%, respectively), but due to delayed maturation of the apophyses and apophysiolysis. Apophysitis associated with adductor brevis-gracilis was more common than adductor longus-associated apophysitis. Pubic apophysitis associated with delayed maturation of the pubic apophyses is common in AFL players and is potentially a significant cause of groin pain in these athletes. Imaging findings in this group are the same as those conventionally describing osteitis pubis. Pubic apophysitis is best visualised with VIBE magnetic resonance imaging (MRI) and may be a more pathologically correct description of early, adductor load-related pubic bone pathology.

The practical application of a method of analysing the variability of within-step accelerations collected via athlete tracking devices

ABSTRACTResolving intra-stride accelerations from training and game data routinely collected by athlete tracking devices is rarely attempted, even though these data can provide important insights into the physical condition of athletes. This study describes in detail a novel analysis tool which uses accelerometer-based measures of step variability to assess longitudinal consistency of stride technique. The performance of the analysis tool was examined by combining results from the analysis of data collected over the course of an Australian Football League season with instances of missed or modified training. Results demonstrated increased gait-related trunk acceleration variability when training was modified due to load or accumulated game and training activity (p = 0.02) and reduced trunk acceleration variability in the week following a training modification due to any injury described as “ankle”, “heel”, or any other description that would loosely refer to a joint or bone in the leg (p = 0.01). These results support the findings of recent research which used a slightly modified version of the same analysis tool to demonstrate an increase in gait-related trunk acceleration variability with the increased acute physical load.

Deceleration, Acceleration, and Impacts Are Strong Contributors to Muscle Damage in Professional Australian Football.

Gastin, PB, Hunkin, SL, Fahrner, B, and Robertson, S. Deceleration, acceleration, and impacts are strong contributors to muscle damage in professional Australian football. J Strength Cond Res 33(12): 3374-3383, 2019-The purpose of this study was to investigate the relationships between serum creatine kinase [CK], an indirect marker of muscle damage, and specific indices of match load in elite Australian football. Twenty-six professional players were assessed during a competitive Australian Football League (AFL) season. [CK] was collected 24-36 hours before match and 34-40 hours after match during 8 in-season rounds. An athlete-tracking technology was used to quantify match load. Generalized estimating equations and random forest models were constructed to determine the extent to which match-load indices and pre-match [CK] explained post-match [CK]. There was a 129 ± 152% increase in [CK] in response to AFL competition. Generalized estimating equations found that number of impacts >3g (p = 0.004) and game time (p = 0.016) were most strongly associated with post-match [CK]. Random forest, with considerably lower errors (130 vs. 316 U·L), found deceleration, acceleration, impacts >3g, and sprint distance to be the strongest predictors. Pre-match [CK] accounted for 11% of post-match [CK], and considerable interindividual and intraindividual variability existed in the data. Creatine kinase, an indicator of muscle damage, was considerably elevated as a result of AFL competition. Parametric and machine-learning analysis techniques found several indices of physical load associated with muscle damage during competition, with impacts >3g and high-intensity running variables as the strongest predictors. [CK] may be used as a global measure of muscle damage in field team sports such as AF, yet with some caution given cost, invasiveness, and inherent variability. Quantifying physical load and the responses to that load can guide athlete management decision-making and is best undertaken within a suite of practical, sport-specific measures, where data are interpreted individually and with an understanding of the limitations.

Multifactorial analysis of factors influencing elite australian football match outcomes: a machine learning approach

Abstract 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.

Comparing subjective and objective evaluations of player performance in Australian Rules football.

Player evaluation plays a fundamental role in the decision-making processes of professional sporting organisations. In the Australian Football League, both subjective and objective evaluations of player match performance are commonplace. This study aimed to identify the extent to which performance indicators can explain subjective ratings of player performance. A secondary aim was to compare subjective and objective ratings of player performance. Inside Football Player Ratings (IFPR) and Australian Football League Player Ratings were collected as subjective and objective evaluations of player performance, respectively, for each player during all 1026 matches throughout the 2013–2017 Australian Football League seasons. Nine common player performance indicators, player role classification, player age and match outcomes were also collected. Standardised linear mixed model and recursive partitioning and regression tree models were undertaken across the whole dataset, as well as separately for each of the seven player roles. The mixed model analysis produced a model associating the performance indicators with IFPR at a root mean square error of 0.98. Random effects accounting for differences between seasons and players ranged by 0.09 and 1.73 IFPR each across the five seasons and 1052 players, respectively. The recursive partitioning and regression tree model explained IFPR exactly in 35.8% of instances, and to within 1.0 IFPR point in 81.0% of instances. When analysed separately by player role, exact explanation varied from 25.2% to 41.7%, and within 1.0 IFPR point from 70.3% to 88.6%. Overall, kicks and handballs were most associated with the IFPR. This study highlights that a select few features account for a majority of the variance when explaining subjective ratings of player performance, and that these vary by player role. Australian Football League organisations should utilise both subjective and objective assessments of performance to gain a better understanding of the differences associated with subjective performance assessment.